{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4a4fdef6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# detect_and_segment.py\n",
    "import torch\n",
    "import supervision as sv\n",
    "from typing import List, Tuple, Optional\n",
    "\n",
    "# ==== 1. One-time global model loading  =====================================\n",
    "from utils.florence import (\n",
    "    load_florence_model,\n",
    "    run_florence_inference,\n",
    "    FLORENCE_OPEN_VOCABULARY_DETECTION_TASK\n",
    ")\n",
    "from utils.sam import load_sam_image_model, run_sam_inference\n",
    "\n",
    "from PIL import Image, ImageDraw, ImageColor\n",
    "import numpy as np\n",
    "\n",
    "DEVICE = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "# load models once – they stay in memory for repeated calls\n",
    "FLORENCE_MODEL, FLORENCE_PROC = load_florence_model(device=DEVICE)\n",
    "SAM_IMAGE_MODEL               = load_sam_image_model(device=DEVICE)\n",
    "\n",
    "# quick annotators\n",
    "COLORS          = ['#FF1493', '#00BFFF', '#FF6347', '#FFD700', '#32CD32', '#8A2BE2']\n",
    "COLOR_PALETTE   = sv.ColorPalette.from_hex(COLORS)\n",
    "BOX_ANNOTATOR   = sv.BoxAnnotator(color=COLOR_PALETTE, color_lookup=sv.ColorLookup.INDEX)\n",
    "LABEL_ANNOTATOR = sv.LabelAnnotator(\n",
    "    color=COLOR_PALETTE,\n",
    "    color_lookup=sv.ColorLookup.INDEX,\n",
    "    text_position=sv.Position.CENTER_OF_MASS,\n",
    "    text_color=sv.Color.from_hex(\"#000000\"),\n",
    "    border_radius=5,\n",
    ")\n",
    "MASK_ANNOTATOR  = sv.MaskAnnotator(color=COLOR_PALETTE, color_lookup=sv.ColorLookup.INDEX)\n",
    "\n",
    "# ==== 2. Inference function  ===============================================\n",
    "\n",
    "@torch.inference_mode()\n",
    "@torch.autocast(device_type=\"cuda\", dtype=torch.bfloat16)\n",
    "def detect_and_segment(\n",
    "    image          : Image.Image,\n",
    "    text_prompts   : str | List[str],\n",
    "    return_image   : bool = True,\n",
    ") -> Tuple[sv.Detections, Optional[Image.Image]]:\n",
    "    \"\"\"\n",
    "    Run Florence-2 open-vocabulary detection + SAM2 mask refinement on a PIL image.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    image : PIL.Image\n",
    "        Input image in RGB.\n",
    "    text_prompts : str | List[str]\n",
    "        Single prompt or comma-separated list (e.g. \"dog, tail, leash\").\n",
    "    return_image : bool\n",
    "        If True, also returns an annotated PIL image.\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    detections : sv.Detections\n",
    "        Supervision object with xyxy, mask, class_id, etc.\n",
    "    annotated  : PIL.Image | None\n",
    "        Annotated image (None if return_image=False)\n",
    "    \"\"\"\n",
    "    # Normalize prompt list\n",
    "    if isinstance(text_prompts, str):\n",
    "        prompts = [p.strip() for p in text_prompts.split(\",\") if p.strip()]\n",
    "    else:\n",
    "        prompts = [p.strip() for p in text_prompts]\n",
    "\n",
    "    if len(prompts) == 0:\n",
    "        raise ValueError(\"Empty prompt list given.\")\n",
    "\n",
    "    # Collect detections from each prompt\n",
    "    det_list: list[sv.Detections] = []\n",
    "    for p in prompts:\n",
    "        _, result = run_florence_inference(\n",
    "            model      = FLORENCE_MODEL,\n",
    "            processor  = FLORENCE_PROC,\n",
    "            device     = DEVICE,\n",
    "            image      = image,\n",
    "            task       = FLORENCE_OPEN_VOCABULARY_DETECTION_TASK,\n",
    "            text       = p,\n",
    "        )\n",
    "        det = sv.Detections.from_lmm(\n",
    "            lmm           = sv.LMM.FLORENCE_2,\n",
    "            result        = result,\n",
    "            resolution_wh = image.size,\n",
    "        )\n",
    "        det = run_sam_inference(SAM_IMAGE_MODEL, image, det)  # SAM2 refinement\n",
    "        det_list.append(det)\n",
    "\n",
    "    detections = sv.Detections.merge(det_list)\n",
    "\n",
    "    annotated_img = None\n",
    "    if return_image:\n",
    "        annotated_img = image.copy()\n",
    "        annotated_img = MASK_ANNOTATOR.annotate(annotated_img, detections)\n",
    "        annotated_img = BOX_ANNOTATOR.annotate(annotated_img, detections)\n",
    "        annotated_img = LABEL_ANNOTATOR.annotate(annotated_img, detections)\n",
    "\n",
    "    return detections, annotated_img\n",
    "\n",
    "\n",
    "\n",
    "def fill_detected_bboxes(\n",
    "    image: Image.Image,\n",
    "    text: str,\n",
    "    inflate_pct: float = 0.10,\n",
    "    fill_color: str | tuple[int, int, int] = \"#00FF00\",\n",
    "):\n",
    "    \"\"\"\n",
    "    Detect objects matching `text`, inflate each bounding-box by `inflate_pct`,\n",
    "    fill the area with `fill_color`, and return the resulting image.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    image        : PIL.Image\n",
    "        Input image (RGB).\n",
    "    text         : str\n",
    "        Comma-separated prompt(s) for open-vocabulary detection.\n",
    "    inflate_pct  : float, default 0.10\n",
    "        Extra margin per side (0.10 = +10 % width & height).\n",
    "    fill_color   : str | tuple, default \"#00FF00\"\n",
    "        Solid color used to fill each inflated bbox (hex or RGB tuple).\n",
    "\n",
    "    Returns\n",
    "    -------\n",
    "    filled_img   : PIL.Image\n",
    "        Image with each detected (inflated) box filled.\n",
    "    detections   : sv.Detections\n",
    "        Original detection object from `detect_and_segment`.\n",
    "    \"\"\"\n",
    "    # run Florence2 + SAM2 pipeline (your helper from earlier)\n",
    "    detections, _ = detect_and_segment(image, text)\n",
    "\n",
    "    w, h = image.size\n",
    "    filled_img = image.copy()\n",
    "    draw = ImageDraw.Draw(filled_img)\n",
    "    fill_rgb = ImageColor.getrgb(fill_color) if isinstance(fill_color, str) else fill_color\n",
    "\n",
    "    for box in detections.xyxy:\n",
    "        # xyxy is numpy array → cast to float for math\n",
    "        x1, y1, x2, y2 = box.astype(float)\n",
    "        dw, dh = (x2 - x1) * inflate_pct, (y2 - y1) * inflate_pct\n",
    "        x1_i = max(0, x1 - dw)\n",
    "        y1_i = max(0, y1 - dh)\n",
    "        x2_i = min(w, x2 + dw)\n",
    "        y2_i = min(h, y2 + dh)\n",
    "        draw.rectangle([x1_i, y1_i, x2_i, y2_i], fill=fill_rgb)\n",
    "\n",
    "    return filled_img, detections\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "00a02394",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "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",
      "image/png": "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",
      "text/plain": [
       "<PIL.Image.Image image mode=RGB size=512x512>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from PIL import Image\n",
    "\n",
    "img = Image.open(\"shoess.webp\").resize((512, 512)).convert(\"RGB\")\n",
    "\n",
    "filled_img, dets = fill_detected_bboxes(\n",
    "    image       = img,\n",
    "    text        = \"person\",\n",
    "    inflate_pct = 0.15,        # 15 % margin\n",
    "    fill_color  = \"#00FF00\"    # chroma green\n",
    ")\n",
    "\n",
    "filled_img\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "864b8731",
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import load_dataset, Image as HFImage\n",
    "from pathlib import Path\n",
    "from tqdm.auto import tqdm\n",
    "import io, random\n",
    "from detect_and_segment_v2 import fill_detected_bboxes   # ← your helper\n",
    "\n",
    "# -------------------------------------------------------\n",
    "# 1. Config\n",
    "# -------------------------------------------------------\n",
    "SOURCE_DATASET   = \"beans\"           # original dataset\n",
    "SPLIT            = \"train\"\n",
    "TEXT_PROMPT      = \"bean leaf\"\n",
    "TARGET_REPO      = \"your-username/beans_bbox_filled\"  # push destination\n",
    "INFLATE_MIN, INFLATE_MAX = 0.10, 0.20\n",
    "\n",
    "# -------------------------------------------------------\n",
    "# 2. Load dataset locally\n",
    "# -------------------------------------------------------\n",
    "ds = load_dataset(SOURCE_DATASET, split=SPLIT, streaming=False)\n",
    "print(\"Samples:\", len(ds))\n",
    "\n",
    "# -------------------------------------------------------\n",
    "# 3. Processing function for .map()\n",
    "# -------------------------------------------------------\n",
    "def add_bfill(sample):\n",
    "    img: HFImage = sample[\"image\"].convert(\"RGB\")\n",
    "\n",
    "    # random inflate 0.10–0.20\n",
    "    inflate_pct = random.uniform(INFLATE_MIN, INFLATE_MAX)\n",
    "\n",
    "    bfilled_img, _ = fill_detected_bboxes(\n",
    "        image       = img,\n",
    "        text        = TEXT_PROMPT,\n",
    "        inflate_pct = inflate_pct,\n",
    "        fill_color  = \"#00FF00\",\n",
    "    )\n",
    "\n",
    "    # store as raw bytes (Hub-friendly) or leave as PIL\n",
    "    buf = io.BytesIO()\n",
    "    bfilled_img.save(buf, format=\"PNG\")\n",
    "    sample[\"bbox_filled\"] = {\"bytes\": buf.getvalue(), \"path\": None}\n",
    "\n",
    "    return sample\n",
    "\n",
    "# -------------------------------------------------------\n",
    "# 4. Map over the dataset (with tqdm progress bar)\n",
    "# -------------------------------------------------------\n",
    "ds = ds.map(add_bfill, batched=False, desc=\"Generating bbox-filled column\")\n",
    "\n",
    "# Tell datasets that `\"bbox_filled\"` is an image column\n",
    "ds = ds.cast_column(\"bbox_filled\", HFImage())\n",
    "\n",
    "# -------------------------------------------------------\n",
    "# 5. Push back to the Hub\n",
    "# -------------------------------------------------------\n",
    "ds.push_to_hub(TARGET_REPO)   # requires you to be logged in (`huggingface-cli login`)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0adaa56b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ERROR! Session/line number was not unique in database. History logging moved to new session 71\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 5764 rows.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "df609901d8104750ae9835dbb73d1bf5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Processing sequentially:   0%|          | 0/5764 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/PIL/ImageFile.py:643\u001b[0m, in \u001b[0;36m_save\u001b[0;34m(im, fp, tile, bufsize)\u001b[0m\n\u001b[1;32m    642\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 643\u001b[0m     fh \u001b[38;5;241m=\u001b[39m \u001b[43mfp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfileno\u001b[49m()\n\u001b[1;32m    644\u001b[0m     fp\u001b[38;5;241m.\u001b[39mflush()\n",
      "\u001b[0;31mAttributeError\u001b[0m: '_idat' object has no attribute 'fileno'",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[9], line 67\u001b[0m\n\u001b[1;32m     64\u001b[0m json_path   \u001b[38;5;241m=\u001b[39m QA_DIR \u001b[38;5;241m/\u001b[39m \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00midx\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m05d\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m_bbox.json\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m     66\u001b[0m filled_img\u001b[38;5;241m.\u001b[39msave(filled_path)\n\u001b[0;32m---> 67\u001b[0m \u001b[43manno\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msave\u001b[49m\u001b[43m(\u001b[49m\u001b[43manno_path\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     68\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mopen\u001b[39m(json_path, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mw\u001b[39m\u001b[38;5;124m\"\u001b[39m) \u001b[38;5;28;01mas\u001b[39;00m fp:\n\u001b[1;32m     69\u001b[0m     json\u001b[38;5;241m.\u001b[39mdump({\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mxyxy\u001b[39m\u001b[38;5;124m\"\u001b[39m: detections\u001b[38;5;241m.\u001b[39mxyxy\u001b[38;5;241m.\u001b[39mtolist()}, fp)\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/PIL/Image.py:2581\u001b[0m, in \u001b[0;36mImage.save\u001b[0;34m(self, fp, format, **params)\u001b[0m\n\u001b[1;32m   2578\u001b[0m     fp \u001b[38;5;241m=\u001b[39m cast(IO[\u001b[38;5;28mbytes\u001b[39m], fp)\n\u001b[1;32m   2580\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 2581\u001b[0m     \u001b[43msave_handler\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfp\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfilename\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   2582\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m:\n\u001b[1;32m   2583\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m open_fp:\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/PIL/PngImagePlugin.py:1492\u001b[0m, in \u001b[0;36m_save\u001b[0;34m(im, fp, filename, chunk, save_all)\u001b[0m\n\u001b[1;32m   1488\u001b[0m     single_im \u001b[38;5;241m=\u001b[39m _write_multiple_frames(\n\u001b[1;32m   1489\u001b[0m         im, fp, chunk, mode, rawmode, default_image, append_images\n\u001b[1;32m   1490\u001b[0m     )\n\u001b[1;32m   1491\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m single_im:\n\u001b[0;32m-> 1492\u001b[0m     \u001b[43mImageFile\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_save\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1493\u001b[0m \u001b[43m        \u001b[49m\u001b[43msingle_im\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1494\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcast\u001b[49m\u001b[43m(\u001b[49m\u001b[43mIO\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;28;43mbytes\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m_idat\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfp\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mchunk\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1495\u001b[0m \u001b[43m        \u001b[49m\u001b[43m[\u001b[49m\u001b[43mImageFile\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_Tile\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mzip\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m+\u001b[39;49m\u001b[43m \u001b[49m\u001b[43msingle_im\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msize\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mrawmode\u001b[49m\u001b[43m)\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1496\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m info:\n\u001b[1;32m   1499\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m info_chunk \u001b[38;5;129;01min\u001b[39;00m info\u001b[38;5;241m.\u001b[39mchunks:\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/PIL/ImageFile.py:647\u001b[0m, in \u001b[0;36m_save\u001b[0;34m(im, fp, tile, bufsize)\u001b[0m\n\u001b[1;32m    645\u001b[0m     _encode_tile(im, fp, tile, bufsize, fh)\n\u001b[1;32m    646\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m (\u001b[38;5;167;01mAttributeError\u001b[39;00m, io\u001b[38;5;241m.\u001b[39mUnsupportedOperation) \u001b[38;5;28;01mas\u001b[39;00m exc:\n\u001b[0;32m--> 647\u001b[0m     \u001b[43m_encode_tile\u001b[49m\u001b[43m(\u001b[49m\u001b[43mim\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfp\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtile\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbufsize\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mexc\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    648\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mhasattr\u001b[39m(fp, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mflush\u001b[39m\u001b[38;5;124m\"\u001b[39m):\n\u001b[1;32m    649\u001b[0m     fp\u001b[38;5;241m.\u001b[39mflush()\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/PIL/ImageFile.py:673\u001b[0m, in \u001b[0;36m_encode_tile\u001b[0;34m(im, fp, tile, bufsize, fh, exc)\u001b[0m\n\u001b[1;32m    670\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m exc:\n\u001b[1;32m    671\u001b[0m     \u001b[38;5;66;03m# compress to Python file-compatible object\u001b[39;00m\n\u001b[1;32m    672\u001b[0m     \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[0;32m--> 673\u001b[0m         errcode, data \u001b[38;5;241m=\u001b[39m \u001b[43mencoder\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mencode\u001b[49m\u001b[43m(\u001b[49m\u001b[43mbufsize\u001b[49m\u001b[43m)\u001b[49m[\u001b[38;5;241m1\u001b[39m:]\n\u001b[1;32m    674\u001b[0m         fp\u001b[38;5;241m.\u001b[39mwrite(data)\n\u001b[1;32m    675\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m errcode:\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "# sequential_furniture_bbox.py\n",
    "# =============================================================\n",
    "# Florence-2 + SAM-2 sequential preprocessing (no multiprocessing)\n",
    "# =============================================================\n",
    "import os, io, json, random\n",
    "from pathlib import Path\n",
    "from tqdm.auto import tqdm\n",
    "from PIL import Image\n",
    "from datasets import load_dataset, Image as HFImage, disable_progress_bar\n",
    "\n",
    "# --- silence tokenizers fork warning -------------------------\n",
    "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
    "disable_progress_bar()\n",
    "\n",
    "# --- your helper (must be importable) ------------------------\n",
    "\n",
    "# ==================== CONFIG =================================\n",
    "DATASET_NAME   = \"fotographerai/furniture_shortlisted_batch1_captioned\"\n",
    "SPLIT          = \"train\"\n",
    "IMAGE_COL      = \"img2\"\n",
    "PROMPT_COL     = \"prompt\"\n",
    "\n",
    "INFLATE_RANGE  = (0.10, 0.20)          # 10-20 % expansion\n",
    "FILL_COLOR     = \"#00FF00\"             # chroma green\n",
    "QA_DIR         = Path(\"bbox_filled_review_seq\")\n",
    "QA_DIR.mkdir(exist_ok=True)\n",
    "\n",
    "# ==================== LOAD DATASET ============================\n",
    "ds = load_dataset(DATASET_NAME, split=SPLIT, streaming=False)\n",
    "print(f\"Loaded {len(ds)} rows.\")\n",
    "\n",
    "bbox_filled_col = []\n",
    "annotated_col   = []\n",
    "bbox_json_col   = []\n",
    "\n",
    "# ==================== MAIN LOOP ===============================\n",
    "for idx, sample in enumerate(tqdm(ds, desc=\"Processing sequentially\")):\n",
    "    # -- trim prompt before first semicolon --------------------\n",
    "    prompt_full = sample[PROMPT_COL]\n",
    "    prompt_trim = prompt_full.split(\";\", 1)[0].strip()\n",
    "\n",
    "    img: Image.Image = sample[IMAGE_COL].convert(\"RGB\")\n",
    "    inflate_pct     = random.uniform(*INFLATE_RANGE)\n",
    "\n",
    "    # -- generate green-screen & detections --------------------\n",
    "    try:\n",
    "        filled_img, detections = fill_detected_bboxes(\n",
    "            image       = img,\n",
    "            text        = prompt_trim,\n",
    "            inflate_pct = inflate_pct,\n",
    "            fill_color  = FILL_COLOR,\n",
    "        )\n",
    "    except:\n",
    "        continue\n",
    "    # -- build annotated overlay ------------------------------\n",
    "    anno = img.copy()\n",
    "    anno = MASK_ANNOTATOR.annotate(anno, detections)\n",
    "    anno = BOX_ANNOTATOR.annotate(anno, detections)\n",
    "    anno = LABEL_ANNOTATOR.annotate(anno, detections)\n",
    "\n",
    "    # -- save files for manual QA ------------------------------\n",
    "    filled_path = QA_DIR / f\"{idx:05d}_green.png\"\n",
    "    anno_path   = QA_DIR / f\"{idx:05d}_anno.png\"\n",
    "    json_path   = QA_DIR / f\"{idx:05d}_bbox.json\"\n",
    "\n",
    "    filled_img.save(filled_path)\n",
    "    anno.save(anno_path)\n",
    "    with open(json_path, \"w\") as fp:\n",
    "        json.dump({\"xyxy\": detections.xyxy.tolist()}, fp)\n",
    "\n",
    "    # -- convert images to bytes for dataset -------------------\n",
    "    def pil2bytes(im: Image.Image) -> dict:\n",
    "        buf = io.BytesIO()\n",
    "        im.save(buf, format=\"PNG\")\n",
    "        return {\"bytes\": buf.getvalue(), \"path\": None}\n",
    "\n",
    "    bbox_filled_col.append(pil2bytes(filled_img))\n",
    "    annotated_col.append(pil2bytes(anno))\n",
    "    bbox_json_col.append(json_path.read_text())\n",
    "\n",
    "# ==================== BUILD NEW DATASET =======================\n",
    "new_ds = ds.add_column(\"bbox_filled\", bbox_filled_col)\n",
    "new_ds = new_ds.add_column(\"annotated\",   annotated_col)\n",
    "new_ds = new_ds.add_column(\"bbox_json\",   bbox_json_col)\n",
    "\n",
    "new_ds = new_ds.cast_column(\"bbox_filled\", HFImage())\n",
    "new_ds = new_ds.cast_column(\"annotated\",   HFImage())\n",
    "\n",
    "print(\"✅ Done! Review PNG/JSON in:\", QA_DIR.resolve())\n",
    "print(\"New dataset columns:\", new_ds.column_names)\n",
    "\n",
    "# === (optional) push after manual QA ==========================\n",
    "# new_ds.push_to_hub(\"fotographerai/furniture_shortlisted_batch1_bboxfilled\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "da1ef4bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 2700 rows.\n",
      "[WARN] idx 6: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 8: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 12: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 4: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 7: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 11: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 10: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 13: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 16: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 14: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 15: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 17: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 24: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 46: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 48: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 57: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 64: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 61: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 62: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 67: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 68: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 69: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 79: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 84: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 99: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 98: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 106: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 103: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 119: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 122: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 126: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 139: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 143: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 129: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 144: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 150: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 158: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 161: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 167: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 148: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 172: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 187: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 197: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 205: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 207: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 223: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 217: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 259: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 254: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 250: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 273: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 281: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 294: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 295: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 290: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 1024\n",
      "[WARN] idx 298: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 302: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 297: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 1024\n",
      "[WARN] idx 312: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 310: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 321: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 323: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 322: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 324: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 283: all the input array dimensions except for the concatenation axis must match exactly, but along dimension 1, the array at index 0 has size 1600 and the array at index 1 has size 1024\n",
      "[WARN] idx 338: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 334: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 343: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 355: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 353: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 358: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 342: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 362: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 360: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 364: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 368: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 371: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 377: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 376: boolean index did not match indexed array along axis 0; size of axis is 1024 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 378: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 379: boolean index did not match indexed array along axis 0; size of axis is 1024 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 380: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 367: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 385: boolean index did not match indexed array along axis 1; size of axis is 1993 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 381: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 383: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 387: boolean index did not match indexed array along axis 0; size of axis is 1935 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 391: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 382: boolean index did not match indexed array along axis 0; size of axis is 1024 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 395: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 394: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1024\n",
      "[WARN] idx 370: all the input array dimensions except for the concatenation axis must match exactly, but along dimension 1, the array at index 0 has size 1024 and the array at index 1 has size 2000\n",
      "[WARN] idx 398: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 399: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 432: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 440: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 448: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 455: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 457: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 465: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 461: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 482: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 484: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 508: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 510: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 519: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 524: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 535: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 547: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 545: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 555: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 553: 'NoneType' object is not subscriptable\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e9f1b157af424c6bbe1fc9ce7aa17945",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Threaded Florence+SAM:   0%|          | 0/2700 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[WARN] idx 560: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 567: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 578: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 581: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 591: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 598: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 613: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 629: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 635: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 643: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 647: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 650: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 649: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 654: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 655: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 657: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 659: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 658: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 653: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 664: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 672: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 674: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 691: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 696: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 707: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 708: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 711: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 714: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 729: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 725: boolean index did not match indexed array along axis 0; size of axis is 1998 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 744: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 748: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 755: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 756: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 759: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 757: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 761: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 762: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 774: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 770: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 778: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 775: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 784: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 795: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 793: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 791: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 799: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 800: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 804: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 803: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 805: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 812: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 807: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 801: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 813: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 817: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 815: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 809: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 814: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 811: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 816: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 819: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 824: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 827: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 828: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 829: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 830: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 825: boolean index did not match indexed array along axis 1; size of axis is 1600 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 835: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 833: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 831: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 836: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 839: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 837: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 840: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 1667\n",
      "[WARN] idx 847: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 848: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 841: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 852: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 850: boolean index did not match indexed array along axis 1; size of axis is 1667 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 846: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 860: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 869: boolean index did not match indexed array along axis 1; size of axis is 1600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 870: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 871: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 872: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 878: boolean index did not match indexed array along axis 1; size of axis is 1600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 875: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 873: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 868: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 877: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 874: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 880: boolean index did not match indexed array along axis 1; size of axis is 1600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 881: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 879: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 882: boolean index did not match indexed array along axis 1; size of axis is 1600 but size of corresponding boolean axis is 1674\n",
      "[WARN] idx 883: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 884: boolean index did not match indexed array along axis 1; size of axis is 1600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 885: boolean index did not match indexed array along axis 1; size of axis is 1600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 890: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 887: boolean index did not match indexed array along axis 1; size of axis is 1674 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 893: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 892: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 902: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 901: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 905: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 916: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 924: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 923: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 931: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 926: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 935: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 933: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 927: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 929: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 932: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 928: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 934: boolean index did not match indexed array along axis 1; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 941: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 942: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 1533\n",
      "[WARN] idx 943: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 1533\n",
      "[WARN] idx 945: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 965: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 974: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 972: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 969: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 966: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 973: boolean index did not match indexed array along axis 1; size of axis is 1333 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 993: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1004: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1009: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1012: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1011: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1010: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1013: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1018: boolean index did not match indexed array along axis 1; size of axis is 1373 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1016: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1028: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1031: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1052: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1048: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1047: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1050: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1055: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1053: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1054: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1060: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1068: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1073: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1077: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1079: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1076: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1082: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1092: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1098: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1096: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1103: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1100: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1106: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1107: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 1192\n",
      "[WARN] idx 1104: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 1192\n",
      "[WARN] idx 1108: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 1186\n",
      "[WARN] idx 1102: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 1186\n",
      "[WARN] idx 1109: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1113: boolean index did not match indexed array along axis 1; size of axis is 1186 but size of corresponding boolean axis is 1192\n",
      "[WARN] idx 1110: boolean index did not match indexed array along axis 1; size of axis is 1500 but size of corresponding boolean axis is 1192\n",
      "[WARN] idx 1114: boolean index did not match indexed array along axis 1; size of axis is 1192 but size of corresponding boolean axis is 1204\n",
      "[WARN] idx 1115: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1116: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1250\n",
      "[WARN] idx 1118: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1250\n",
      "[WARN] idx 1120: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1250\n",
      "[WARN] idx 1117: boolean index did not match indexed array along axis 0; size of axis is 1250 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1124: boolean index did not match indexed array along axis 0; size of axis is 900 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1122: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1125: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1119: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1127: boolean index did not match indexed array along axis 1; size of axis is 1116 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1121: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1131: boolean index did not match indexed array along axis 0; size of axis is 2113 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1130: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1134: boolean index did not match indexed array along axis 1; size of axis is 1186 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1132: boolean index did not match indexed array along axis 1; size of axis is 1175 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1137: boolean index did not match indexed array along axis 0; size of axis is 2119 but size of corresponding boolean axis is 2113\n",
      "[WARN] idx 1133: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2113\n",
      "[WARN] idx 1136: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2113\n",
      "[WARN] idx 1126: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2113\n",
      "[WARN] idx 1140: boolean index did not match indexed array along axis 0; size of axis is 2113 but size of corresponding boolean axis is 1882\n",
      "[WARN] idx 1141: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1142: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1143: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1145: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1139: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1148: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1771\n",
      "[WARN] idx 1147: boolean index did not match indexed array along axis 0; size of axis is 2113 but size of corresponding boolean axis is 1771\n",
      "[WARN] idx 1146: boolean index did not match indexed array along axis 0; size of axis is 2138 but size of corresponding boolean axis is 1771\n",
      "[WARN] idx 1144: boolean index did not match indexed array along axis 0; size of axis is 1771 but size of corresponding boolean axis is 1882\n",
      "[WARN] idx 1150: boolean index did not match indexed array along axis 0; size of axis is 1882 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1159: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1892\n",
      "[WARN] idx 1151: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1892\n",
      "[WARN] idx 1160: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1892\n",
      "[WARN] idx 1154: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1892\n",
      "[WARN] idx 1158: boolean index did not match indexed array along axis 0; size of axis is 2305 but size of corresponding boolean axis is 1892\n",
      "[WARN] idx 1138: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1153: boolean index did not match indexed array along axis 1; size of axis is 1024 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1157: boolean index did not match indexed array along axis 0; size of axis is 1024 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1149: boolean index did not match indexed array along axis 0; size of axis is 2113 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1155: boolean index did not match indexed array along axis 1; size of axis is 991 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1174: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1177: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1183: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1194: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1199: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1200: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1201: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1215: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1213: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1221: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1235: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1242: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1254: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1260: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1259: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1258: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1875\n",
      "[WARN] idx 1261: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1875\n",
      "[WARN] idx 1257: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1875\n",
      "[WARN] idx 1264: boolean index did not match indexed array along axis 0; size of axis is 1875 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1262: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1875\n",
      "[WARN] idx 1268: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1266: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1267: boolean index did not match indexed array along axis 0; size of axis is 1875 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1263: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2048\n",
      "[WARN] idx 1275: boolean index did not match indexed array along axis 0; size of axis is 2048 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1276: boolean index did not match indexed array along axis 0; size of axis is 2048 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1272: boolean index did not match indexed array along axis 0; size of axis is 2048 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1283: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1290: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2048\n",
      "[WARN] idx 1274: boolean index did not match indexed array along axis 0; size of axis is 2048 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1292: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2048\n",
      "[WARN] idx 1286: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 2048\n",
      "[WARN] idx 1269: boolean index did not match indexed array along axis 0; size of axis is 2048 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1296: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1302: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1306: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1305: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1317: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1326: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1330: boolean index did not match indexed array along axis 0; size of axis is 2113 but size of corresponding boolean axis is 1365\n",
      "[WARN] idx 1334: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1338: boolean index did not match indexed array along axis 0; size of axis is 1365 but size of corresponding boolean axis is 1500\n",
      "[WARN] idx 1339: boolean index did not match indexed array along axis 0; size of axis is 1500 but size of corresponding boolean axis is 1365\n",
      "[WARN] idx 1343: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1352: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1355: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1371: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1365: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1374: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1373: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1390: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1389: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1394: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1377: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1401: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1406: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1414: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1422: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1438: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1442: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1444: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1457: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1453: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1454: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1464: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1468: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1458: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1480: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1483: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1493: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1503: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1496: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1499: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1508: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1525: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1533: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1545: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1543: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1558: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1560: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1539: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1564: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1572: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1570: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1577: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1587: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1586: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1592: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1596: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1602: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1601: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1603: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1614: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1615: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1620: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1621: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1632: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1635: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1644: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1643: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1657: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1660: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1658: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1671: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1667: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1668: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1663: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1669: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1672: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1678: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1677: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1682: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1680: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1693: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1705: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1713: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1716: 'NoneType' object is not subscriptable[WARN] idx 1715: 'NoneType' object is not subscriptable\n",
      "\n",
      "[WARN] idx 1719: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1725: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1733: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1732: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1744: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1751: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1749: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1757: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1761: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1763: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1766: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1770: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1772: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1776: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1787: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1783: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1782: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1780: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1794: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1795: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1799: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1793: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1806: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1810: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1819: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1824: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1825: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1820: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1833: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1836: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1838: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1839: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1845: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1864: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1868: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1869: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1876: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1882: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1897: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1888: boolean index did not match indexed array along axis 0; size of axis is 1365 but size of corresponding boolean axis is 2002\n",
      "[WARN] idx 1900: boolean index did not match indexed array along axis 0; size of axis is 1365 but size of corresponding boolean axis is 2002\n",
      "[WARN] idx 1879: boolean index did not match indexed array along axis 0; size of axis is 1365 but size of corresponding boolean axis is 2868\n",
      "[WARN] idx 1902: boolean index did not match indexed array along axis 0; size of axis is 2002 but size of corresponding boolean axis is 3820\n",
      "[WARN] idx 1904: boolean index did not match indexed array along axis 0; size of axis is 3820 but size of corresponding boolean axis is 1365\n",
      "[WARN] idx 1901: boolean index did not match indexed array along axis 0; size of axis is 1365 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1907: boolean index did not match indexed array along axis 0; size of axis is 2868 but size of corresponding boolean axis is 3000\n",
      "[WARN] idx 1899: boolean index did not match indexed array along axis 0; size of axis is 1365 but size of corresponding boolean axis is 3000\n",
      "[WARN] idx 1909: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 2002\n",
      "[WARN] idx 1910: boolean index did not match indexed array along axis 0; size of axis is 3000 but size of corresponding boolean axis is 2180\n",
      "[WARN] idx 1903: boolean index did not match indexed array along axis 0; size of axis is 3000 but size of corresponding boolean axis is 2180\n",
      "[WARN] idx 1912: boolean index did not match indexed array along axis 0; size of axis is 2180 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1908: boolean index did not match indexed array along axis 0; size of axis is 2002 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1914: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1906: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1916: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 1911: boolean index did not match indexed array along axis 0; size of axis is 2420 but size of corresponding boolean axis is 2992\n",
      "[WARN] idx 1915: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 2992\n",
      "[WARN] idx 1921: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1913: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1922: boolean index did not match indexed array along axis 0; size of axis is 2398 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 1929: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1918: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 1932: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 1933: boolean index did not match indexed array along axis 0; size of axis is 2160 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 1917: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 2100\n",
      "[WARN] idx 1919: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 2100\n",
      "[WARN] idx 1935: boolean index did not match indexed array along axis 0; size of axis is 3200 but size of corresponding boolean axis is 2964\n",
      "[WARN] idx 1939: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1936: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 1937: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1938: boolean index did not match indexed array along axis 0; size of axis is 2100 but size of corresponding boolean axis is 2002\n",
      "[WARN] idx 1923: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1943: boolean index did not match indexed array along axis 0; size of axis is 3200 but size of corresponding boolean axis is 2088\n",
      "[WARN] idx 1920: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1927: boolean index did not match indexed array along axis 0; size of axis is 2970 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1945: boolean index did not match indexed array along axis 0; size of axis is 2948 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1946: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1944: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 3000\n",
      "[WARN] idx 1940: boolean index did not match indexed array along axis 0; size of axis is 3000 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1930: boolean index did not match indexed array along axis 0; size of axis is 3000 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1934: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1947: boolean index did not match indexed array along axis 0; size of axis is 2160 but size of corresponding boolean axis is 2001\n",
      "[WARN] idx 1942: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1951: boolean index did not match indexed array along axis 0; size of axis is 2144 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1950: boolean index did not match indexed array along axis 0; size of axis is 3200 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1953: boolean index did not match indexed array along axis 0; size of axis is 2001 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 1960: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1952: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 1954: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1958: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1957: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 4000[WARN] idx 1961: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 4000\n",
      "\n",
      "[WARN] idx 1959: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 1949: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 1969: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2400\n",
      "[WARN] idx 1970: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1965: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2400\n",
      "[WARN] idx 1972: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1978: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 1971: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2100\n",
      "[WARN] idx 1973: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2100\n",
      "[WARN] idx 1974: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2100\n",
      "[WARN] idx 1976: boolean index did not match indexed array along axis 0; size of axis is 2100 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 1975: boolean index did not match indexed array along axis 0; size of axis is 2100 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 1982: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2100\n",
      "[WARN] idx 1984: boolean index did not match indexed array along axis 0; size of axis is 2100 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 1986: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2800\n",
      "[WARN] idx 1979: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2800\n",
      "[WARN] idx 1994: boolean index did not match indexed array along axis 0; size of axis is 2800 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 1985: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1987: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2800\n",
      "[WARN] idx 1988: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2200\n",
      "[WARN] idx 1992: boolean index did not match indexed array along axis 0; size of axis is 2800 but size of corresponding boolean axis is 2200\n",
      "[WARN] idx 1990: boolean index did not match indexed array along axis 0; size of axis is 2200 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1993: boolean index did not match indexed array along axis 0; size of axis is 4400 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1998: boolean index did not match indexed array along axis 0; size of axis is 2200 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 1997: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 1989: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 2007: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2010: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 1995: boolean index did not match indexed array along axis 0; size of axis is 2800 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 2014: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2048: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2044: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 3600\n",
      "[WARN] idx 2051: boolean index did not match indexed array along axis 0; size of axis is 3600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 2052: boolean index did not match indexed array along axis 0; size of axis is 1780 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 2036: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2054: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2047: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 2880\n",
      "[WARN] idx 2059: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2050: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2058: boolean index did not match indexed array along axis 0; size of axis is 2880 but size of corresponding boolean axis is 3600\n",
      "[WARN] idx 2063: boolean index did not match indexed array along axis 0; size of axis is 3600 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 2067: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2065: boolean index did not match indexed array along axis 0; size of axis is 3600 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 2071: boolean index did not match indexed array along axis 0; size of axis is 3200 but size of corresponding boolean axis is 3600\n",
      "[WARN] idx 2055: boolean index did not match indexed array along axis 0; size of axis is 3600 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 2069: boolean index did not match indexed array along axis 0; size of axis is 3200 but size of corresponding boolean axis is 2640\n",
      "[WARN] idx 2073: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 2640\n",
      "[WARN] idx 2076: boolean index did not match indexed array along axis 0; size of axis is 2958 but size of corresponding boolean axis is 3600\n",
      "[WARN] idx 2070: boolean index did not match indexed array along axis 0; size of axis is 3200 but size of corresponding boolean axis is 3600\n",
      "[WARN] idx 2064: boolean index did not match indexed array along axis 0; size of axis is 3600 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2072: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2080: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2084: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 1502\n",
      "[WARN] idx 2087: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 1502\n",
      "[WARN] idx 2074: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 1502\n",
      "[WARN] idx 2089: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2936\n",
      "[WARN] idx 2088: boolean index did not match indexed array along axis 0; size of axis is 1502 but size of corresponding boolean axis is 3200\n",
      "[WARN] idx 2083: boolean index did not match indexed array along axis 0; size of axis is 2936 but size of corresponding boolean axis is 2966\n",
      "[WARN] idx 2078: boolean index did not match indexed array along axis 0; size of axis is 2170 but size of corresponding boolean axis is 2966\n",
      "[WARN] idx 2081: boolean index did not match indexed array along axis 0; size of axis is 3200 but size of corresponding boolean axis is 2730\n",
      "[WARN] idx 2082: boolean index did not match indexed array along axis 0; size of axis is 2966 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2085: boolean index did not match indexed array along axis 0; size of axis is 3200 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2068: boolean index did not match indexed array along axis 0; size of axis is 2730 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2077: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2079: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2093: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 4000\n",
      "[WARN] idx 2086: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2097: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 1504\n",
      "[WARN] idx 2090: boolean index did not match indexed array along axis 0; size of axis is 4000 but size of corresponding boolean axis is 1504\n",
      "[WARN] idx 2098: boolean index did not match indexed array along axis 0; size of axis is 1504 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2100: boolean index did not match indexed array along axis 0; size of axis is 2000 but size of corresponding boolean axis is 1600\n",
      "[WARN] idx 2099: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2104: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2095: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 2096: boolean index did not match indexed array along axis 0; size of axis is 1600 but size of corresponding boolean axis is 2000\n",
      "[WARN] idx 2110: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2131: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2130: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2149: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2148: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2142: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2139: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2155: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2160: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2166: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2172: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2173: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2174: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2167: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2178: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2182: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2187: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2190: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2198: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2199: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2214: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2213: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2206: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2236: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2231: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2229: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2242: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2260: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2243: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2254: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2279: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2259: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2278: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2285: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2288: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2291: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2300: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2304: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2301: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2306: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2308: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2297: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2310: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2323: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2324: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2329: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2330: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2346: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2362: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2371: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2395: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2404: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2403: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2409: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2417: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2402: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2439: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2441: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2438: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2457: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2454: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2458: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2448: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2467: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2469: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2445: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2470: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2473: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2471: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2484: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2492: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2482: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2495: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2513: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2521: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2523: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2554: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2549: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2570: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2574: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2596: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2580: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2601: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2608: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2610: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2605: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2606: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2616: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2617: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2630: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2632: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2640: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2663: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2666: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2664: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2665: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2671: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2677: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2668: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2687: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2676: An image must be set with .set_image(...) before mask prediction.\n",
      "[WARN] idx 2684: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2695: 'NoneType' object is not subscriptable\n",
      "[WARN] idx 2688: 'NoneType' object is not subscriptable\n",
      "✅ Completed. Review PNG/JSON in: /home/comdoleger1/florence-sam/bbox_filled_review_fast_batch6\n",
      "New dataset columns: ['id', 'img1', 'img2', 'size', 'category', 'prompt', 'bbox_filled', 'annotated', 'bbox_json']\n"
     ]
    }
   ],
   "source": [
    "# furniture_bbox_fast.py  ────────────────────────────────────────────\n",
    "# Florence-2 + SAM-2 dataset preprocessing with threaded acceleration\n",
    "# -------------------------------------------------------------------\n",
    "import os, io, json, random\n",
    "from pathlib import Path\n",
    "from concurrent.futures import ThreadPoolExecutor, as_completed\n",
    "\n",
    "from tqdm.auto import tqdm\n",
    "from datasets import load_dataset, Image as HFImage, disable_progress_bar\n",
    "from PIL import Image\n",
    "\n",
    "# ── silence HF-tokenizers fork warning ──────────────────────────────\n",
    "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
    "disable_progress_bar()\n",
    "\n",
    "# ── import your Florence+SAM helpers  ───────────────────────────────\n",
    "# assumes detect_and_segment_v2.py & detect_and_segment.py are on PYTHONPATH\n",
    "\n",
    "\n",
    "# ── CONFIGURATION ───────────────────────────────────────────────────\n",
    "DATASET_NAME   = \"fotographerai/surroundingomnibatch6revised_amr\"\n",
    "SPLIT          = \"train\"\n",
    "IMAGE_COL      = \"img2\"\n",
    "PROMPT_COL     = \"prompt\"  # use \"prompt\" for full prompt text\n",
    "\n",
    "INFLATE_RANGE  = (0.01, 0.05)          # 10–20 % expansion\n",
    "FILL_COLOR     = \"#00FF00\"             # chroma key green\n",
    "QA_DIR         = Path(\"bbox_filled_review_fast_batch6\")\n",
    "QA_DIR.mkdir(exist_ok=True)\n",
    "\n",
    "MAX_WORKERS    = 4                   # threads (tune for your GPU)\n",
    "\n",
    "# ── load dataset locally ────────────────────────────────────────────\n",
    "ds = load_dataset(DATASET_NAME, split=SPLIT, streaming=False)\n",
    "print(f\"Loaded {len(ds)} rows.\")\n",
    "\n",
    "# pre-allocate result holders to keep original order\n",
    "bbox_filled_col = [None] * len(ds)\n",
    "annotated_col   = [None] * len(ds)\n",
    "bbox_json_col   = [None] * len(ds)\n",
    "\n",
    "# ── helper to convert PIL → bytes dict for datasets image column\n",
    "def pil2bytes(im: Image.Image) -> dict:\n",
    "    buf = io.BytesIO()\n",
    "    im.save(buf, format=\"PNG\")\n",
    "    return {\"bytes\": buf.getvalue(), \"path\": None}\n",
    "\n",
    "# ── thread worker ---------------------------------------------------\n",
    "def process_row(idx_sample):\n",
    "    idx, sample = idx_sample\n",
    "    #prompt_trim = sample[PROMPT_COL].split(\";\", 1)[0].strip()\n",
    "    prompt_trim = sample[PROMPT_COL]\n",
    "    if prompt_trim==\"sofas_lazy_chairs\":\n",
    "        prompt_trim = \"sofa, lazy chair, armchair\"\n",
    "    elif prompt_trim==\"beds\":   \n",
    "        prompt_trim = \"bed\"\n",
    "    elif prompt_trim==\"desks\":\n",
    "        prompt_trim = \"desk\"\n",
    "    elif prompt_trim==\"office_chairs\":\n",
    "        prompt_trim = \"chair\"\n",
    "    else:\n",
    "        prompt_trim = sample[PROMPT_COL].split(\";\", 1)[0].strip()\n",
    "    img = sample[IMAGE_COL].convert(\"RGB\")\n",
    "\n",
    "    try:\n",
    "        filled_img, detections = fill_detected_bboxes(\n",
    "            image       = img,\n",
    "            text        = prompt_trim,\n",
    "            inflate_pct = random.uniform(*INFLATE_RANGE),\n",
    "            fill_color  = FILL_COLOR,\n",
    "        )\n",
    "    except Exception as e:\n",
    "        print(f\"[WARN] idx {idx}: {e}\")\n",
    "        return idx, None, None, None\n",
    "\n",
    "    # annotated overlay\n",
    "    anno = img.copy()\n",
    "    anno = MASK_ANNOTATOR.annotate(anno, detections)\n",
    "    anno = BOX_ANNOTATOR.annotate(anno, detections)\n",
    "    anno = LABEL_ANNOTATOR.annotate(anno, detections)\n",
    "\n",
    "    # local QA files\n",
    "    filled_path = QA_DIR / f\"{idx:05d}_green.png\"\n",
    "    anno_path   = QA_DIR / f\"{idx:05d}_anno.png\"\n",
    "    json_path   = QA_DIR / f\"{idx:05d}_bbox.json\"\n",
    "\n",
    "    filled_img.save(filled_path)\n",
    "    anno.save(anno_path)\n",
    "    with open(json_path, \"w\") as fp:\n",
    "        json.dump({\"xyxy\": detections.xyxy.tolist()}, fp)\n",
    "\n",
    "    return idx, pil2bytes(filled_img), pil2bytes(anno), json_path.read_text()\n",
    "\n",
    "# ── threaded execution ----------------------------------------------\n",
    "with ThreadPoolExecutor(max_workers=MAX_WORKERS) as pool:\n",
    "    futures = {\n",
    "        pool.submit(process_row, (i, ds[i])): i\n",
    "        for i in range(len(ds))\n",
    "    }\n",
    "\n",
    "    for fut in tqdm(as_completed(futures),\n",
    "                    total=len(futures),\n",
    "                    desc=\"Threaded Florence+SAM\"):\n",
    "        i, bfill, anno, jtxt = fut.result()\n",
    "        if bfill is None:            # skipped rows\n",
    "            continue\n",
    "        bbox_filled_col[i] = bfill\n",
    "        annotated_col[i]   = anno\n",
    "        bbox_json_col[i]   = jtxt\n",
    "\n",
    "# ── build enriched dataset ------------------------------------------\n",
    "new_ds = ds.add_column(\"bbox_filled\", bbox_filled_col)\n",
    "new_ds = new_ds.add_column(\"annotated\",   annotated_col)\n",
    "new_ds = new_ds.add_column(\"bbox_json\",   bbox_json_col)\n",
    "\n",
    "new_ds = new_ds.cast_column(\"bbox_filled\", HFImage())\n",
    "new_ds = new_ds.cast_column(\"annotated\",   HFImage())\n",
    "\n",
    "print(\"✅ Completed. Review PNG/JSON in:\", QA_DIR.resolve())\n",
    "print(\"New dataset columns:\", new_ds.column_names)\n",
    "\n",
    "# optional push after QA\n",
    "# new_ds.push_to_hub(\"fotographerai/furniture_shortlisted_batch1_bboxfilled\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "871d5206",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rows: 5760\n",
      "Processing idx 0: Sofa (1500x1500)\n",
      "[WARN] idx 0 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 1: Sofa (1500x1500)\n",
      "[WARN] idx 1 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 2: Sofa (1500x1500)\n",
      "[WARN] idx 2 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 3: Couch (1500x1500)\n",
      "[WARN] idx 3 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 0 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 1 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 2 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 3 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 0 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 1 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 2 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 4: Couch (1500x1500)\n",
      "[WARN] idx 4 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 3 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 5: Sofa (1500x1500)\n",
      "[WARN] idx 5 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 6: Sofa (1500x1500)\n",
      "[WARN] idx 6 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 7: Sofa (1500x1500)\n",
      "[WARN] idx 7 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 4 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 5 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 6 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 7 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 4 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 5 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 6 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 8: Sofa (1500x1500)\n",
      "[WARN] idx 8 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 9: Sofa (1500x1500)\n",
      "[WARN] idx 9 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 7 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 10: Sofa (1500x1500)\n",
      "[WARN] idx 10 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 11: Sofa (1500x1500)\n",
      "[WARN] idx 11 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 8 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 9 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 10 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 11 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 8 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 9 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 10 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 12: Sofa (1500x1500)\n",
      "[WARN] idx 12 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 13: Sofa (1500x1500)\n",
      "[WARN] idx 13 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 11 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 14: Couch (1500x1500)\n",
      "[WARN] idx 14 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 15: Sofa (1500x1500)\n",
      "[WARN] idx 15 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 12 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 13 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 14 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 15 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 12 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 13 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 14 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 16: Sofa (1500x1500)\n",
      "[WARN] idx 16 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 15 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 17: Couch (1500x1500)\n",
      "[WARN] idx 17 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 18: Sofa (1500x1500)\n",
      "[WARN] idx 18 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 19: Couch (1500x1500)\n",
      "[WARN] idx 19 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 16 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 17 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 18 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 19 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 16 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 17 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 18 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 20: Sofa (1500x1500)\n",
      "[WARN] idx 20 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 19 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 22: Leather (1500x1500)\n",
      "[WARN] idx 22 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 21: Sofa (1500x1500)\n",
      "[WARN] idx 21 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 23: Leather (1500x1500)\n",
      "[WARN] idx 23 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 20 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 22 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 21 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 23 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 20 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 22 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 21 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 24: Sofa (1500x1500)\n",
      "[WARN] idx 24 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 23 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 25: Storage (1500x1500)\n",
      "[WARN] idx 25 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 26: Recliner (1500x1500)\n",
      "[WARN] idx 26 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 27: Chair (1500x1500)\n",
      "[WARN] idx 27 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 24 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 25 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 26 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 27 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 24 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 25 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 26 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 28: Couch (1500x1500)\n",
      "[WARN] idx 28 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 29: Chair (1500x1500)\n",
      "[WARN] idx 29 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 27 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 30: Chair (1500x1500)\n",
      "[WARN] idx 30 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 31: Sofa (1500x1500)\n",
      "[WARN] idx 31 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 28 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 29 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 30 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 31 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 28 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 29 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 30 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 32: Sofa (1500x1500)\n",
      "[WARN] idx 32 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 33: Chair (1500x1500)\n",
      "[WARN] idx 33 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 31 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 34: Sofa (1500x1500)\n",
      "[WARN] idx 34 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 35: Sofa (1500x1500)\n",
      "[WARN] idx 35 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 32 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 33 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 34 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 35 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 32 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 33 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 34 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 36: Sofa (1500x1500)\n",
      "[WARN] idx 36 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 35 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 37: Sofa (1500x1500)\n",
      "[WARN] idx 37 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 38: Leather (1500x1500)\n",
      "[WARN] idx 38 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 39: Sofa (1500x1500)\n",
      "[WARN] idx 39 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 36 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 37 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 38 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 39 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 36 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 37 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 38 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 40: Sofa (1500x1500)\n",
      "[WARN] idx 40 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 39 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 42: Sofa (1500x1500)\n",
      "[WARN] idx 42 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 41: Sofa (1500x1500)\n",
      "[WARN] idx 41 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 43: Leather (1500x1500)\n",
      "[WARN] idx 43 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 40 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 42 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 41 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 43 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 40 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 42 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 41 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 44: Sofa (1500x1500)\n",
      "[WARN] idx 44 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 43 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 45: Sofa (1500x1500)\n",
      "[WARN] idx 45 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 46: Sofa (1500x1500)\n",
      "[WARN] idx 46 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 47: Sofa (1500x1500)\n",
      "[WARN] idx 47 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 44 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 45 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 46 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 47 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 44 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 45 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 48: Sofa (1500x1500)\n",
      "[WARN] idx 48 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 46 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 47 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 49: Sofa (1500x1500)\n",
      "[WARN] idx 49 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 50: Leather (1500x1500)\n",
      "[WARN] idx 50 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 51: Brown (1500x1500)\n",
      "[WARN] idx 51 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 48 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 49 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 50 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 51 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 48 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 52: Sofa (1500x1500)\n",
      "[WARN] idx 52 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 49 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 50 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 51 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 53: Sofa (1500x1500)\n",
      "[WARN] idx 53 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 54: Sofa (1500x1500)\n",
      "[WARN] idx 54 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 55: Sofa (1500x1500)\n",
      "[WARN] idx 55 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 52 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 53 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 54 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 55 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 52 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 56: Sofa (1500x1500)\n",
      "[WARN] idx 56 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 53 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 54 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 55 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 57: Sofa (1500x1500)\n",
      "[WARN] idx 57 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 58: Sofa (1500x1500)\n",
      "[WARN] idx 58 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 59: Sofa (1500x1500)\n",
      "[WARN] idx 59 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 56 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 57 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 58 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 59 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 56 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 60: Chair (1500x1500)\n",
      "[WARN] idx 60 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 57 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 58 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 59 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 62: Sofa (1500x1500)\n",
      "[WARN] idx 62 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 61: Sofa (1500x1500)\n",
      "[WARN] idx 61 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 63: Sofa (1500x1500)\n",
      "[WARN] idx 63 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 60 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 62 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 61 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 63 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 60 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 64: Chair (1500x1500)\n",
      "[WARN] idx 64 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 62 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 61 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 63 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 65: Chair (1500x1500)\n",
      "[WARN] idx 65 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 66: Swivel (1500x1500)\n",
      "[WARN] idx 66 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 67: Armchair (1500x1500)\n",
      "[WARN] idx 67 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 64 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 65 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 66 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 67 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 64 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 68: Chair (1500x1500)\n",
      "[WARN] idx 68 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 65 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 66 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 67 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 69: Chair (1500x1500)\n",
      "[WARN] idx 69 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 70: Chair (1500x1500)\n",
      "[WARN] idx 70 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 71: Chair (1500x1500)\n",
      "[WARN] idx 71 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 68 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 69 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 70 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 71 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 68 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 72: Contemporary (1500x1500)\n",
      "[WARN] idx 72 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 69 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 70 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 73: Chair (1500x1500)\n",
      "[WARN] idx 73 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 71 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 74: Armchair (1500x1500)\n",
      "[WARN] idx 74 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 75: Chair (1500x1500)\n",
      "[WARN] idx 75 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 72 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 73 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 74 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 75 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 72 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 76: Chair (1500x1500)\n",
      "[WARN] idx 76 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 73 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 74 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 77: Chair (1500x1500)\n",
      "[WARN] idx 77 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 75 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 78: Chair (1500x1500)\n",
      "[WARN] idx 78 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 79: Rotating (1500x1500)\n",
      "[WARN] idx 79 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 76 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 77 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 78 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 79 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 76 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 80: Chair (1500x1500)\n",
      "[WARN] idx 80 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 77 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 78 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 81: Chair (1500x1500)\n",
      "[WARN] idx 81 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 79 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 82: Swivel (1500x1500)\n",
      "[WARN] idx 82 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 83: Chair (1500x1500)\n",
      "[WARN] idx 83 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 80 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 81 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 82 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 83 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 80 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 84: Armchair (1500x1500)\n",
      "[WARN] idx 84 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 81 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 82 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 85: Chair (1500x1500)\n",
      "[WARN] idx 85 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 83 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 86: Chair (1500x1500)\n",
      "[WARN] idx 86 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 87: Chair (1500x1500)\n",
      "[WARN] idx 87 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 84 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 85 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 86 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 87 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 84 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 88: Armchair (1500x1500)\n",
      "[WARN] idx 88 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 85 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 89: Chair (1500x1500)\n",
      "[WARN] idx 89 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 86 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 87 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 90: Chair (1500x1500)\n",
      "[WARN] idx 90 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 91: Chair (1500x1500)\n",
      "[WARN] idx 91 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 88 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 89 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 90 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 91 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 88 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "Processing idx 92: Chair (1500x1500)\n",
      "[WARN] idx 92 try 1/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 89 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 90 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 91 try 3/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 92 try 2/3: name 'fill_detected_bboxes' is not defined\n",
      "[WARN] idx 92 try 3/3: name 'fill_detected_bboxes' is not defined\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[1], line 104\u001b[0m\n\u001b[1;32m    102\u001b[0m fails \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m    103\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m ThreadPoolExecutor(MAX_WORKERS) \u001b[38;5;28;01mas\u001b[39;00m pool:\n\u001b[0;32m--> 104\u001b[0m     fut2idx \u001b[38;5;241m=\u001b[39m {pool\u001b[38;5;241m.\u001b[39msubmit(process_row, i, ds[i]): i \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(\u001b[38;5;28mlen\u001b[39m(ds))}\n\u001b[1;32m    105\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m fut \u001b[38;5;129;01min\u001b[39;00m tqdm(as_completed(fut2idx), total\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mlen\u001b[39m(fut2idx),\n\u001b[1;32m    106\u001b[0m                     desc\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFlorence+SAM\u001b[39m\u001b[38;5;124m\"\u001b[39m):\n\u001b[1;32m    107\u001b[0m         i \u001b[38;5;241m=\u001b[39m fut2idx[fut]\n",
      "Cell \u001b[0;32mIn[1], line 104\u001b[0m, in \u001b[0;36m<dictcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    102\u001b[0m fails \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m    103\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m ThreadPoolExecutor(MAX_WORKERS) \u001b[38;5;28;01mas\u001b[39;00m pool:\n\u001b[0;32m--> 104\u001b[0m     fut2idx \u001b[38;5;241m=\u001b[39m {pool\u001b[38;5;241m.\u001b[39msubmit(process_row, i, \u001b[43mds\u001b[49m\u001b[43m[\u001b[49m\u001b[43mi\u001b[49m\u001b[43m]\u001b[49m): i \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(\u001b[38;5;28mlen\u001b[39m(ds))}\n\u001b[1;32m    105\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m fut \u001b[38;5;129;01min\u001b[39;00m tqdm(as_completed(fut2idx), total\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mlen\u001b[39m(fut2idx),\n\u001b[1;32m    106\u001b[0m                     desc\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFlorence+SAM\u001b[39m\u001b[38;5;124m\"\u001b[39m):\n\u001b[1;32m    107\u001b[0m         i \u001b[38;5;241m=\u001b[39m fut2idx[fut]\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/datasets/arrow_dataset.py:2777\u001b[0m, in \u001b[0;36mDataset.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   2775\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21m__getitem__\u001b[39m(\u001b[38;5;28mself\u001b[39m, key):  \u001b[38;5;66;03m# noqa: F811\u001b[39;00m\n\u001b[1;32m   2776\u001b[0m \u001b[38;5;250m    \u001b[39m\u001b[38;5;124;03m\"\"\"Can be used to index columns (by string names) or rows (by integer index or iterable of indices or bools).\"\"\"\u001b[39;00m\n\u001b[0;32m-> 2777\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_getitem\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/datasets/arrow_dataset.py:2762\u001b[0m, in \u001b[0;36mDataset._getitem\u001b[0;34m(self, key, **kwargs)\u001b[0m\n\u001b[1;32m   2760\u001b[0m formatter \u001b[38;5;241m=\u001b[39m get_formatter(format_type, features\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_info\u001b[38;5;241m.\u001b[39mfeatures, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mformat_kwargs)\n\u001b[1;32m   2761\u001b[0m pa_subtable \u001b[38;5;241m=\u001b[39m query_table(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_data, key, indices\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_indices)\n\u001b[0;32m-> 2762\u001b[0m formatted_output \u001b[38;5;241m=\u001b[39m \u001b[43mformat_table\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   2763\u001b[0m \u001b[43m    \u001b[49m\u001b[43mpa_subtable\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mkey\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mformatter\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mformatter\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mformat_columns\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mformat_columns\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43moutput_all_columns\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moutput_all_columns\u001b[49m\n\u001b[1;32m   2764\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   2765\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m formatted_output\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/datasets/formatting/formatting.py:653\u001b[0m, in \u001b[0;36mformat_table\u001b[0;34m(table, key, formatter, format_columns, output_all_columns)\u001b[0m\n\u001b[1;32m    651\u001b[0m python_formatter \u001b[38;5;241m=\u001b[39m PythonFormatter(features\u001b[38;5;241m=\u001b[39mformatter\u001b[38;5;241m.\u001b[39mfeatures)\n\u001b[1;32m    652\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m format_columns \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m--> 653\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mformatter\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpa_table\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mquery_type\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mquery_type\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    654\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m query_type \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcolumn\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[1;32m    655\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m key \u001b[38;5;129;01min\u001b[39;00m format_columns:\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/datasets/formatting/formatting.py:406\u001b[0m, in \u001b[0;36mFormatter.__call__\u001b[0;34m(self, pa_table, query_type)\u001b[0m\n\u001b[1;32m    404\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21m__call__\u001b[39m(\u001b[38;5;28mself\u001b[39m, pa_table: pa\u001b[38;5;241m.\u001b[39mTable, query_type: \u001b[38;5;28mstr\u001b[39m) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Union[RowFormat, ColumnFormat, BatchFormat]:\n\u001b[1;32m    405\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m query_type \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mrow\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[0;32m--> 406\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mformat_row\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpa_table\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    407\u001b[0m     \u001b[38;5;28;01melif\u001b[39;00m query_type \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcolumn\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[1;32m    408\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mformat_column(pa_table)\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/datasets/formatting/formatting.py:454\u001b[0m, in \u001b[0;36mPythonFormatter.format_row\u001b[0;34m(self, pa_table)\u001b[0m\n\u001b[1;32m    452\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mlazy:\n\u001b[1;32m    453\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m LazyRow(pa_table, \u001b[38;5;28mself\u001b[39m)\n\u001b[0;32m--> 454\u001b[0m row \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpython_arrow_extractor\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mextract_row\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpa_table\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    455\u001b[0m row \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mpython_features_decoder\u001b[38;5;241m.\u001b[39mdecode_row(row)\n\u001b[1;32m    456\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m row\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/datasets/formatting/formatting.py:143\u001b[0m, in \u001b[0;36mPythonArrowExtractor.extract_row\u001b[0;34m(self, pa_table)\u001b[0m\n\u001b[1;32m    142\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21mextract_row\u001b[39m(\u001b[38;5;28mself\u001b[39m, pa_table: pa\u001b[38;5;241m.\u001b[39mTable) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m \u001b[38;5;28mdict\u001b[39m:\n\u001b[0;32m--> 143\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m _unnest(\u001b[43mpa_table\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mto_pydict\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m)\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/pyarrow/table.pxi:2308\u001b[0m, in \u001b[0;36mpyarrow.lib._Tabular.to_pydict\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/pyarrow/table.pxi:1380\u001b[0m, in \u001b[0;36mpyarrow.lib.ChunkedArray.to_pylist\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/pyarrow/array.pxi:1775\u001b[0m, in \u001b[0;36mpyarrow.lib.Array.to_pylist\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/site-packages/pyarrow/scalar.pxi:1035\u001b[0m, in \u001b[0;36mpyarrow.lib.StructScalar.as_py\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/_collections_abc.py:836\u001b[0m, in \u001b[0;36mMapping.keys\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    833\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    834\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m\n\u001b[0;32m--> 836\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21mkeys\u001b[39m(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m    837\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mD.keys() -> a set-like object providing a view on D\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124ms keys\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    838\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m KeysView(\u001b[38;5;28mself\u001b[39m)\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "# furniture_bbox_fast.py  ────────────────────────────────────────────\n",
    "# Florence-2 + SAM-2 preprocessing with square-padding + auto-retries\n",
    "# -------------------------------------------------------------------\n",
    "import os, io, json, random, time\n",
    "from pathlib import Path\n",
    "from concurrent.futures import ThreadPoolExecutor, as_completed\n",
    "\n",
    "from tqdm.auto import tqdm\n",
    "from datasets import load_dataset, Image as HFImage, disable_progress_bar\n",
    "from PIL import Image, ImageOps\n",
    "\n",
    "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
    "disable_progress_bar()\n",
    "\n",
    "# ═══════════════════ CONFIG ════════════════════════════════════════\n",
    "DATASET_NAME  = \"fotographerai/furniture_captioned_segment_prompt\"\n",
    "SPLIT         = \"train\"\n",
    "IMAGE_COL     = \"img2\"\n",
    "PROMPT_COL    = \"segmenting_prompt\"\n",
    "\n",
    "INFLATE_RANGE = (0.01, 0.05)\n",
    "FILL_COLOR    = \"#00FF00\"\n",
    "\n",
    "QA_DIR        = Path(\"bbox_review_recaptioned\"); QA_DIR.mkdir(exist_ok=True)\n",
    "MAX_WORKERS   = 4\n",
    "MAX_RETRIES   = 3\n",
    "RETRY_SLEEP   = 0.3          # seconds between retries\n",
    "TARGET_SIDE   = 1500         # final square size (px)\n",
    "\n",
    "FAILED_LOG    = QA_DIR / \"failed_rows.jsonl\"\n",
    "\n",
    "# ═══════════════════ LOAD DATA ═════════════════════════════════════\n",
    "ds = load_dataset(DATASET_NAME, split=SPLIT, streaming=False)\n",
    "print(\"Rows:\", len(ds))\n",
    "\n",
    "filled_col, anno_col, json_col = [None]*len(ds), [None]*len(ds), [None]*len(ds)\n",
    "\n",
    "def pil2bytes(im: Image.Image):\n",
    "    buf = io.BytesIO(); im.save(buf, format=\"PNG\"); buf.seek(0)\n",
    "    return {\"bytes\": buf.getvalue(), \"path\": None}\n",
    "\n",
    "# map some categories → explicit prompts\n",
    "PROMPT_MAP = {\n",
    "\n",
    "}\n",
    "\n",
    "# pad & resize to TARGET_SIDE × TARGET_SIDE\n",
    "def make_square(im: Image.Image, side: int = TARGET_SIDE) -> Image.Image:\n",
    "    # keep aspect ratio, add padding to reach 'side'\n",
    "    im = ImageOps.contain(im, (side, side))          # resize w/ ratio\n",
    "    pad_w = side - im.width\n",
    "    pad_h = side - im.height\n",
    "    # center pad with reflection to avoid solid borders that might confuse model\n",
    "    im = ImageOps.expand(im,\n",
    "                         border=(pad_w//2, pad_h//2,\n",
    "                                 pad_w - pad_w//2, pad_h - pad_h//2),\n",
    "                         fill=im.getpixel((0,0)))\n",
    "    return im\n",
    "\n",
    "# ── worker with square-padding + retries ───────────────────────────\n",
    "def process_row(i, sample):\n",
    "    prompt = PROMPT_MAP.get(sample[PROMPT_COL],\n",
    "                            sample[PROMPT_COL].split(\",\", 1)[0].strip())\n",
    "    img0   = sample[IMAGE_COL].convert(\"RGB\")\n",
    "    img_sq = make_square(img0)\n",
    "    \n",
    "    print(f\"Processing idx {i}: {prompt} ({img_sq.size[0]}x{img_sq.size[1]})\")\n",
    "\n",
    "    for attempt in range(1, MAX_RETRIES+1):\n",
    "        try:\n",
    "            filled, dets = fill_detected_bboxes(\n",
    "                image       = img_sq,\n",
    "                text        = prompt,\n",
    "                inflate_pct = random.uniform(*INFLATE_RANGE),\n",
    "                fill_color  = FILL_COLOR,\n",
    "            )\n",
    "            if dets is None or len(dets.xyxy) == 0:\n",
    "                raise ValueError(\"no detections\")\n",
    "\n",
    "            anno = img_sq.copy()\n",
    "            anno = MASK_ANNOTATOR.annotate(anno, dets)\n",
    "            anno = BOX_ANNOTATOR.annotate(anno, dets)\n",
    "            anno = LABEL_ANNOTATOR.annotate(anno, dets)\n",
    "\n",
    "            # QA artefacts\n",
    "            filled.save(QA_DIR / f\"{i:06d}_green.png\")\n",
    "            anno  .save(QA_DIR / f\"{i:06d}_anno.png\")\n",
    "            bbox_path = QA_DIR / f\"{i:06d}_bbox.json\"\n",
    "            bbox_path.write_text(json.dumps({\"xyxy\": dets.xyxy.tolist()}))\n",
    "\n",
    "            return (\"ok\", pil2bytes(filled), pil2bytes(anno),\n",
    "                    bbox_path.read_text())\n",
    "\n",
    "        except Exception as e:\n",
    "            print(f\"[WARN] idx {i} try {attempt}/{MAX_RETRIES}: {e}\")\n",
    "            if attempt < MAX_RETRIES:\n",
    "                time.sleep(RETRY_SLEEP)\n",
    "            else:\n",
    "                return (\"fail\", str(e))\n",
    "\n",
    "# ═══════════════════ THREAD POOL ═══════════════════════════════════\n",
    "fails = 0\n",
    "with ThreadPoolExecutor(MAX_WORKERS) as pool:\n",
    "    fut2idx = {pool.submit(process_row, i, ds[i]): i for i in range(len(ds))}\n",
    "    for fut in tqdm(as_completed(fut2idx), total=len(fut2idx),\n",
    "                    desc=\"Florence+SAM\"):\n",
    "        i = fut2idx[fut]\n",
    "        result = fut.result()\n",
    "        if result[0] == \"ok\":\n",
    "            _, bfill, anno, jtxt = result\n",
    "            filled_col[i], anno_col[i], json_col[i] = bfill, anno, jtxt\n",
    "        else:\n",
    "            fails += 1\n",
    "            with FAILED_LOG.open(\"a\") as fp:\n",
    "                fp.write(json.dumps({\"idx\": i, \"reason\": result[1]})+\"\\n\")\n",
    "\n",
    "print(f\"❌ permanently failed rows: {fails}\")\n",
    "\n",
    "# ═══════════════════ BUILD SUCCESS DATASET ═════════════════════════\n",
    "keep = [i for i,x in enumerate(filled_col) if x is not None]\n",
    "new_ds = ds.select(keep)\n",
    "new_ds = new_ds.add_column(\"bbox_filled\", [filled_col[i] for i in keep])\n",
    "new_ds = new_ds.add_column(\"annotated\",   [anno_col[i]   for i in keep])\n",
    "new_ds = new_ds.add_column(\"bbox_json\",   [json_col[i]   for i in keep])\n",
    "new_ds = new_ds.cast_column(\"bbox_filled\", HFImage())\n",
    "new_ds = new_ds.cast_column(\"annotated\",   HFImage())\n",
    "\n",
    "print(f\"✅ successes: {len(new_ds)} / {len(ds)}\")\n",
    "print(\"QA artefacts in:\", QA_DIR.resolve())\n",
    "print(\"Columns:\", new_ds.column_names)\n",
    "\n",
    "# new_ds.push_to_hub(\"fotographerai/surround_omnibatch6_bboxfilled\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3c4dc9be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d1a6b002c48a441795b3967d917301cc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Resolving data files:   0%|          | 0/38 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "531845f59f6b4a84a678967f422b4cec",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Resolving data files:   0%|          | 0/38 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "576b1110d4ed459186012f2f37af19df",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Loading dataset shards:   0%|          | 0/26 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Original rows: 5760\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e021b42b0ba14d15b6f62e8b153b381f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Re-assembling:   0%|          | 0/5760 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "reassemble_bbox_dataset.py\n",
    "Re-build bbox_filled / annotated / bbox_json columns from QA folder.\n",
    "Missing rows receive a 1500×1500 green mask & full-frame bbox.\n",
    "\"\"\"\n",
    "from pathlib import Path\n",
    "import json, io, os\n",
    "from tqdm.auto import tqdm\n",
    "from datasets import load_dataset, Dataset, Image as HFImage\n",
    "from PIL import Image\n",
    "\n",
    "# ═════ CONFIG ═══════════════════════════════════════════════════════\n",
    "DATASET_NAME  = \"fotographerai/furniture_captioned_segment_prompt\"\n",
    "SPLIT         = \"train\"\n",
    "IMAGE_COL     = \"img2\"            # keep original image column(s)\n",
    "ID_COL        = \"id\"              # unique identifier, or use row index\n",
    "QA_DIR        = Path(\"bbox_review_recaptioned\")   # where *_green.png live\n",
    "TARGET_SIDE   = 1500\n",
    "GREEN_RGB     = (0, 255, 0)\n",
    "\n",
    "# ═════ LOAD ORIGINAL DATASET (no internet after first cache) ════════\n",
    "ds = load_dataset(DATASET_NAME, split=SPLIT, streaming=False)\n",
    "print(\"Original rows:\", len(ds))\n",
    "\n",
    "# ═════ HELPERS ══════════════════════════════════════════════════════\n",
    "def img_ref(path: Path):\n",
    "    \"\"\"Return datasets “image” dict referring to a file on disk.\"\"\"\n",
    "    return {\"path\": str(path), \"bytes\": None}\n",
    "\n",
    "def make_green_mask(path: Path, side: int = TARGET_SIDE):\n",
    "    \"\"\"Create and save a solid-green PNG mask.\"\"\"\n",
    "    if not path.exists():\n",
    "        img = Image.new(\"RGB\", (side, side), GREEN_RGB)\n",
    "        img.save(path)\n",
    "\n",
    "def full_bbox_json(path: Path, side: int = TARGET_SIDE):\n",
    "    if not path.exists():\n",
    "        with open(path, \"w\") as fp:\n",
    "            json.dump({\"xyxy\": [[0, 0, side, side]]}, fp)\n",
    "\n",
    "# ═════ BUILD RECORDS LIST ═══════════════════════════════════════════\n",
    "records = []\n",
    "for i in tqdm(range(len(ds)), desc=\"Re-assembling\"):\n",
    "    # file paths for this row\n",
    "    green_png = QA_DIR / f\"{i:06d}_green.png\"\n",
    "    anno_png  = QA_DIR / f\"{i:06d}_anno.png\"\n",
    "    bbox_js   = QA_DIR / f\"{i:06d}_bbox.json\"\n",
    "\n",
    "    if not green_png.exists() or not anno_png.exists() or not bbox_js.exists():\n",
    "        # create fallback artefacts\n",
    "        mask_png = QA_DIR / f\"{i:06d}_mask.png\"\n",
    "        make_green_mask(mask_png)\n",
    "        green_png = anno_png = mask_png\n",
    "        full_bbox_json(bbox_js)\n",
    "\n",
    "    records.append({\n",
    "        **{k: ds[i][k] for k in ds[i].keys()},   # copy original cols\n",
    "        \"bbox_filled\": img_ref(green_png),\n",
    "        \"annotated\"  : img_ref(anno_png),\n",
    "        \"bbox_json\"  : bbox_js.read_text(),\n",
    "    })\n",
    "\n",
    "# ═════ DATASET FROM LIST ════════════════════════════════════════════\n",
    "new_ds = Dataset.from_list(records)\n",
    "new_ds = new_ds.cast_column(\"bbox_filled\", HFImage())\n",
    "new_ds = new_ds.cast_column(\"annotated\",   HFImage())\n",
    "\n",
    "print(\"✅ rebuilt rows:\", len(new_ds))\n",
    "print(\"Columns:\", new_ds.column_names)\n",
    "\n",
    "# (optional) push\n",
    "# new_ds.push_to_hub(\"fotographerai/furniture_bboxfilled_rebuild\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7dc15f5b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "863622dbf6c9487a815810070672f6f2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00000-of-00046.parquet:   0%|          | 0.00/587M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3432a28e5c064bfb90a49ae3d1c79b5d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00001-of-00046.parquet:   0%|          | 0.00/626M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ca2ce71106954e85b24f8385b5b84889",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00002-of-00046.parquet:   0%|          | 0.00/555M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d493cf4ebb2c40fc804ddef960ed2ec1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00003-of-00046.parquet:   0%|          | 0.00/619M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1fac74960e5f46febbb34b48ba51eab1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00004-of-00046.parquet:   0%|          | 0.00/637M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0a91fc58f9e24b20b5c89fe863704cee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00005-of-00046.parquet:   0%|          | 0.00/583M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "9bb9ed8e3f3b442ebd7c98ed82c2ceae",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00006-of-00046.parquet:   0%|          | 0.00/559M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ad0e8c7743cf456db036c72845e12a25",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00007-of-00046.parquet:   0%|          | 0.00/406M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "77e3f3eae0f8447bb889891fac711b47",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00008-of-00046.parquet:   0%|          | 0.00/294M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "99e08b20a5d84afbac1ff2ee31386b76",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00009-of-00046.parquet:   0%|          | 0.00/354M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "91b9fe1f17c644028ef20c7ab2ba97ad",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00010-of-00046.parquet:   0%|          | 0.00/362M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cb76ddd7095f43e08a7aa0ea9eedd560",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00011-of-00046.parquet:   0%|          | 0.00/387M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "618f7a6b7cbe499fbc8e5ef87cdbf9d1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00012-of-00046.parquet:   0%|          | 0.00/493M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e9a8738263534c0e8b1bbab2e6dfe495",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00013-of-00046.parquet:   0%|          | 0.00/528M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ce9fb5814f6948688c91aedc55e72b59",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00014-of-00046.parquet:   0%|          | 0.00/470M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "83496f8bc3c2425687004238d15f9e39",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00015-of-00046.parquet:   0%|          | 0.00/455M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4852190ac54b42b38add9e32310493f4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00016-of-00046.parquet:   0%|          | 0.00/481M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed814e8637e348b585043492c1b9e791",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00017-of-00046.parquet:   0%|          | 0.00/455M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "54f6d0a61b09465d959b25124c3ff98f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00018-of-00046.parquet:   0%|          | 0.00/468M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6ce1bc2018bc489e9af0c33bbc5afc86",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00019-of-00046.parquet:   0%|          | 0.00/455M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f93039bb4a58456f9df515b1bdbe3932",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00020-of-00046.parquet:   0%|          | 0.00/442M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bdcae62083bb44a3b66ba78df537adde",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00021-of-00046.parquet:   0%|          | 0.00/465M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "267a27ba6d314649ab3244418fe3186a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00022-of-00046.parquet:   0%|          | 0.00/437M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da2984d6298e42c2ac8a780baa539d58",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00023-of-00046.parquet:   0%|          | 0.00/503M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a1a6d751d15247859fadb649b825ddf4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00024-of-00046.parquet:   0%|          | 0.00/490M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bf5d6a580a4f4b88bc43961d1e9b442d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00025-of-00046.parquet:   0%|          | 0.00/490M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2d5df099d34a4ed885ab937e9c28e717",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00026-of-00046.parquet:   0%|          | 0.00/490M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8b71a9648dbd4ffea33af1a125a4fde8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00027-of-00046.parquet:   0%|          | 0.00/459M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00a02f36fd2d439ebf7363e09cd91afe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00028-of-00046.parquet:   0%|          | 0.00/482M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "52222e83f8084a5f9a2fd6d6d19473ab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00029-of-00046.parquet:   0%|          | 0.00/508M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f75d038bf4404c9e8eaadb368b330088",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00030-of-00046.parquet:   0%|          | 0.00/435M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "96d1fca2c47c4f048eea596df76aec13",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00031-of-00046.parquet:   0%|          | 0.00/388M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "85a2cceee42d4db4859bb1fe1c27bce1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00032-of-00046.parquet:   0%|          | 0.00/408M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "38a7065e42534b8b817457dbda415376",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00033-of-00046.parquet:   0%|          | 0.00/426M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a5e43870359149188f8135e1e6651a76",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00034-of-00046.parquet:   0%|          | 0.00/582M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e7bfdbf132bb4aedb0c39da5b0c2f00b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00035-of-00046.parquet:   0%|          | 0.00/457M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b980476558c94f18a764b2a882de1e1e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00036-of-00046.parquet:   0%|          | 0.00/609M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e4a2adb228f547b89448f6b101805b52",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00037-of-00046.parquet:   0%|          | 0.00/412M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d748ab068a1242d2b8c9a9a5b2a70437",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00038-of-00046.parquet:   0%|          | 0.00/411M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ae4a3cecd88744338144ac87d55eb79b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00039-of-00046.parquet:   0%|          | 0.00/419M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "343b564c620f44b78945657c9fb22136",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00040-of-00046.parquet:   0%|          | 0.00/389M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ea60b09267374c749bfe307df45dd7dc",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00041-of-00046.parquet:   0%|          | 0.00/472M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cae69a3f2e2e488d9a2318302f493369",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00042-of-00046.parquet:   0%|          | 0.00/673M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "310388208ecd46989478d7d2e3412836",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00043-of-00046.parquet:   0%|          | 0.00/495M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a09cb883f10549929d2ee63ddd682f7c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00044-of-00046.parquet:   0%|          | 0.00/454M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "73bde28ddb8c40b58a10d4fc3b49f1e0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "train-00045-of-00046.parquet:   0%|          | 0.00/509M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded 5760 rows from fotographerai/furniture_bboxfilled_rebuild[train]\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "19296d23a97547d3a0408c2c69a50218",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Indexing:   0%|          | 0/5760 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Categories: 21\n",
      "Total sampled rows: 63\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "653022984b1b43ddac76547d4508ed42",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Copying images:   0%|          | 0/63 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "TypeError",
     "evalue": "expected str, bytes or os.PathLike object, not PngImageFile",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[2], line 60\u001b[0m\n\u001b[1;32m     58\u001b[0m \u001b[38;5;66;03m# src_info may already be {\"path\": …} or full Image object; handle both\u001b[39;00m\n\u001b[1;32m     59\u001b[0m src_path \u001b[38;5;241m=\u001b[39m src_info[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpath\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(src_info, \u001b[38;5;28mdict\u001b[39m) \u001b[38;5;28;01melse\u001b[39;00m src_info\n\u001b[0;32m---> 60\u001b[0m dst_path \u001b[38;5;241m=\u001b[39m OUTPUT_DIR \u001b[38;5;241m/\u001b[39m \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mrow_idx\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m06d\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m_\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[43mPath\u001b[49m\u001b[43m(\u001b[49m\u001b[43msrc_path\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241m.\u001b[39mname\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m     61\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m COPY_METHOD \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mlink\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[1;32m     62\u001b[0m     dst_path\u001b[38;5;241m.\u001b[39mhardlink_to(src_path) \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m dst_path\u001b[38;5;241m.\u001b[39mexists() \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/pathlib.py:960\u001b[0m, in \u001b[0;36mPath.__new__\u001b[0;34m(cls, *args, **kwargs)\u001b[0m\n\u001b[1;32m    958\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mcls\u001b[39m \u001b[38;5;129;01mis\u001b[39;00m Path:\n\u001b[1;32m    959\u001b[0m     \u001b[38;5;28mcls\u001b[39m \u001b[38;5;241m=\u001b[39m WindowsPath \u001b[38;5;28;01mif\u001b[39;00m os\u001b[38;5;241m.\u001b[39mname \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnt\u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m PosixPath\n\u001b[0;32m--> 960\u001b[0m \u001b[38;5;28mself\u001b[39m \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mcls\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_from_parts\u001b[49m\u001b[43m(\u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    961\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_flavour\u001b[38;5;241m.\u001b[39mis_supported:\n\u001b[1;32m    962\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mNotImplementedError\u001b[39;00m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcannot instantiate \u001b[39m\u001b[38;5;132;01m%r\u001b[39;00m\u001b[38;5;124m on your system\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    963\u001b[0m                               \u001b[38;5;241m%\u001b[39m (\u001b[38;5;28mcls\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m,))\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/pathlib.py:594\u001b[0m, in \u001b[0;36mPurePath._from_parts\u001b[0;34m(cls, args)\u001b[0m\n\u001b[1;32m    589\u001b[0m \u001b[38;5;129m@classmethod\u001b[39m\n\u001b[1;32m    590\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[38;5;21m_from_parts\u001b[39m(\u001b[38;5;28mcls\u001b[39m, args):\n\u001b[1;32m    591\u001b[0m     \u001b[38;5;66;03m# We need to call _parse_args on the instance, so as to get the\u001b[39;00m\n\u001b[1;32m    592\u001b[0m     \u001b[38;5;66;03m# right flavour.\u001b[39;00m\n\u001b[1;32m    593\u001b[0m     \u001b[38;5;28mself\u001b[39m \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mobject\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__new__\u001b[39m(\u001b[38;5;28mcls\u001b[39m)\n\u001b[0;32m--> 594\u001b[0m     drv, root, parts \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_parse_args\u001b[49m\u001b[43m(\u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    595\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_drv \u001b[38;5;241m=\u001b[39m drv\n\u001b[1;32m    596\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_root \u001b[38;5;241m=\u001b[39m root\n",
      "File \u001b[0;32m/opt/conda/envs/zen2c/lib/python3.10/pathlib.py:578\u001b[0m, in \u001b[0;36mPurePath._parse_args\u001b[0;34m(cls, args)\u001b[0m\n\u001b[1;32m    576\u001b[0m     parts \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m a\u001b[38;5;241m.\u001b[39m_parts\n\u001b[1;32m    577\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m--> 578\u001b[0m     a \u001b[38;5;241m=\u001b[39m \u001b[43mos\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfspath\u001b[49m\u001b[43m(\u001b[49m\u001b[43ma\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    579\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(a, \u001b[38;5;28mstr\u001b[39m):\n\u001b[1;32m    580\u001b[0m         \u001b[38;5;66;03m# Force-cast str subclasses to str (issue #21127)\u001b[39;00m\n\u001b[1;32m    581\u001b[0m         parts\u001b[38;5;241m.\u001b[39mappend(\u001b[38;5;28mstr\u001b[39m(a))\n",
      "\u001b[0;31mTypeError\u001b[0m: expected str, bytes or os.PathLike object, not PngImageFile"
     ]
    }
   ],
   "source": [
    "#!/usr/bin/env python\n",
    "\"\"\"\n",
    "sample_per_category.py  ────────────────────────────────────────────────\n",
    "Sample K images per category from a Hugging Face dataset (local or hub).\n",
    "\n",
    "• Keeps original rows intact (no re-export of PNG bytes).\n",
    "• Produces:\n",
    "      ├─ sample_manifest.csv   # (dataset_idx, category, image_path …)\n",
    "      └─ sampled_images/       # (optional) hard-linked / copied PNGs\n",
    "\n",
    "Adjust the CONFIG section and run.\n",
    "\"\"\"\n",
    "import random, csv, shutil\n",
    "from pathlib import Path\n",
    "from collections import defaultdict\n",
    "from tqdm.auto import tqdm\n",
    "from datasets import load_dataset, Dataset, disable_progress_bar\n",
    "\n",
    "disable_progress_bar()\n",
    "\n",
    "# ═════ CONFIG ─────────────────────────────────────────────────══════\n",
    "DATASET_ID      = \"fotographerai/furniture_bboxfilled_rebuild\"  # hub path OR local dir\n",
    "SPLIT           = \"train\"\n",
    "CATEGORY_COL    = \"category\"        # e.g. \"category\"  (string)\n",
    "IMAGE_COL       = \"annotated\"     # column to copy (HF image type or path dict)\n",
    "K_PER_CATEGORY  = 3                 # how many to sample\n",
    "OUTPUT_DIR      = Path(\"sampled_images\")    # set None to skip copying\n",
    "MANIFEST_CSV    = Path(\"sample_manifest.csv\")\n",
    "SEED            = 42\n",
    "COPY_METHOD     = \"link\"            # \"link\", \"copy\" or \"symlink\"\n",
    "# ════════════════════════════════════════════════════════════════════\n",
    "\n",
    "random.seed(SEED)\n",
    "ds = load_dataset(DATASET_ID, split=SPLIT, streaming=False)\n",
    "print(f\"Loaded {len(ds)} rows from {DATASET_ID}[{SPLIT}]\")\n",
    "\n",
    "# ── group indices by category ───────────────────────────────────────\n",
    "by_cat: dict[str, list[int]] = defaultdict(list)\n",
    "for idx, row in enumerate(tqdm(ds, desc=\"Indexing\")):\n",
    "    by_cat[row[CATEGORY_COL]].append(idx)\n",
    "\n",
    "print(\"Categories:\", len(by_cat))\n",
    "\n",
    "# ── sample -------------------------------------------------------------------\n",
    "samples: list[tuple[int,dict]] = []      # (row_idx, row_dict)\n",
    "for cat, idxs in by_cat.items():\n",
    "    chosen = random.sample(idxs, min(K_PER_CATEGORY, len(idxs)))\n",
    "    for i in chosen:\n",
    "        samples.append((i, ds[i]))\n",
    "\n",
    "print(f\"Total sampled rows: {len(samples)}\")\n",
    "\n",
    "# ── optional: copy images -----------------------------------------------------\n",
    "if OUTPUT_DIR:\n",
    "    OUTPUT_DIR.mkdir(exist_ok=True)\n",
    "    for row_idx, row in tqdm(samples, desc=\"Copying images\"):\n",
    "        src_info = row[IMAGE_COL]\n",
    "        # src_info may already be {\"path\": …} or full Image object; handle both\n",
    "        src_path = src_info[\"path\"] if isinstance(src_info, dict) else src_info\n",
    "        dst_path = OUTPUT_DIR / f\"{row_idx:06d}_{Path(src_path).name}\"\n",
    "        if COPY_METHOD == \"link\":\n",
    "            dst_path.hardlink_to(src_path) if not dst_path.exists() else None\n",
    "        elif COPY_METHOD == \"symlink\":\n",
    "            dst_path.symlink_to(src_path)   if not dst_path.exists() else None\n",
    "        else:  # \"copy\"\n",
    "            shutil.copy2(src_path, dst_path)\n",
    "\n",
    "# ── write CSV manifest --------------------------------------------------------\n",
    "with MANIFEST_CSV.open(\"w\", newline=\"\") as fp:\n",
    "    writer = csv.writer(fp)\n",
    "    writer.writerow([\"row_idx\", CATEGORY_COL, \"image_path\"])\n",
    "    for idx, row in samples:\n",
    "        img_path = row[IMAGE_COL][\"path\"] if isinstance(row[IMAGE_COL], dict) else row[IMAGE_COL]\n",
    "        writer.writerow([idx, row[CATEGORY_COL], img_path])\n",
    "\n",
    "print(f\"✅ Manifest written to {MANIFEST_CSV.resolve()}\")\n",
    "if OUTPUT_DIR:\n",
    "    print(f\"📂 Images saved / linked in {OUTPUT_DIR.resolve()}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4072ec64",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "21 unique categories:\n",
      " ['Furniture_Interior_batch3', 'Furniture_nordic_nest', 'Illumination_nordic_nest', 'Interior_goods', 'Interior_nordic_nest', 'Kitchen_bathroom', 'Tableware', 'beds', 'castlery-Accessories', 'castlery-Beds', 'castlery-Chairs', 'castlery-Furniture Sets', 'castlery-Outdoor Furniture', 'castlery-Sofas', 'castlery-Storage', 'castlery-Tables', 'decoration-vase_batch3', 'desks', 'lamps_chandeliers_batch2', 'office_chairs', 'sofas_lazy_chairs']\n"
     ]
    }
   ],
   "source": [
    "# ds is the Dataset you've loaded\n",
    "categories: list[str] = sorted(set(ds[\"category\"]))   # or whatever your column is called\n",
    "print(f\"{len(categories)} unique categories:\\n\", categories)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "zen2c",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}