Training in progress, epoch 1

config.json

@@ -0,0 +1,121 @@
+{
+  "_name_or_path": "jozhang97/deta-swin-large-o365",
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "architectures": [
+    "DetaForObjectDetection"
+  ],
+  "assign_first_stage": true,
+  "attention_dropout": 0.0,
+  "auxiliary_loss": false,
+  "backbone_config": {
+    "attention_probs_dropout_prob": 0.0,
+    "depths": [
+      2,
+      2,
+      18,
+      2
+    ],
+    "drop_path_rate": 0.1,
+    "embed_dim": 192,
+    "encoder_stride": 32,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 1536,
+    "image_size": 224,
+    "initializer_range": 0.02,
+    "layer_norm_eps": 1e-05,
+    "mlp_ratio": 4.0,
+    "model_type": "swin",
+    "num_heads": [
+      6,
+      12,
+      24,
+      48
+    ],
+    "num_layers": 4,
+    "out_features": [
+      "stage2",
+      "stage3",
+      "stage4"
+    ],
+    "out_indices": [
+      2,
+      3,
+      4
+    ],
+    "patch_size": 4,
+    "path_norm": true,
+    "qkv_bias": true,
+    "use_absolute_embeddings": false,
+    "window_size": 12
+  },
+  "bbox_cost": 5,
+  "bbox_loss_coefficient": 5,
+  "class_cost": 1,
+  "d_model": 256,
+  "decoder_attention_heads": 8,
+  "decoder_ffn_dim": 2048,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_n_points": 4,
+  "dice_loss_coefficient": 1,
+  "dropout": 0.1,
+  "encoder_attention_heads": 8,
+  "encoder_ffn_dim": 2048,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "encoder_n_points": 4,
+  "eos_coefficient": 0.1,
+  "focal_alpha": 0.25,
+  "giou_cost": 2,
+  "giou_loss_coefficient": 2,
+  "id2label": {
+    "0": "pet_bottle",
+    "1": "hm_ldpe",
+    "2": "pp_w",
+    "3": "ldpe_wrapper",
+    "4": "hdpe_bottle",
+    "5": "paper",
+    "6": "pp",
+    "7": "aluminium_foil",
+    "8": "multilayer_plastic",
+    "9": "ps",
+    "10": "cardboard",
+    "11": "blister_pack",
+    "12": "aluminium_can",
+    "13": "tetrapack",
+    "14": "others"
+  },
+  "init_std": 0.02,
+  "init_xavier_std": 1.0,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "aluminium_can": 12,
+    "aluminium_foil": 7,
+    "blister_pack": 11,
+    "cardboard": 10,
+    "hdpe_bottle": 4,
+    "hm_ldpe": 1,
+    "ldpe_wrapper": 3,
+    "multilayer_plastic": 8,
+    "others": 14,
+    "paper": 5,
+    "pet_bottle": 0,
+    "pp": 6,
+    "pp_w": 2,
+    "ps": 9,
+    "tetrapack": 13
+  },
+  "mask_loss_coefficient": 1,
+  "max_position_embeddings": 2048,
+  "model_type": "deta",
+  "num_feature_levels": 5,
+  "num_queries": 900,
+  "position_embedding_type": "sine",
+  "torch_dtype": "float32",
+  "transformers_version": "4.32.1",
+  "two_stage": true,
+  "two_stage_num_proposals": 300,
+  "with_box_refine": true
+}

preprocessor_config.json

@@ -0,0 +1,24 @@
+{
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "format": "coco_detection",
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "DetaImageProcessor",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 1333,
+    "shortest_edge": 800
+  }
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54dfad0443bb7583382e1c68844ac8c11b2831dfb69a66f06958afa6a64e5115
+size 878823725

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b04caf3b254c2bbb092751e2a7a48e5e3ef3a7fc6025f4c942f02b853fbb7c8
+size 4091