exdysa/EVA-CLIP-18B-SAFETENSORS

Original Model Link : https://huggingface.co/BAAI/EVA-CLIP-18B

name: EVA-CLIP-18B-SFT
base_model: BAAI/EVA-CLIP-18B
license: apache-2.0
pipeline_tag: image-to-text
size : 
- 35255802928
- 36GB
- 10GB x 7
- 5GB x 1
tasks :
 - image-to-text
 - image-text-to-text
 - visual-question-answering
 - zero-shot-image-classification
language: en
datasets :
 - laion/laion2B-en
 - kakaobrain/coyo-700m

EVA-CLIP-18B-SFT

EVA-CLIP-18B in Safetensors format with no quantization applied

Usage

Huggingface Version

from PIL import Image
from transformers import AutoModel, AutoConfig
from transformers import CLIPImageProcessor, pipeline, CLIPTokenizer
import torch
import torchvision.transforms as T
from torchvision.transforms import InterpolationMode

image_path = "CLIP.png"
model_name_or_path = "BAAI/EVA-CLIP-18B" # or /path/to/local/EVA-CLIP-18B
image_size = 224

processor = CLIPImageProcessor.from_pretrained("openai/clip-vit-large-patch14")

# use image processor with conig
# processor = CLIPImageProcessor(size={"shortest_edge":image_size}, do_center_crop=True, crop_size=image_size)

## you can also directly use the image processor by torchvision
## squash
# processor = T.Compose(
#     [
#         T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
#         T.Resize((image_size, image_size), interpolation=InterpolationMode.BICUBIC),
#         T.ToTensor(),
#         T.Normalize(mean=(0.48145466, 0.4578275, 0.40821073), std=(0.26862954, 0.26130258, 0.27577711))
#     ]
# )
## shortest
## processor = T.Compose(
#     [
#         T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
#         T.Resize(image_size, interpolation=InterpolationMode.BICUBIC),
#         T.CenterCrop(image_size),
#         T.ToTensor(),
#         T.Normalize(mean=(0.48145466, 0.4578275, 0.40821073), std=(0.26862954, 0.26130258, 0.27577711))
#     ]
# )

model = AutoModel.from_pretrained(
    model_name_or_path, 
    torch_dtype=torch.float16,
    trust_remote_code=True).to('cuda').eval()

image = Image.open(image_path)
captions = ["a diagram", "a dog", "a cat"]
tokenizer = CLIPTokenizer.from_pretrained(model_name_or_path)
input_ids = tokenizer(captions,  return_tensors="pt", padding=True).input_ids.to('cuda')
input_pixels = processor(images=image, return_tensors="pt", padding=True).pixel_values.to('cuda')

with torch.no_grad(), torch.cuda.amp.autocast():
    image_features = model.encode_image(input_pixels)
    text_features = model.encode_text(input_ids)
    image_features /= image_features.norm(dim=-1, keepdim=True)
    text_features /= text_features.norm(dim=-1, keepdim=True)

label_probs = (100.0 * image_features @ text_features.T).softmax(dim=-1)
print(f"Label probs: {label_probs}")

PyTorch Version

GitHub

import torch
from eva_clip import create_model_and_transforms, get_tokenizer
from PIL import Image

model_name = "EVA-CLIP-18B" 
pretrained = "eva_clip" # or "/path/to/EVA_CLIP_18B_psz14_s6B.fp16.pt"

image_path = "CLIP.png"
caption = ["a diagram", "a dog", "a cat"]

device = "cuda" if torch.cuda.is_available() else "cpu"
model, _, processor = create_model_and_transforms(model_name, pretrained, force_custom_clip=True)
tokenizer = get_tokenizer(model_name)
model = model.to(device)

image = processor(Image.open(image_path)).unsqueeze(0).to(device)
text = tokenizer(["a diagram", "a dog", "a cat"]).to(device)

with torch.no_grad(), torch.cuda.amp.autocast():
    image_features = model.encode_image(image)
    text_features = model.encode_text(text)
    image_features /= image_features.norm(dim=-1, keepdim=True)
    text_features /= text_features.norm(dim=-1, keepdim=True)

    text_probs = (100.0 * image_features @ text_features.T).softmax(dim=-1)

print("Label probs:", text_probs)

You can leverage deepspeed.zero.Init() with deepspeed zero stage 3 if you have limited CPU memory. For loading a pretrained checkpoint in the context of using deepspeed.zero.Init(), it's advised to use the load_zero_partitions() function in eva_clip/factory.py.

BibTeX & Citation

@article{EVA-CLIP-18B,
  title={EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters}, 
  author={Quan Sun and Jinsheng Wang and Qiying Yu and Yufeng Cui and Fan Zhang and Xiaosong Zhang and Xinlong Wang},
  journal={arXiv preprint arXiv:2402.04252},
  year={2023}
}