disabled models except blip and blip2_8bit models

app.py

@@ -12,11 +12,11 @@ torch.hub.download_url_to_file('https://cdn.openai.com/dall-e-2/demos/text2im/as
 # git_processor_base = AutoProcessor.from_pretrained("microsoft/git-base-coco")
 # git_model_base = AutoModelForCausalLM.from_pretrained("microsoft/git-base-coco")
 
-git_processor_large_coco = AutoProcessor.from_pretrained("microsoft/git-large-coco")
-git_model_large_coco = AutoModelForCausalLM.from_pretrained("microsoft/git-large-coco")
+# git_processor_large_coco = AutoProcessor.from_pretrained("microsoft/git-large-coco")
+# git_model_large_coco = AutoModelForCausalLM.from_pretrained("microsoft/git-large-coco")
 
-git_processor_large_textcaps = AutoProcessor.from_pretrained("microsoft/git-large-r-textcaps")
-git_model_large_textcaps = AutoModelForCausalLM.from_pretrained("microsoft/git-large-r-textcaps")
+# git_processor_large_textcaps = AutoProcessor.from_pretrained("microsoft/git-large-r-textcaps")
+# git_model_large_textcaps = AutoModelForCausalLM.from_pretrained("microsoft/git-large-r-textcaps")
 
 # blip_processor_base = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
 # blip_model_base = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
@@ -34,20 +34,20 @@ blip2_model_8_bit = Blip2ForConditionalGeneration.from_pretrained("Salesforce/bl
 # vitgpt_model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
 # vitgpt_tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
 
-coca_model, _, coca_transform = open_clip.create_model_and_transforms(
-  model_name="coca_ViT-L-14",
-  pretrained="mscoco_finetuned_laion2B-s13B-b90k"
-)
+# coca_model, _, coca_transform = open_clip.create_model_and_transforms(
+#   model_name="coca_ViT-L-14",
+#   pretrained="mscoco_finetuned_laion2B-s13B-b90k"
+# )
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
 # git_model_base.to(device)
 # blip_model_base.to(device)
-git_model_large_coco.to(device)
-git_model_large_textcaps.to(device)
+# git_model_large_coco.to(device)
+# git_model_large_textcaps.to(device)
 blip_model_large.to(device)
 # vitgpt_model.to(device)
-coca_model.to(device)
+# coca_model.to(device)
 # blip2_model.to(device)
 
 def generate_caption(processor, model, image, tokenizer=None, use_float_16=False):
@@ -76,9 +76,9 @@ def generate_caption_coca(model, transform, image):
 def generate_captions(image):
     # caption_git_base = generate_caption(git_processor_base, git_model_base, image)
 
-    caption_git_large_coco = generate_caption(git_processor_large_coco, git_model_large_coco, image)
+    # caption_git_large_coco = generate_caption(git_processor_large_coco, git_model_large_coco, image)
 
-    caption_git_large_textcaps = generate_caption(git_processor_large_textcaps, git_model_large_textcaps, image)
+    # caption_git_large_textcaps = generate_caption(git_processor_large_textcaps, git_model_large_textcaps, image)
 
     # caption_blip_base = generate_caption(blip_processor_base, blip_model_base, image)
 
@@ -86,17 +86,20 @@ def generate_captions(image):
 
     # caption_vitgpt = generate_caption(vitgpt_processor, vitgpt_model, image, vitgpt_tokenizer)
 
-    caption_coca = generate_caption_coca(coca_model, coca_transform, image)
+    # caption_coca = generate_caption_coca(coca_model, coca_transform, image)
 
     # caption_blip2 = generate_caption(blip2_processor, blip2_model, image, use_float_16=True).strip()
 
     caption_blip2_8_bit = generate_caption(blip2_processor_8_bit, blip2_model_8_bit, image, use_float_16=True).strip()
 
-    return caption_git_large_coco, caption_git_large_textcaps, caption_blip_large, caption_coca, caption_blip2_8_bit
+    # return caption_git_large_coco, caption_git_large_textcaps, caption_blip_large, caption_coca, caption_blip2_8_bit
+    return caption_blip_large, caption_blip2_8_bit
+
 
    
 examples = [["cats.jpg"], ["stop_sign.png"], ["astronaut.jpg"]]
-outputs = [gr.outputs.Textbox(label="Caption generated by GIT-large fine-tuned on COCO"), gr.outputs.Textbox(label="Caption generated by GIT-large fine-tuned on TextCaps"), gr.outputs.Textbox(label="Caption generated by BLIP-large"), gr.outputs.Textbox(label="Caption generated by CoCa"), gr.outputs.Textbox(label="Caption generated by BLIP-2 OPT 6.7b")] 
+# outputs = [gr.outputs.Textbox(label="Caption generated by GIT-large fine-tuned on COCO"), gr.outputs.Textbox(label="Caption generated by GIT-large fine-tuned on TextCaps"), gr.outputs.Textbox(label="Caption generated by BLIP-large"), gr.outputs.Textbox(label="Caption generated by CoCa"), gr.outputs.Textbox(label="Caption generated by BLIP-2 OPT 6.7b")] 
+outputs = [gr.outputs.Textbox(label="Caption generated by BLIP-large"), gr.outputs.Textbox(label="Caption generated by BLIP-2 OPT 6.7b")] 
 
 title = "Interactive demo: comparing image captioning models"
 description = "Gradio Demo to compare GIT, BLIP, CoCa, and BLIP-2, 4 state-of-the-art vision+language models. To use it, simply upload your image and click 'submit', or click one of the examples to load them. Read more at the links below."