Update app.py

app.py

@@ -6,9 +6,9 @@ import torch
 from PIL import Image
 import gradio as gr
 import subprocess
-from LLaVA.llava.model.builder import load_pretrained_model
-from LLaVA.llava.mm_utils import get_model_name_from_path
-from LLaVA.llava.eval.run_llava import eval_model
+from llava.model.builder import load_pretrained_model
+from llava.mm_utils import get_model_name_from_path
+from llava.eval.run_llava import eval_model
 
 # Load the LLaVA model and processor
 llava_model_path = "/My_new_LLaVA/llava-fine_tune_model"
@@ -32,30 +32,6 @@ class Args:
         self.num_beams = num_beams
         self.max_new_tokens = max_new_tokens
 
-# # Function to predict using Idefics2
-# def predict_idefics2(image, question, temperature, max_tokens):
-#     image = image.convert("RGB")
-#     images = [image]
-
-#     messages = [
-#         {
-#             "role": "user",
-#             "content": [
-#                 {"type": "image"},
-#                 {"type": "text", "text": question}
-#             ]
-#         }
-#     ]
-#     input_text = idefics2_processor.apply_chat_template(messages, add_generation_prompt=False).strip()
-
-#     inputs = idefics2_processor(text=[input_text], images=images, return_tensors="pt", padding=True).to("cuda:0")
-
-#     with torch.no_grad():
-#         outputs = idefics2_model.generate(**inputs, max_length=max_tokens, max_new_tokens=max_tokens, temperature=temperature)
-
-#     predictions = idefics2_processor.decode(outputs[0], skip_special_tokens=True)
-
-#     return predictions
 
 # Function to predict using LLaVA
 def predict_llava(image, question, temperature, max_tokens):