changed app.py

app.py

@@ -1,13 +1,26 @@
-# Step 2: Verify GPU
+# Step 1: Clone the DeepSeek-VL2 repository (Only for Colab environments)
+import os
+if not os.path.exists("/content/DeepSeek-VL2"):
+    !git clone https://github.com/deepseek-ai/DeepSeek-VL2
+    %cd DeepSeek-VL2
+else:
+    print("Repository already cloned.")
+
+# Install dependencies if not already installed (you can add more if necessary)
+!pip install -r /content/DeepSeek-VL2/requirements.txt
+
+# Step 2: Ensure DeepSeek-VL2 is in the Python path
+import sys
+sys.path.append('/content/DeepSeek-VL2')  # Add the DeepSeek-VL2 repository to the path
+
+# Step 3: Verify GPU (Optional)
 import torch
 print("CUDA Available:", torch.cuda.is_available())
 print("Device:", torch.cuda.current_device() if torch.cuda.is_available() else "CPU")
 print("Device Name:", torch.cuda.get_device_name(0) if torch.cuda.is_available() else "N/A")
 
-# Step 3: Modified app.py for Colab with debugging
-import os
+# Step 4: Rest of your app.py code
 import gradio as gr
-import torch
 from PIL import Image
 from deepseek_vl2.serve.inference import load_model, deepseek_generate, convert_conversation_to_prompts
 from deepseek_vl2.models.conversation import SeparatorStyle
@@ -125,4 +138,4 @@ with gr.Blocks() as demo:
     submit_btn = gr.Button("Extract Text")
     submit_btn.click(upload_and_process, inputs=image_input, outputs=[output_text, history_state])
 
-demo.launch(share=True, debug=True)  # Added debug=True for more Gradio logs
+demo.launch(share=True, debug=True)  # Added debug=True for more Gradio logs