fix build error

Dockerfile

@@ -9,7 +9,7 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
 
 # Install dependencies
 COPY requirements.txt .
-RUN pip install --no-cache-dir -r requirements.txt
+RUN pip cache purge && pip install --no-cache-dir -r requirements.txt
 
 # Copy source code
 COPY app.py .
@@ -20,7 +20,7 @@ RUN mkdir -p /app/cache && chmod -R 777 /app/cache
 # Set env vars
 ENV HF_HOME=/app/cache \
     PYTHONUNBUFFERED=1 \
-    PYTHONWARNINGS=ignore::FutureWarning
+    PYTHONWARNINGS=ignore::FutureWarning,ignore::UserWarning:torch._utils
 
 EXPOSE 7860
 

app.py

@@ -1,4 +1,7 @@
 import os
+import warnings
+warnings.filterwarnings("ignore", category=UserWarning, module="torch._utils")
+
 from transformers import GPT2LMHeadModel, GPT2Tokenizer
 import torch
 import gradio as gr
@@ -17,6 +20,10 @@ device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 model.to(device)
 model.eval()
 
+# Print device and memory info for debugging
+print(f"Device: {device}")
+print(f"Memory allocated: {torch.cuda.memory_allocated(device)/1e9:.2f} GB" if torch.cuda.is_available() else "CPU only")
+
 def generate_text(prompt, max_length=100, temperature=1.0):
     try:
         inputs = tokenizer.encode(prompt, return_tensors="pt").to(device)

requirements.txt

@@ -1,16 +1,6 @@
-# CPU-friendly PyTorch
-torch==2.1.2+cpu
--f https://download.pytorch.org/whl/torch_stable.html
-
-# Transformers / Sentence Transformers
 transformers==4.40.0
-sentence-transformers==2.7.0
-
-# Gradio app
+torch==2.1.2
 gradio==4.44.1
-fastapi
-uvicorn
-
-# Tools
-pandas
-numpy
+fastapi==0.111.0
+pydantic==2.7.1
+numpy<2.0.0