Add environment variable configuration and directory creation for ML libraries

- Introduced environment variables for TORCH_HOME, TENSORFLOW_HOME, and KERAS_HOME to manage library directories.
- Implemented robust directory creation with error handling and fallback mechanisms to ensure stability.
- Enhanced logging to provide feedback on directory creation processes and any encountered issues.

Dockerfile

@@ -10,7 +10,7 @@ RUN apt-get update && apt-get install -y \
     && rm -rf /var/lib/apt/lists/*
 
 # Create necessary directories with proper permissions
-RUN mkdir -p /app/.streamlit /tmp/docling_temp /tmp/easyocr_models /tmp/cache /tmp/config /tmp/data /tmp/huggingface /tmp/huggingface_cache /tmp/transformers_cache /tmp/datasets_cache && \
+RUN mkdir -p /app/.streamlit /tmp/docling_temp /tmp/easyocr_models /tmp/cache /tmp/config /tmp/data /tmp/huggingface /tmp/huggingface_cache /tmp/transformers_cache /tmp/datasets_cache /tmp/torch /tmp/tensorflow /tmp/keras && \
     chmod 755 /app/.streamlit && \
     chmod 777 /tmp/docling_temp && \
     chmod 777 /tmp/easyocr_models && \
@@ -20,7 +20,10 @@ RUN mkdir -p /app/.streamlit /tmp/docling_temp /tmp/easyocr_models /tmp/cache /t
     chmod 777 /tmp/huggingface && \
     chmod 777 /tmp/huggingface_cache && \
     chmod 777 /tmp/transformers_cache && \
-    chmod 777 /tmp/datasets_cache
+    chmod 777 /tmp/datasets_cache && \
+    chmod 777 /tmp/torch && \
+    chmod 777 /tmp/tensorflow && \
+    chmod 777 /tmp/keras
 
 COPY requirements.txt ./
 COPY src/ ./src/

src/streamlit_app.py

@@ -120,6 +120,29 @@ for env_var in hf_dirs:
         except Exception as e2:
             logging.error(f"Failed to create fallback Hugging Face directory for {env_var}: {e2}")
 
+# Additional environment variables for other libraries that might access root directories
+os.environ['TORCH_HOME'] = os.path.join(TEMP_DIR, 'torch')
+os.environ['TENSORFLOW_HOME'] = os.path.join(TEMP_DIR, 'tensorflow')
+os.environ['KERAS_HOME'] = os.path.join(TEMP_DIR, 'keras')
+os.environ['MLFLOW_TRACKING_URI'] = 'file:' + os.path.join(TEMP_DIR, 'mlruns')
+
+# Create additional library directories
+lib_dirs = ['TORCH_HOME', 'TENSORFLOW_HOME', 'KERAS_HOME']
+for env_var in lib_dirs:
+    try:
+        os.makedirs(os.environ[env_var], exist_ok=True)
+        logging.info(f"Created library directory for {env_var}: {os.environ[env_var]}")
+    except Exception as e:
+        logging.warning(f"Could not create library directory for {env_var}: {e}")
+        # Fallback to /tmp
+        fallback_path = os.path.join('/tmp', env_var.lower())
+        os.environ[env_var] = fallback_path
+        try:
+            os.makedirs(fallback_path, exist_ok=True)
+            logging.info(f"Using fallback library directory for {env_var}: {fallback_path}")
+        except Exception as e2:
+            logging.error(f"Failed to create fallback library directory for {env_var}: {e2}")
+
 # Log startup information
 logging.info("=" * 50)
 logging.info("Docling Streamlit App Starting")