Update app.py

app.py

@@ -6,7 +6,6 @@ from llama_index.embeddings.mixedbreadai import MixedbreadAIEmbedding
 from llama_index.embeddings.huggingface import HuggingFaceEmbedding
 from llama_index.llms.groq import Groq
 from llama_parse import LlamaParse
-import mixedbread_ai
 from mixedbread_ai.core.api_error import ApiError
 
 # API keys
@@ -18,19 +17,9 @@ if not (llama_cloud_key and groq_key and mxbai_key):
 
 # Model names
 llm_model_name = "llama-3.1-70b-versatile"
-embed_model_name = "mxbai-embed-large-v1"  # Mixedbread AI model
+embed_model_name = "mixedbread-ai/mxbai-embed-large-v1"
 fallback_embed_model = "sentence-transformers/all-MiniLM-L6-v2"  # Fallback model
 
-# Configure Mixedbread AI SDK
-mixedbread_config = mixedbread_ai.Configuration(
-    api_key=mxbai_key,
-    retry_on=[503],  # Retry on 503 Service Unavailable
-    max_retries=3,
-    retry_delay=2.0,  # Seconds between retries
-    timeout=30.0,  # Request timeout
-)
-mixedbread_client = mixedbread_ai.Client(configuration=mixedbread_config)
-
 # Initialize the parser
 parser = LlamaParse(api_key=llama_cloud_key, result_type="markdown")
 
@@ -52,17 +41,22 @@ file_extractor = {
 }
 
 # Initialize models with error handling
-def initialize_embed_model():
-    try:
-        return MixedbreadAIEmbedding(
-            api_key=mxbai_key,
-            model_name=embed_model_name,
-            mxbai_client=mixedbread_client,  # Use configured SDK client
-        )
-    except Exception as e:
-        print(f"Failed to initialize Mixedbread AI embedding: {str(e)}")
-        print("Falling back to local HuggingFace embedding model.")
-        return HuggingFaceEmbedding(model_name=fallback_embed_model)
+def initialize_embed_model(max_retries=3, delay=2):
+    for attempt in range(max_retries):
+        try:
+            return MixedbreadAIEmbedding(api_key=mxbai_key, model_name=embed_model_name)
+        except ApiError as e:
+            if attempt == max_retries - 1:
+                print(f"Failed to initialize Mixedbread AI embedding after {max_retries} attempts: {str(e)}")
+                print("Falling back to local HuggingFace embedding model.")
+                return HuggingFaceEmbedding(model_name=fallback_embed_model)
+            time.sleep(delay)
+        except Exception as e:
+            print(f"Unexpected error initializing embedding model: {str(e)}")
+            if attempt == max_retries - 1:
+                print("Falling back to local HuggingFace embedding model.")
+                return HuggingFaceEmbedding(model_name=fallback_embed_model)
+            time.sleep(delay)
 
 try:
     embed_model = initialize_embed_model()
@@ -74,7 +68,7 @@ except Exception as e:
 vector_index = None
 
 # File processing function
-def load_files(file_path: str):
+def load_files(file_path: str, max_retries=3, delay=2):
     global vector_index
     if not file_path:
         return "No file path provided. Please upload a file."
@@ -89,17 +83,22 @@ def load_files(file_path: str):
             file_extractor=file_extractor
         ).load_data()
         
-        try:
-            vector_index = VectorStoreIndex.from_documents(
-                document, 
-                embed_model=embed_model
-            )
-            filename = os.path.basename(file_path)
-            return f"Ready to provide responses based on: {filename}"
-        except ApiError as e:
-            return f"Error processing file with Mixedbread AI API: {str(e)}. Status code: {e.status_code}"
-        except Exception as e:
-            return f"Unexpected error processing file: {str(e)}"
+        # Retry logic for creating vector index
+        for attempt in range(max_retries):
+            try:
+                vector_index = VectorStoreIndex.from_documents(
+                    document, 
+                    embed_model=embed_model
+                )
+                filename = os.path.basename(file_path)
+                return f"Ready to provide responses based on: {filename}"
+            except ApiError as e:
+                if attempt == max_retries - 1:
+                    return f"Error processing file after {max_retries} attempts: {str(e)}"
+                print(f"Attempt {attempt + 1} failed: {str(e)}. Retrying in {delay} seconds...")
+                time.sleep(delay)
+            except Exception as e:
+                return f"Unexpected error processing file: {str(e)}"
     except Exception as e:
         return f"Error loading file: {str(e)}"