fixed issues

.gitignore

@@ -0,0 +1,3 @@
+env.sh
+
+test_request.py

app/chains.py

@@ -31,12 +31,12 @@ history_chain = None
 # TODO: Let's construct the standalone_chain by piping standalone_prompt_formatted with the LLM
 standalone_chain = None
 
-input_1 = RunnablePassthrough.assign(new_question=standalone_chain)
-input_2 = {
-    'context': lambda x: format_context(data_indexer.search(x['new_question'])),
-    'standalone_question': lambda x: x['new_question']
-}
-input_to_rag_chain = input_1 | input_2
+# input_1 = RunnablePassthrough.assign(new_question=standalone_chain)
+# input_2 = {
+#     'context': lambda x: format_context(data_indexer.search(x['new_question'])),
+#     'standalone_question': lambda x: x['new_question']
+# }
+# input_to_rag_chain = input_1 | input_2
 
 # TODO: use input_to_rag_chain, rag_prompt_formatted, 
 # HistoryInput and the LLM to build the rag_chain.

app/data_indexing.py

@@ -27,11 +27,19 @@ class DataIndexer:
         if index_name not in self.pinecone_client.list_indexes().names():
             # TODO: create your index if it doesn't exist. Use the create_index function. 
             # Make sure to choose the dimension that corresponds to your embedding model
-            pass
+            self.pinecone_client.create_index(
+                name=self.index_name,
+                dimension=1536,
+                metric="cosine",
+                spec=ServerlessSpec(
+                    cloud="aws",
+                    region="us-east-1",
+                ),
+            )
 
         self.index = self.pinecone_client.Index(self.index_name)
         # TODO: make sure to build the index.
-        self.source_index = None
+        self.source_index = self.get_source_index()
 
     def get_source_index(self):
         if not os.path.isfile(self.source_file):