initial commit

app.py

@@ -6,6 +6,7 @@ import pinecone
 from langchain.vectorstores import Pinecone
 import os
 from langchain.embeddings.huggingface import HuggingFaceEmbeddings
+import time
 
 API_URL = "https://api-inference.huggingface.co/models/HuggingFaceH4/zephyr-7b-beta"
 # retrieval = Client("https://ishaan-mital-ncert-helper-vector-db.hf.space/--replicas/149bg26k5/")
@@ -28,7 +29,21 @@ pinecone.init(
 index_name = 'llama-rag'
 index = pinecone.Index(index_name)
 text_field = 'text'  # field in metadata that contains text content
-
+docs = [
+    "this is one document",
+    "and another document"
+]
+
+embeddings = embed_model.embed_documents(docs)
+if index_name not in pinecone.list_indexes():
+    pinecone.create_index(
+        index_name,
+        dimension=len(embeddings[0]),
+        metric='cosine'
+    )
+    # wait for index to finish initialization
+    while not pinecone.describe_index(index_name).status['ready']:
+        time.sleep(1)
 vectorstore = Pinecone(
     index, embed_model.embed_query, text_field
 )