Dependency updates

Dockerfile.gradio.prod

@@ -2,8 +2,10 @@
 # BUILDER #
 ###########
 
+ARG AWS_ACCOUNT_ID
+
 # pull official base image
-FROM 224427659724.dkr.ecr.us-east-1.amazonaws.com/gradio-python:3.11.10-slim as builder
+FROM ${AWS_ACCOUNT_ID}.dkr.ecr.us-east-1.amazonaws.com/gradio-python:3.11.10-slim as builder
 
 # set work directory
 WORKDIR /app
@@ -27,8 +29,10 @@ RUN pip wheel --no-cache-dir --no-deps --wheel-dir /app/wheels -r requirements.t
 # FINAL #
 #########
 
+ARG AWS_ACCOUNT_ID
+
 # pull official base image
-FROM 224427659724.dkr.ecr.us-east-1.amazonaws.com/gradio-python:3.11.10-slim
+FROM ${AWS_ACCOUNT_ID}.dkr.ecr.us-east-1.amazonaws.com/gradio-python:3.11.10-slim
 
 # create directory for the app user
 RUN mkdir -p /home/backend-app

README.md

@@ -121,7 +121,7 @@ export AWS_ACCOUNT_ID=aws_account_id # ie. 2243838xxxxxx
 Login into the AWS ECR (Elastic Container Registry) via the commandline
 
 ```sh
-aws ecr get-login-password --region $AWS_DEFAULT_REGION | docker login --username AWS --password-stdin "$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com
+aws ecr get-login-password --region $AWS_DEFAULT_REGION | docker login --username AWS --password-stdin "$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com"
 ```
 
 Create a python image and push to ECR. This image will be used as the base image for the application image deployed on AWS ECS.
@@ -181,7 +181,7 @@ echo $ECR_BACKEND_FASTAPI_URL
 - Build the docker image for the production and push to ECR
 
 ```sh
-docker build -f Dockerfile.api.prod -t fastapi-api-prod .
+docker build --build-arg AWS_ACCOUNT_ID=$AWS_ACCOUNT_ID -f Dockerfile.api.prod -t fastapi-api-prod .
 docker tag fastapi-api-prod:latest "${ECR_BACKEND_FASTAPI_URL}:latest"
 docker push "${ECR_BACKEND_FASTAPI_URL}:latest"
 ```

bin/cfn/ecs-deploy

@@ -22,4 +22,4 @@ aws cloudformation deploy \
   --tags group=acres-rag \
   --parameter-overrides $PARAMETERS \
   --capabilities CAPABILITY_NAMED_IAM \
-  --profile sunbirdai
+  --profile acres

rag/rag_pipeline.py

@@ -1,10 +1,12 @@
 # rag/rag_pipeline.py
 import json
 import logging
+import os
 import re
 from typing import Any, Dict, List, Optional, Tuple
 
 import chromadb
+from dotenv import load_dotenv
 from llama_index.core import Document, PromptTemplate, VectorStoreIndex
 from llama_index.core.node_parser import SentenceSplitter, SentenceWindowNodeParser
 from llama_index.embeddings.openai import OpenAIEmbedding
@@ -14,6 +16,8 @@ from llama_index.vector_stores.chroma import ChromaVectorStore
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
 
+load_dotenv()
+
 
 class RAGPipeline:
     def __init__(
@@ -28,7 +32,9 @@ class RAGPipeline:
         self.documents = None
         self.client = chromadb.Client()
         self.collection = self.client.get_or_create_collection(self.collection_name)
-        self.embedding_model = OpenAIEmbedding(model_name="text-embedding-ada-002")
+        self.embedding_model = OpenAIEmbedding(
+            model_name="text-embedding-ada-002", api_key=os.getenv("OPENAI_API_KEY")
+        )
         self.is_pdf = self._check_if_pdf_collection()
         self.load_documents()
         self.build_index()
@@ -175,7 +181,7 @@ class RAGPipeline:
             text_qa_template=prompt_template,
             similarity_top_k=n_documents if n_documents <= 17 else 15,
             response_mode="tree_summarize",
-            llm=OpenAI(model="gpt-4o-mini"),
+            llm=OpenAI(model="gpt-4o-mini", api_key=os.getenv("OPENAI_API_KEY")),
         )
 
         response = query_engine.query(context)

requirements.txt

@@ -1,17 +1,17 @@
-chromadb
-fastapi
-gradio
-gradio_client
-llama-index
-llama-index-vector-stores-chroma
+chromadb==0.5.20
+fastapi==0.115.5
+gradio==5.6.0
+gradio_client==1.4.3
+llama-index==0.12.3
+llama-index-vector-stores-chroma==0.4.0
 nest-asyncio==1.6.0
-openai
-pandas
-pydantic
+openai==1.57.0
+pandas==2.2.3
+pydantic==2.9.2
 python-dotenv
 pyzotero
 python-slugify
 PyMuPDF==1.23.8  
 Pillow==10.2.0
-sqlmodel
+sqlmodel==0.0.22
 cachetools