Merge pull request #1 from SunbirdAI/api

.dockerignore

@@ -0,0 +1,13 @@
+**/.git
+**/.gitignore
+**/.vscode
+**/coverage
+**/.aws
+**/.ssh
+**/.terraform
+Dockerfile
+README.md
+docker-compose.yml
+**/.DS_Store
+**/venv
+**/env

.env.example

@@ -0,0 +1,6 @@
+OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxx
+HF_TOKEN_PATRICK=hf_xxxxxxxxxxxxxxxxx
+ZOTERO_LIBRARY_ID=1120xxxx
+ZOTERO_API_ACCESS_KEY=Ky5RGxxxxxxxxxxxxxxxxxx
+GRADIO_URL_=http://gradio:7860/
+

.flake8

@@ -0,0 +1,5 @@
+[flake8]
+    ignore = D203, E402, F403, F405, W503, W605
+    exclude = .git,env,__pycache__,docs/source/conf.py,old,build,dist, *migrations*,env,venv,alembic
+    max-complexity = 10
+    max-line-length = 119

.gitignore

@@ -173,5 +173,9 @@ poetry.toml
 pyrightconfig.json
 
 # data
-# data/
+data/
 study_export_*
+study_files.db
+study_files.json
+
+infra/ecs_config.toml

.isort.cfg

@@ -0,0 +1,7 @@
+[settings]
+multi_line_output=3
+include_trailing_comma=True
+force_grid_wrap=0
+use_parentheses=True
+line_length=88
+skip=env,migrations,alembic,venv

Dockerfile.api

@@ -0,0 +1,19 @@
+# FastAPI Dockerfile
+FROM python:3.11.10-slim
+
+ENV PYTHONUNBUFFERED=1
+ENV OMP_NUM_THREADS=1
+
+# Set working directory
+WORKDIR /app
+
+# Copy app files
+COPY requirements.txt ./
+RUN pip install --no-cache-dir -r requirements.txt
+COPY . .
+
+# Expose port
+EXPOSE 8000
+
+# Command to run the FastAPI app
+CMD ["uvicorn", "api:app", "--host", "0.0.0.0", "--port", "8000"]

Dockerfile.api.prod

@@ -0,0 +1,72 @@
+###########
+# BUILDER #
+###########
+
+ARG AWS_ACCOUNT_ID
+
+# pull official base image
+FROM ${AWS_ACCOUNT_ID}.dkr.ecr.us-east-1.amazonaws.com/gradio-python:3.11.10-slim as builder
+
+# set work directory
+WORKDIR /app
+
+# set environment variables
+ENV PYTHONDONTWRITEBYTECODE 1
+ENV PYTHONUNBUFFERED 1
+ENV OMP_NUM_THREADS=1
+
+
+# install dependencies
+RUN apt-get update \
+    && apt-get -y install libpq-dev gcc \
+    && pip install psycopg
+
+RUN pip install --upgrade pip
+COPY ./requirements.txt /app/requirements.txt
+RUN pip wheel --no-cache-dir --no-deps --wheel-dir /app/wheels -r requirements.txt
+
+#########
+# FINAL #
+#########
+
+ARG AWS_ACCOUNT_ID
+
+# pull official base image
+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
+
+# create the app user
+RUN addgroup --system app && adduser --system --group app
+
+# create the appropriate directories
+ENV HOME=/home/app
+ENV BACKEND_APP_HOME=/home/app
+# RUN mkdir $BACKEND_APP_HOME
+WORKDIR $BACKEND_APP_HOME
+
+# install dependencies
+RUN apt-get update \
+    && apt-get -y install libpq-dev gcc \
+    && pip install psycopg
+
+COPY --from=builder /app/wheels /wheels
+COPY --from=builder /app/requirements.txt .
+RUN pip install --upgrade pip
+RUN pip install --no-cache /wheels/*
+
+# copy project
+COPY . $BACKEND_APP_HOME
+
+# chown all the files to the app user
+RUN chown -R app:app $BACKEND_APP_HOME
+
+# change to the app user
+USER app
+
+## Expose port
+EXPOSE 8000
+
+# Command to run the FastAPI app
+CMD ["uvicorn", "api:app", "--host", "0.0.0.0", "--port", "8000"]

Dockerfile.gradio

@@ -0,0 +1,21 @@
+# Gradio Dockerfile
+FROM python:3.11.10-slim
+
+ENV PYTHONUNBUFFERED=1
+ENV OMP_NUM_THREADS=1
+
+# Set working directory
+WORKDIR /app
+
+# Copy app files
+COPY requirements.txt ./
+RUN pip install --no-cache-dir -r requirements.txt
+COPY . .
+
+# Expose port
+EXPOSE 7860
+ENV GRADIO_SERVER_NAME="0.0.0.0"
+
+# Command to run the Gradio app
+CMD ["gradio", "app.py"]
+# CMD ["python", "app.py"]

Dockerfile.gradio.prod

@@ -0,0 +1,68 @@
+###########
+# BUILDER #
+###########
+
+# pull official base image
+FROM 224427659724.dkr.ecr.us-east-1.amazonaws.com/gradio-python:3.11.10-slim as builder
+
+# set work directory
+WORKDIR /app
+
+# set environment variables
+ENV PYTHONDONTWRITEBYTECODE 1
+ENV PYTHONUNBUFFERED 1
+ENV OMP_NUM_THREADS=1
+
+
+# install dependencies
+RUN apt-get update \
+    && apt-get -y install libpq-dev gcc \
+    && pip install psycopg
+
+RUN pip install --upgrade pip
+COPY ./requirements.txt /app/requirements.txt
+RUN pip wheel --no-cache-dir --no-deps --wheel-dir /app/wheels -r requirements.txt
+
+#########
+# FINAL #
+#########
+
+# pull official base image
+FROM 224427659724.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
+
+# create the app user
+RUN addgroup --system app && adduser --system --group app
+
+# create the appropriate directories
+ENV HOME=/home/app
+ENV BACKEND_APP_HOME=/home/app
+# RUN mkdir $BACKEND_APP_HOME
+WORKDIR $BACKEND_APP_HOME
+
+# install dependencies
+RUN apt-get update \
+    && apt-get -y install libpq-dev gcc \
+    && pip install psycopg
+
+COPY --from=builder /app/wheels /wheels
+COPY --from=builder /app/requirements.txt .
+RUN pip install --upgrade pip
+RUN pip install --no-cache /wheels/*
+
+# copy project
+COPY . $BACKEND_APP_HOME
+
+# chown all the files to the app user
+RUN chown -R app:app $BACKEND_APP_HOME
+
+# change to the app user
+USER app
+
+# Expose port
+EXPOSE 7860
+ENV GRADIO_SERVER_NAME="0.0.0.0"
+
+CMD ["gradio", "app.py"]

Makefile

@@ -0,0 +1,12 @@
+.PHONY: lint-apply lint-check
+
+lint-check:
+	@echo "Checking for lint errors..."
+	flake8 .
+	black --check .
+	isort --check-only .
+
+lint-apply:
+	@echo "apply linting ..."
+	black .
+	isort .

README.md

@@ -10,4 +10,177 @@ pinned: false
 license: apache-2.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+
+# ACRES RAG Project
+
+## Project Setup
+
+To test and run the project locally. Clone the project from github and change directoory to `acres`.
+
+```sh
+git clone https://github.com/SunbirdAI/acres.git
+cd acres
+```
+
+Create python virtual environment and activate it.
+
+```sh
+python -m venv env
+source env/bin/activate
+```
+
+Install project dependencies
+
+```sh
+pip install -r requirements.txt
+```
+
+## Run project locally
+To test the project locally follow the steps below.
+
+Copy `.env.example` to `.env` and provide the correct enviroment variable values.
+
+```sh
+cp .env.example .env
+```
+
+Run the application
+
+```sh
+python app.py
+```
+
+OR
+
+```sh
+gradio app.py
+```
+
+Browse the application with the link `http://localhost:7860/`
+
+
+## Run with docker
+To run the application with docker locally, first make sure you have docker installed. See [link](https://docs.docker.com/)
+
+Build the project docker image
+
+```sh
+docker build -f Dockerfile.gradio -t gradio-app .
+```
+
+Create docker network
+
+```sh
+docker network create gradio-fastapi-network
+```
+
+Run the docker container 
+
+```sh
+docker run -it -p 7860:7860 --rm --name gradio --network=gradio-fastapi-network gradio-app
+```
+
+Browse the application with the link `http://localhost:7860/`
+
+
+## Deploy to AWS ECS (Elastic Container Service) with Fargate
+
+Install and configure the AWS CLI and aws credentials. See [link](https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-welcome.html)
+
+OR: See the pdf document [here](./aws-cli.pdf)
+
+Now follow the steps below to deploy to AWS ECS
+
+Setup the default region and your aws account id
+
+```sh
+export AWS_DEFAULT_REGION=region # i.e us-east-1, eu-west-1
+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.$AWS_DEFAULT_REGION.amazonaws.com
+```
+
+Create a python image and push to ECR. This image will be used as the base image for the application image deployed on AWS ECS.
+
+- Create python repository
+
+```sh
+aws ecr create-repository \
+  --repository-name gradio-python \
+  --image-tag-mutability MUTABLE
+```
+
+```sh
+export ECR_PYTHON_URL="$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com/gradio-python"
+echo $ECR_PYTHON_URL
+```
+
+- Pull python image and tag it to the ECR url
+
+```sh
+docker pull python:3.11.10-slim
+docker tag python:3.11.10-slim $ECR_PYTHON_URL:3.11.10-slim
+
+docker push $ECR_PYTHON_URL:3.11.10-slim
+```
+
+- Now create application repostory
+
+```sh
+aws ecr create-repository \
+  --repository-name gradio-app-prod \
+  --image-tag-mutability MUTABLE
+
+export ECR_BACKEND_GRADIO_URL="$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com/gradio-app-prod"
+echo $ECR_BACKEND_GRADIO_URL
+```
+
+- Build the docker image for the production and push to ECR
+
+```sh
+docker build --build-arg AWS_ACCOUNT_ID=$AWS_ACCOUNT_ID -f Dockerfile.gradio.prod -t gradio-app-prod .
+docker tag gradio-app-prod:latest "${ECR_BACKEND_GRADIO_URL}:latest"
+docker push "${ECR_BACKEND_GRADIO_URL}:latest"
+```
+
+### Setup and Provision AWS ECS infra using AWS Cloudformation (IaC)
+
+#### Install
+To install the CFN-CLI run the command below
+
+```sh
+pip install cloudformation-cli cloudformation-cli-java-plugin cloudformation-cli-go-plugin cloudformation-cli-python-plugin cloudformation-cli-typescript-plugin
+```
+
+#### CFN-Toml
+
+```sh
+gem install cfn-toml
+```
+
+
+Copy `infra/ecs_config.template` to `infra/ecs_config.toml` and provide the correct `AWS Account ID` for the `ContainerImageGradio`
+
+```sh
+cp infra/ecs_config.template infra/ecs_config.toml
+```
+
+
+#### Deploy
+
+To deploy the ECS infra run the commands below. It provisions the cloudformation stack changeset for review. 
+
+Log into your aws console and search for `cloudformation`. See and review the changeset. If everything is good execute the changeset to finish with the infra deployment.
+
+Then look for the outputs to the link for the deployed application.
+
+```sh
+chmod u+x bin/cfn/*
+./bin/cfn/ecs-deploy
+```

api.py

@@ -0,0 +1,116 @@
+import logging
+import os
+from enum import Enum
+from typing import List, Optional
+
+from dotenv import load_dotenv
+from fastapi import FastAPI, HTTPException
+from fastapi.responses import FileResponse
+from gradio_client import Client
+from pydantic import BaseModel, ConfigDict, Field, constr
+
+from docs import description, tags_metadata
+
+load_dotenv()
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+app = FastAPI(
+    title="ACRES RAG API",
+    description=description,
+    openapi_tags=tags_metadata,
+)
+GRADIO_URL = os.getenv("GRADIO_URL", "http://localhost:7860/")
+logger.info(f"GRADIO_URL: {GRADIO_URL}")
+client = Client(GRADIO_URL)
+
+
+class StudyVariables(str, Enum):
+    ebola_virus = "Ebola Virus"
+    vaccine_coverage = "Vaccine coverage"
+    genexpert = "GeneXpert"
+
+
+class PromptType(str, Enum):
+    default = "Default"
+    highlight = "Highlight"
+    evidence_based = "Evidence-based"
+
+
+class StudyVariableRequest(BaseModel):
+    study_variable: StudyVariables
+    prompt_type: PromptType
+    text: constr(min_length=1, strip_whitespace=True)  # type: ignore
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+class DownloadCSV(BaseModel):
+    text: constr(min_length=1, strip_whitespace=True)  # type: ignore
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+class Study(BaseModel):
+    study_name: constr(min_length=1, strip_whitespace=True)  # type: ignore
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+class ZoteroCredentials(BaseModel):
+    library_id: constr(min_length=1, strip_whitespace=True)  # type: ignore
+    api_access_key: constr(min_length=1, strip_whitespace=True)  # type: ignore
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+@app.post("/process_zotero_library_items", tags=["zotero"])
+def process_zotero_library_items(zotero_credentials: ZoteroCredentials):
+    result = client.predict(
+        zotero_library_id=zotero_credentials.library_id,
+        zotero_api_access_key=zotero_credentials.api_access_key,
+        api_name="/process_zotero_library_items",
+    )
+    return {"result": result}
+
+
+@app.post("/get_study_info", tags=["zotero"])
+def get_study_info(study: Study):
+    result = client.predict(study_name=study.study_name, api_name="/get_study_info")
+    # print(result)
+    return {"result": result}
+
+
+@app.post("/study_variables", tags=["zotero"])
+def process_study_variables(
+    study_request: StudyVariableRequest,
+):
+    result = client.predict(
+        text=study_request.text,  # "study id, study title, study design, study summary",
+        study_name=study_request.study_variable,  # "Ebola Virus",
+        prompt_type=study_request.prompt_type,  # "Default",
+        api_name="/process_multi_input",
+    )
+    print(type(result))
+    return {"result": result[0]}
+
+
+@app.post("/download_csv", tags=["zotero"])
+def download_csv(download_request: DownloadCSV):
+    result = client.predict(
+        markdown_content=download_request.text, api_name="/download_as_csv"
+    )
+    print(result)
+
+    file_path = result
+    if not file_path or not os.path.exists(file_path):
+        raise HTTPException(status_code=404, detail="File not found")
+
+    # Use FileResponse to send the file to the client
+    return FileResponse(
+        file_path,
+        media_type="text/csv",  # Specify the correct MIME type for CSV
+        filename=os.path.basename(
+            file_path
+        ),  # Provide a default filename for the download
+    )

app.py

@@ -1,33 +1,38 @@
 # app.py
 
 import csv
-
 import datetime
-
 # from datetime import datetime
 import io
 import json
 import logging
 import os
-from typing import Tuple, List, Any
+from typing import Any, List, Tuple
 
 import gradio as gr
 import openai
+from cachetools import LRUCache
 from dotenv import load_dotenv
 from slugify import slugify
 
-from config import STUDY_FILES, OPENAI_API_KEY
+from config import OPENAI_API_KEY, STUDY_FILES
+from interface import create_chat_interface
 from rag.rag_pipeline import RAGPipeline
+from utils.db import (
+    add_study_files_to_db,
+    create_db_and_tables,
+    get_all_study_files,
+    get_study_file_by_name,
+    get_study_files_by_library_id,
+)
 from utils.helpers import (
-    append_to_study_files,
     add_study_files_to_chromadb,
+    append_to_study_files,
     chromadb_client,
 )
-from utils.prompts import highlight_prompt, evidence_based_prompt
-from utils.zotero_manager import ZoteroManager
-
-from interface import create_chat_interface
 from utils.pdf_processor import PDFProcessor
+from utils.prompts import evidence_based_prompt, highlight_prompt
+from utils.zotero_manager import ZoteroManager
 
 # Configure logging
 logging.basicConfig(level=logging.INFO)
@@ -39,9 +44,27 @@ openai.api_key = OPENAI_API_KEY
 # Initialize ChromaDB with study files
 add_study_files_to_chromadb("study_files.json", "study_files_collection")
 
+# Create sqlite study file data table
+create_db_and_tables()
+
+
 # Cache for RAG pipelines
 rag_cache = {}
 
+cache = LRUCache(maxsize=100)
+
+# with open("study_files.json", "w") as file:
+#     data_ = {}
+#     json.dump(data_, file, indent=4)
+
+
+def get_cache_value(key):
+    return cache.get(key)
+
+
+zotero_library_id = get_cache_value("zotero_library_id")
+logger.info(f"zotero_library_id: {zotero_library_id}")
+
 
 def get_rag_pipeline(study_name: str) -> RAGPipeline:
     """Get or create a RAGPipeline instance for the given study by querying ChromaDB."""
@@ -61,18 +84,26 @@ def get_rag_pipeline(study_name: str) -> RAGPipeline:
     return rag_cache[study_name]
 
 
-def get_study_info(study_name: str) -> str:
+def get_study_info(study_name: str | list) -> str:
     """Retrieve information about the specified study."""
+    if isinstance(study_name, list):
+        study_name = study_name[0] if study_name else None
+
+    if not study_name:
+        return "No study selected"
+
+    study = get_study_file_by_name(study_name)
+    logger.info(f"Study: {study}")
 
     collection = chromadb_client.get_or_create_collection("study_files_collection")
     result = collection.get(ids=[study_name])  # Query by study name (as a list)
-    logging.info(f"Result: ======> {result}")
+    logger.info(f"Result: {result}")
 
     if not result or len(result["metadatas"]) == 0:
         raise ValueError(f"Invalid study name: {study_name}")
 
     study_file = result["metadatas"][0].get("file_path")
-    logging.info(f"study_file: =======> {study_file}")
+    logger.info(f"study_file: {study_file}")
     if not study_file:
         raise ValueError(f"File path not found for study name: {study_name}")
 
@@ -116,9 +147,9 @@ def cleanup_temp_files():
                     try:
                         os.remove(file)
                     except Exception as e:
-                        logging.warning(f"Failed to remove temp file {file}: {e}")
+                        logger.warning(f"Failed to remove temp file {file}: {e}")
     except Exception as e:
-        logging.warning(f"Error during cleanup: {e}")
+        logger.warning(f"Error during cleanup: {e}")
 
 
 def chat_function(message: str, study_name: str, prompt_type: str) -> str:
@@ -128,7 +159,7 @@ def chat_function(message: str, study_name: str, prompt_type: str) -> str:
         return "Please enter a valid query."
 
     rag = get_rag_pipeline(study_name)
-    logging.info(f"rag: ==> {rag}")
+    logger.info(f"rag: {rag}")
     prompt = {
         "Highlight": highlight_prompt,
         "Evidence-based": evidence_based_prompt,
@@ -139,12 +170,14 @@ def chat_function(message: str, study_name: str, prompt_type: str) -> str:
 
 
 def process_zotero_library_items(
-    zotero_library_id: str, zotero_api_access_key: str
+    zotero_library_id_param: str, zotero_api_access_key: str
 ) -> str:
-    if not zotero_library_id or not zotero_api_access_key:
+    global zotero_library_id
+    if not zotero_library_id_param or not zotero_api_access_key:
         return "Please enter your zotero library Id and API Access Key"
 
-    zotero_library_id = zotero_library_id
+    zotero_library_id = zotero_library_id_param
+    cache["zotero_library_id"] = zotero_library_id
     zotero_library_type = "user"  # or "group"
     zotero_api_access_key = zotero_api_access_key
 
@@ -192,10 +225,18 @@ def process_zotero_library_items(
 
                 # Update in-memory STUDY_FILES for reference in current session
                 STUDY_FILES.update({collection_name: f"data/{export_file}"})
-                logging.info(f"STUDY_FILES: {STUDY_FILES}")
+                logger.info(f"STUDY_FILES: {STUDY_FILES}")
 
         # After loop, add all collected data to ChromaDB
         add_study_files_to_chromadb("study_files.json", "study_files_collection")
+        # Add collected data to sqlite
+        add_study_files_to_db("study_files.json", zotero_library_id)
+
+        # Dynamically update study choices
+        global study_choices
+        study_choices = [
+            file.name for file in get_study_files_by_library_id([zotero_library_id])
+        ]
         message = "Successfully processed items in your zotero library"
     except Exception as e:
         message = f"Error process your zotero library: {str(e)}"
@@ -203,11 +244,27 @@ def process_zotero_library_items(
     return message
 
 
+def refresh_study_choices():
+    """
+    Refresh study choices for a specific dropdown instance.
+
+    :return: Updated Dropdown with current study choices
+    """
+    global study_choices
+    zotero_library_id = get_cache_value("zotero_library_id")
+    logger.info(f"zotero_library_id: {zotero_library_id}")
+    study_choices = [
+        file.name for file in get_study_files_by_library_id([zotero_library_id])
+    ]
+    logger.info(f"Study choices: {study_choices}")
+    return study_choices
+
+
 def process_multi_input(text, study_name, prompt_type):
     # Split input based on commas and strip any extra spaces
     variable_list = [word.strip().upper() for word in text.split(",")]
     user_message = f"Extract and present in a tabular format the following variables for each {study_name} study: {', '.join(variable_list)}"
-    logging.info(f"User message: ==> {user_message}")
+    logger.info(f"User message: {user_message}")
     response = chat_function(user_message, study_name, prompt_type)
     return [response, gr.update(visible=True)]
 
@@ -311,7 +368,8 @@ def chat_response(
 
 def create_gr_interface() -> gr.Blocks:
     """Create and configure the Gradio interface for the RAG platform."""
-    with gr.Blocks() as demo:
+    global zotero_library_id
+    with gr.Blocks(theme=gr.themes.Base()) as demo:
         gr.Markdown("# ACRES RAG Platform")
 
         with gr.Tabs() as tabs:
@@ -320,7 +378,7 @@ def create_gr_interface() -> gr.Blocks:
                 with gr.Row():
                     with gr.Column(scale=1):
                         gr.Markdown("### Zotero Credentials")
-                        zotero_library_id = gr.Textbox(
+                        zotero_library_id_param = gr.Textbox(
                             label="Zotero Library ID",
                             type="password",
                             placeholder="Enter Your Zotero Library ID here...",
@@ -346,11 +404,26 @@ def create_gr_interface() -> gr.Blocks:
                             if all_documents
                         ]
 
+                        print(f"zotero_library_id: {zotero_library_id_param.value}")
+                        zotero_library_id = zotero_library_id_param.value
+                        if zotero_library_id is None:
+                            zotero_library_id = get_cache_value("zotero_library_id")
+                        logger.info(f"zotero_library_id: =====> {zotero_library_id}")
+                        study_choices_db = get_study_files_by_library_id(
+                            [zotero_library_id]
+                        )
+                        logger.info(f"study_choices_db: =====> {study_choices_db}")
+                        study_files = get_all_study_files()
+                        logger.info(f"study_files: =====> {study_files}")
+
                         study_dropdown = gr.Dropdown(
                             choices=study_choices,
                             label="Select Study",
                             value=(study_choices[0] if study_choices else None),
                         )
+                        # In Gradio interface setup
+                        refresh_button = gr.Button("Refresh Studies")
+
                         study_info = gr.Markdown(label="Study Details")
                         prompt_type = gr.Radio(
                             ["Default", "Highlight", "Evidence-based"],
@@ -420,7 +493,7 @@ def create_gr_interface() -> gr.Blocks:
         # Event handlers for Study Analysis tab
         process_zotero_btn.click(
             process_zotero_library_items,
-            inputs=[zotero_library_id, zotero_api_access_key],
+            inputs=[zotero_library_id_param, zotero_api_access_key],
             outputs=[zotero_output],
         )
 
@@ -438,6 +511,11 @@ def create_gr_interface() -> gr.Blocks:
             fn=download_as_csv, inputs=[answer_output], outputs=[download_btn]
         ).then(fn=cleanup_temp_files, inputs=None, outputs=None)
 
+        refresh_button.click(
+            fn=refresh_study_choices,
+            outputs=[study_dropdown],  # Update the same dropdown
+        )
+
         # Event handlers for PDF Chat tab
 
         def handle_pdf_upload(files, name):

bin/cfn/ecs-delete

@@ -0,0 +1,14 @@
+#! /usr/bin/env bash
+set -e # stop the execution of the script if it fails
+
+CONFIG_PATH="/Users/patrickcmd/Projects/sunbirdai/Acres/infra/ecs_config.toml"
+
+
+REGION=$(cfn-toml key deploy.region -t $CONFIG_PATH)
+STACK_NAME=$(cfn-toml key deploy.stack_name -t $CONFIG_PATH)
+
+
+aws cloudformation delete-stack \
+    --stack-name $STACK_NAME \
+    --region $REGION \
+    --profile sunbirdai

bin/cfn/ecs-deploy

@@ -0,0 +1,25 @@
+#! /usr/bin/env bash
+set -e # stop the execution of the script if it fails
+
+CFN_PATH="/Users/patrickcmd/Projects/sunbirdai/Acres/infra/ecs_fargate.yml"
+CONFIG_PATH="/Users/patrickcmd/Projects/sunbirdai/Acres/infra/ecs_config.toml"
+echo $CFN_PATH
+
+cfn-lint $CFN_PATH
+
+BUCKET=$(cfn-toml key deploy.bucket -t $CONFIG_PATH)
+REGION=$(cfn-toml key deploy.region -t $CONFIG_PATH)
+STACK_NAME=$(cfn-toml key deploy.stack_name -t $CONFIG_PATH)
+PARAMETERS=$(cfn-toml params v2 -t $CONFIG_PATH)
+
+aws cloudformation deploy \
+  --stack-name $STACK_NAME \
+  --s3-bucket $BUCKET \
+  --s3-prefix acres-rag \
+  --region $REGION \
+  --template-file "$CFN_PATH" \
+  --no-execute-changeset \
+  --tags group=acres-rag \
+  --parameter-overrides $PARAMETERS \
+  --capabilities CAPABILITY_NAMED_IAM \
+  --profile sunbirdai

commands.md

@@ -0,0 +1,53 @@
+docker network create gradio-fastapi-network
+
+docker run -it -p 7860:7860 --rm --name gradio --network=gradio-fastapi-network gradio-app
+
+docker run -it -p 7860:7860 --rm --name gradio --network=gradio-fastapi-network gradio-app-prod
+
+
+export AWS_DEFAULT_REGION=us-east-1
+export AWS_ACCOUNT_ID=2244276xxxxx
+aws ecr get-login-password --region $AWS_DEFAULT_REGION | docker login --username AWS --password-stdin "$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com"
+
+aws ecr create-repository \
+  --repository-name gradio-python \
+  --image-tag-mutability MUTABLE
+
+export ECR_PYTHON_URL="$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com/gradio-python"
+echo $ECR_PYTHON_URL
+
+docker pull python:3.11.10-slim
+docker tag python:3.11.10-slim $ECR_PYTHON_URL:3.11.10-slim
+
+docker push $ECR_PYTHON_URL:3.11.10-slim
+
+
+aws ecr create-repository \
+  --repository-name gradio-app-prod \
+  --image-tag-mutability MUTABLE
+
+export ECR_BACKEND_GRADIO_URL="$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com/gradio-app-prod"
+echo $ECR_BACKEND_GRADIO_URL
+
+docker build --build-arg AWS_ACCOUNT_ID=2244276xxxxx -t your-image-name .
+docker build -f Dockerfile.gradio.prod -t gradio-app-prod .
+
+docker build --build-arg AWS_ACCOUNT_ID=$AWS_ACCOUNT_ID -f Dockerfile.gradio.prod -t gradio-app-prod .
+docker tag gradio-app-prod:latest "${ECR_BACKEND_GRADIO_URL}:latest"
+docker push "${ECR_BACKEND_GRADIO_URL}:latest"
+
+
+docker build -f Dockerfile.api -t fastapi-app .
+docker run -it -p 8000:8000 --rm --name fastapi --network=gradio-fastapi-network fastapi-app
+
+aws ecr create-repository \
+  --repository-name fastapi-api-prod \
+  --image-tag-mutability MUTABLE
+
+export ECR_BACKEND_FASTAPI_URL="$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com/fastapi-api-prod"
+echo $ECR_BACKEND_FASTAPI_URL
+
+docker build -f Dockerfile.api.prod -t fastapi-api-prod .
+docker tag fastapi-api-prod:latest "${ECR_BACKEND_FASTAPI_URL}:latest"
+docker push "${ECR_BACKEND_FASTAPI_URL}:latest"
+

docs.py

@@ -0,0 +1,13 @@
+description = """
+Welcome to the Acres AI RAG API documentation.
+
+### RAG Tasks
+- Use the `/process_zotero_library_items`: Process zotero library items with your zotero credentials.
+- Use the `/get_study_info`: Get number of documents in a zotero study.
+- Use the `/study_variables`: Get research summary from the study provided the study variables.
+- Use the `/download_csv`: Export the markdown text to a csv file.
+"""
+
+tags_metadata = [
+    {"name": "ACRES RAG", "description": "AI RAG Application"},
+]

infra/ecs_config.template

@@ -0,0 +1,7 @@
+[deploy]
+bucket = 'dev-acres-gradio-bucket'
+region = 'us-east-1'
+stack_name = 'AcresRag'
+
+[parameters]
+ContainerImageGradio = '224427659xxxx.dkr.ecr.us-east-1.amazonaws.com/gradio-app-prod:latest'

infra/ecs_fargate.yml

@@ -0,0 +1,581 @@
+AWSTemplateFormatVersion: '2010-09-09'
+Description: Deploy Gradio and FastAPI services on AWS ECS Fargate
+
+Parameters:
+  Environment:
+    Type: String
+    Default: dev
+    AllowedValues: [dev, prod]
+  
+  # VPC Configuration
+  VpcCIDR:
+    Type: String
+    Default: 10.0.0.0/16
+  PublicSubnet1CIDR:
+    Type: String
+    Default: 10.0.1.0/24
+  PublicSubnet2CIDR:
+    Type: String
+    Default: 10.0.2.0/24
+  
+  # ECS Configuration
+  ECSClusterName:
+    Type: String
+    Default: rag-ecs-cluster
+  GradioTaskDefinitionCPU:
+    Type: Number
+    Default: 512
+  GradioTaskDefinitionMemory:
+    Type: Number 
+    Default: 1024
+  FastAPITaskDefinitionCPU:
+    Type: Number
+    Default: 256
+  FastAPITaskDefinitionMemory: 
+    Type: Number
+    Default: 512
+  
+  # Container Images
+  ContainerImageGradio:
+    Type: String
+    Description: URI of the Gradio container image in ECR
+  ContainerImageFastAPI:
+    Type: String
+    Description: URI of the FastAPI container image in ECR
+  # CertificateArn:
+  #   Type: String
+
+Resources:
+  # VPC and Networking
+  VPC:
+    Type: AWS::EC2::VPC
+    Properties:
+      CidrBlock: !Ref VpcCIDR
+      EnableDnsHostnames: true
+      EnableDnsSupport: true
+      Tags:
+        - Key: Name
+          Value: !Sub ${Environment}-acres-vpc
+
+  InternetGateway:
+    Type: AWS::EC2::InternetGateway
+    Properties:
+      Tags:
+        - Key: Name
+          Value: !Sub ${Environment}-acres-igw
+
+  AttachGateway:
+    Type: AWS::EC2::VPCGatewayAttachment
+    Properties:
+      VpcId: !Ref VPC
+      InternetGatewayId: !Ref InternetGateway
+
+  PublicSubnet1:
+    Type: AWS::EC2::Subnet
+    Properties:
+      VpcId: !Ref VPC
+      AvailabilityZone: !Select [0, !GetAZs '']
+      CidrBlock: !Ref PublicSubnet1CIDR
+      MapPublicIpOnLaunch: true
+      Tags:
+        - Key: Name
+          Value: !Sub ${Environment}-acres-public-subnet-1
+
+  PublicSubnet2:
+    Type: AWS::EC2::Subnet
+    Properties:
+      VpcId: !Ref VPC
+      AvailabilityZone: !Select [1, !GetAZs '']
+      CidrBlock: !Ref PublicSubnet2CIDR
+      MapPublicIpOnLaunch: true
+      Tags:
+        - Key: Name
+          Value: !Sub ${Environment}-acres-public-subnet-2
+
+  PublicRouteTable:
+    Type: AWS::EC2::RouteTable
+    Properties:
+      VpcId: !Ref VPC
+      Tags:
+        - Key: Name
+          Value: !Sub ${Environment}-acres-public-rt
+
+  PublicRoute:
+    Type: AWS::EC2::Route
+    DependsOn: AttachGateway
+    Properties:
+      RouteTableId: !Ref PublicRouteTable
+      DestinationCidrBlock: 0.0.0.0/0
+      GatewayId: !Ref InternetGateway
+
+  PublicSubnet1RouteTableAssociation:
+    Type: AWS::EC2::SubnetRouteTableAssociation
+    Properties:
+      SubnetId: !Ref PublicSubnet1
+      RouteTableId: !Ref PublicRouteTable
+
+  PublicSubnet2RouteTableAssociation:
+    Type: AWS::EC2::SubnetRouteTableAssociation
+    Properties:
+      SubnetId: !Ref PublicSubnet2
+      RouteTableId: !Ref PublicRouteTable
+
+  # Security Groups
+  GradioSecurityGroup:
+    Type: AWS::EC2::SecurityGroup
+    Properties:
+      GroupDescription: Security group for Gradio service
+      VpcId: !Ref VPC
+      SecurityGroupIngress:
+        - IpProtocol: tcp
+          FromPort: 7860
+          ToPort: 7860
+          CidrIp: 0.0.0.0/0
+          Description: INTERNET HTTPS
+        - IpProtocol: tcp
+          FromPort: 80
+          ToPort: 80
+          CidrIp: 0.0.0.0/0
+          Description: INTERNET HTTP
+      SecurityGroupEgress:
+        - IpProtocol: -1
+          CidrIp: 0.0.0.0/0
+  
+  FastAPISecurityGroup:
+    Type: AWS::EC2::SecurityGroup
+    Properties:
+      GroupDescription: Security group for FastAPI service
+      VpcId: !Ref VPC
+      SecurityGroupIngress:
+        - IpProtocol: tcp
+          FromPort: 8000
+          ToPort: 8000
+          CidrIp: 0.0.0.0/0
+          Description: INTERNET HTTPS
+        - IpProtocol: tcp
+          FromPort: 80
+          ToPort: 80
+          CidrIp: 0.0.0.0/0
+          Description: INTERNET HTTP
+      SecurityGroupEgress:
+        - IpProtocol: -1
+          CidrIp: 0.0.0.0/0
+
+
+  # IAM Roles and Policies
+  # Gradio Execution Role - for pulling images and logging
+  GradioTaskExecutionRole:
+    Type: AWS::IAM::Role
+    Properties:
+      AssumeRolePolicyDocument:
+        Version: '2012-10-17'
+        Statement:
+          - Effect: Allow
+            Principal:
+              Service: ecs-tasks.amazonaws.com
+            Action: sts:AssumeRole
+      ManagedPolicyArns:
+        - arn:aws:iam::aws:policy/service-role/AmazonECSTaskExecutionRolePolicy
+      Policies:
+        - PolicyName: GradioExecutionPolicy
+          PolicyDocument:
+            Version: '2012-10-17'
+            Statement:
+              - Effect: Allow
+                Action:
+                  - ecr:GetAuthorizationToken
+                  - ecr:BatchCheckLayerAvailability
+                  - ecr:GetDownloadUrlForLayer
+                  - ecr:BatchGetImage
+                Resource: '*'
+              - Effect: Allow
+                Action:
+                  - logs:CreateLogStream
+                  - logs:PutLogEvents
+                Resource: 
+                  - !Sub arn:aws:logs:${AWS::Region}:${AWS::AccountId}:log-group:/ecs/${Environment}-acres-gradio:*
+                  - !Sub arn:aws:logs:${AWS::Region}:${AWS::AccountId}:log-group:/ecs/${Environment}-acres-gradio:log-stream:*
+
+  # Gradio Task Role - for runtime permissions
+  GradioTaskRole:
+    Type: AWS::IAM::Role
+    Properties:
+      AssumeRolePolicyDocument:
+        Version: '2012-10-17'
+        Statement:
+          - Effect: Allow
+            Principal:
+              Service: ecs-tasks.amazonaws.com
+            Action: sts:AssumeRole
+      Policies:
+        - PolicyName: GradioTaskPolicy
+          PolicyDocument:
+            Version: '2012-10-17'
+            Statement:
+              # Add specific permissions needed by your Gradio application at runtime
+              - Effect: Allow
+                Action:
+                  - s3:GetObject
+                  - s3:PutObject
+                Resource: !Sub arn:aws:s3:::${Environment}-acres-gradio-bucket/*
+  
+  # FastAPI Execution Role - for pulling images and logging
+  FastAPITaskExecutionRole:
+    Type: AWS::IAM::Role
+    Properties:
+      AssumeRolePolicyDocument:
+        Version: '2012-10-17'
+        Statement:
+          - Effect: Allow
+            Principal:
+              Service: ecs-tasks.amazonaws.com
+            Action: sts:AssumeRole
+      ManagedPolicyArns:
+        - arn:aws:iam::aws:policy/service-role/AmazonECSTaskExecutionRolePolicy
+      Policies:
+        - PolicyName: FastAPIExecutionPolicy
+          PolicyDocument:
+            Version: '2012-10-17'
+            Statement:
+              - Effect: Allow
+                Action:
+                  - ecr:GetAuthorizationToken
+                  - ecr:BatchCheckLayerAvailability
+                  - ecr:GetDownloadUrlForLayer
+                  - ecr:BatchGetImage
+                Resource: '*'
+              - Effect: Allow
+                Action:
+                  - logs:CreateLogStream
+                  - logs:PutLogEvents
+                Resource: 
+                  - !Sub arn:aws:logs:${AWS::Region}:${AWS::AccountId}:log-group:/ecs/${Environment}-acres-fastapi:*
+                  - !Sub arn:aws:logs:${AWS::Region}:${AWS::AccountId}:log-group:/ecs/${Environment}-acres-fastapi:log-stream:*
+  
+  # FastAPI Task Role - for runtime permissions
+  FastAPITaskRole:
+    Type: AWS::IAM::Role
+    Properties:
+      AssumeRolePolicyDocument:
+        Version: '2012-10-17'
+        Statement:
+          - Effect: Allow
+            Principal:
+              Service: ecs-tasks.amazonaws.com
+            Action: sts:AssumeRole
+      Policies:
+        - PolicyName: FastAPITaskPolicy
+          PolicyDocument:
+            Version: '2012-10-17'
+            Statement:
+              # Add specific permissions needed by your FastAPI application at runtime
+              - Effect: Allow
+                Action:
+                  - dynamodb:GetItem
+                  - dynamodb:PutItem
+                  - dynamodb:Query
+                Resource: !Sub arn:aws:dynamodb:${AWS::Region}:${AWS::AccountId}:table/${Environment}-acres-fastapi-table
+              # Allow FastAPI to make HTTP calls to Gradio service
+              - Effect: Allow
+                Action:
+                  - execute-api:Invoke
+                Resource: !Sub arn:aws:execute-api:${AWS::Region}:${AWS::AccountId}:*
+
+  # ECS Cluster
+  ECSCluster:
+    Type: AWS::ECS::Cluster
+    Properties:
+      ClusterName: !Ref ECSClusterName
+      Tags:
+        - Key: Environment
+          Value: !Ref Environment
+
+  # Load Balancer for Gradio
+  GradioALB:
+    Type: AWS::ElasticLoadBalancingV2::LoadBalancer
+    Properties:
+      Name: !Sub ${Environment}-acres-gradio-alb
+      Scheme: internet-facing
+      LoadBalancerAttributes:
+        - Key: idle_timeout.timeout_seconds
+          Value: '60'
+      Subnets:
+        - !Ref PublicSubnet1
+        - !Ref PublicSubnet2
+      SecurityGroups:
+        - !Ref GradioSecurityGroup
+
+  GradioTargetGroup:
+    Type: AWS::ElasticLoadBalancingV2::TargetGroup
+    Properties:
+      HealthCheckEnabled: true
+      HealthCheckIntervalSeconds: 30
+      HealthCheckPath: /
+      HealthCheckPort: 7860
+      HealthCheckTimeoutSeconds: 20
+      HealthyThresholdCount: 2
+      Name: !Sub ${Environment}-acres-gradio-tg
+      Port: 7860
+      Protocol: HTTP
+      TargetType: ip
+      UnhealthyThresholdCount: 5
+      VpcId: !Ref VPC
+      TargetGroupAttributes:
+        - Key: deregistration_delay.timeout_seconds
+          Value: '30'
+
+  GradioHTTPSListener:
+    # https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-elasticloadbalancingv2-listener.html
+    Type: AWS::ElasticLoadBalancingV2::Listener
+    Properties:
+      DefaultActions:
+        - Type: forward
+          TargetGroupArn: !Ref GradioTargetGroup
+      LoadBalancerArn: !Ref GradioALB
+      # Certificates:
+      #   - CertificateArn: !Ref CertificateArn
+      Port: 7860
+      Protocol: HTTP
+  # GradioHTTPListener:
+  #   # https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-elasticloadbalancingv2-listener.html
+  #   Type: AWS::ElasticLoadBalancingV2::Listener
+  #   Properties:
+  #     Protocol: HTTP
+  #     Port: 80
+  #     LoadBalancerArn: !Ref GradioALB
+  #     DefaultActions:
+  #       - Type: redirect
+  #         RedirectConfig:
+  #           Protocol: "HTTPS"
+  #           Port: 7860
+  #           Host: "#{host}"
+  #           Path: "/#{path}"
+  #           Query: "#{query}"
+  #           StatusCode: "HTTP_301"
+  
+  # Load Balancer for FastAPI
+  FastAPIALB:
+    Type: AWS::ElasticLoadBalancingV2::LoadBalancer
+    Properties:
+      Name: !Sub ${Environment}-acres-fastapi-alb
+      Scheme: internet-facing
+      LoadBalancerAttributes:
+        - Key: idle_timeout.timeout_seconds
+          Value: '60'
+      Subnets:
+        - !Ref PublicSubnet1
+        - !Ref PublicSubnet2
+      SecurityGroups:
+        - !Ref FastAPISecurityGroup
+
+  FastAPITargetGroup:
+    Type: AWS::ElasticLoadBalancingV2::TargetGroup
+    Properties:
+      HealthCheckEnabled: true
+      HealthCheckIntervalSeconds: 30
+      HealthCheckPath: /docs  # FastAPI's Swagger UI path
+      HealthCheckPort: 8000
+      HealthCheckTimeoutSeconds: 20
+      HealthyThresholdCount: 2
+      Name: !Sub ${Environment}-acres-fastapi-tg
+      Port: 8000
+      Protocol: HTTP
+      TargetType: ip
+      UnhealthyThresholdCount: 5
+      VpcId: !Ref VPC
+      TargetGroupAttributes:
+        - Key: deregistration_delay.timeout_seconds
+          Value: '30'
+
+  FastAPIHTTPSListener:
+    # https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-elasticloadbalancingv2-listener.html
+    Type: AWS::ElasticLoadBalancingV2::Listener
+    Properties:
+      DefaultActions:
+        - Type: forward
+          TargetGroupArn: !Ref FastAPITargetGroup
+      LoadBalancerArn: !Ref FastAPIALB
+      # Certificates:
+      #   - CertificateArn: !Ref CertificateArn
+      Port: 8000
+      Protocol: HTTP
+  # FastAPIHTTPListener:
+  #   # https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-elasticloadbalancingv2-listener.html
+  #   Type: AWS::ElasticLoadBalancingV2::Listener
+  #   Properties:
+  #     Protocol: HTTP
+  #     Port: 80
+  #     LoadBalancerArn: !Ref FastAPIALB
+  #     DefaultActions:
+  #       - Type: redirect
+  #         RedirectConfig:
+  #           Protocol: "HTTPS"
+  #           Port: 8000
+  #           Host: "#{host}"
+  #           Path: "/#{path}"
+  #           Query: "#{query}"
+  #           StatusCode: "HTTP_301"
+
+  # ECS Task Definitions
+  GradioTaskDefinition:
+    Type: AWS::ECS::TaskDefinition
+    Properties:
+      Family: !Sub ${Environment}-acres-gradio
+      RequiresCompatibilities:
+        - FARGATE
+      Cpu: !Ref GradioTaskDefinitionCPU
+      Memory: !Ref GradioTaskDefinitionMemory
+      NetworkMode: awsvpc
+      ExecutionRoleArn: !GetAtt GradioTaskExecutionRole.Arn
+      TaskRoleArn: !GetAtt GradioTaskRole.Arn
+      ContainerDefinitions:
+        - Name: gradio
+          Image: !Ref ContainerImageGradio
+          PortMappings:
+            - ContainerPort: 7860
+          LogConfiguration:
+            LogDriver: awslogs
+            Options:
+              awslogs-group: !Ref GradioLogGroup
+              awslogs-region: !Ref AWS::Region
+              awslogs-stream-prefix: gradio
+  
+  FastAPITaskDefinition:
+    Type: AWS::ECS::TaskDefinition
+    Properties:
+      Family: !Sub ${Environment}-acres-fastapi
+      RequiresCompatibilities:
+        - FARGATE
+      Cpu: !Ref FastAPITaskDefinitionCPU
+      Memory: !Ref FastAPITaskDefinitionMemory
+      NetworkMode: awsvpc
+      ExecutionRoleArn: !GetAtt FastAPITaskExecutionRole.Arn
+      TaskRoleArn: !GetAtt FastAPITaskRole.Arn
+      ContainerDefinitions:
+        - Name: fastapi
+          Image: !Ref ContainerImageFastAPI
+          PortMappings:
+            - ContainerPort: 8000
+          Environment:
+            - Name: GRADIO_URL
+              Value: !Sub http://${GradioALB.DNSName}:7860/
+          LogConfiguration:
+            LogDriver: awslogs
+            Options:
+              awslogs-group: !Ref FastAPILogGroup
+              awslogs-region: !Ref AWS::Region
+              awslogs-stream-prefix: fastapi
+
+  # CloudWatch Log Groups
+  GradioLogGroup:
+    Type: AWS::Logs::LogGroup
+    Properties:
+      LogGroupName: !Sub /ecs/${Environment}-acres-gradio
+      RetentionInDays: 30
+  
+  FastAPILogGroup:
+    Type: AWS::Logs::LogGroup
+    Properties:
+      LogGroupName: !Sub /ecs/${Environment}-acres-fastapi
+      RetentionInDays: 30
+
+
+  # ECS Services
+  GradioService:
+    Type: AWS::ECS::Service
+    DependsOn: 
+      - GradioHTTPSListener
+      # - GradioHTTPListener
+    Properties:
+      ServiceName: !Sub ${Environment}-acres-gradio
+      Cluster: !Ref ECSCluster
+      TaskDefinition: !Ref GradioTaskDefinition
+      DesiredCount: 1
+      LaunchType: FARGATE
+      HealthCheckGracePeriodSeconds: 180
+      LoadBalancers:
+        - ContainerName: gradio
+          ContainerPort: 7860
+          TargetGroupArn: !Ref GradioTargetGroup
+      NetworkConfiguration:
+        AwsvpcConfiguration:
+          AssignPublicIp: ENABLED
+          SecurityGroups:
+            - !Ref GradioSecurityGroup
+          Subnets:
+            - !Ref PublicSubnet1
+            - !Ref PublicSubnet2
+      DeploymentConfiguration:
+        DeploymentCircuitBreaker:
+          Enable: true
+          Rollback: true
+        MaximumPercent: 200
+        MinimumHealthyPercent: 100
+  
+  FastAPIService:
+    Type: AWS::ECS::Service
+    DependsOn: 
+      - GradioService
+      - FastAPIHTTPSListener
+      # - FastAPIHTTPListener
+    Properties:
+      ServiceName: !Sub ${Environment}-acres-fastapi
+      Cluster: !Ref ECSCluster
+      TaskDefinition: !Ref FastAPITaskDefinition
+      DesiredCount: 1
+      LaunchType: FARGATE
+      HealthCheckGracePeriodSeconds: 180
+      LoadBalancers:
+        - ContainerName: fastapi
+          ContainerPort: 8000
+          TargetGroupArn: !Ref FastAPITargetGroup
+      NetworkConfiguration:
+        AwsvpcConfiguration:
+          AssignPublicIp: ENABLED
+          SecurityGroups:
+            - !Ref FastAPISecurityGroup
+          Subnets:
+            - !Ref PublicSubnet1
+            - !Ref PublicSubnet2
+      DeploymentConfiguration:
+        DeploymentCircuitBreaker:
+          Enable: true
+          Rollback: true
+        MaximumPercent: 200
+        MinimumHealthyPercent: 100
+      # Add deployment controller for better rollout control
+      DeploymentController:
+        Type: ECS
+
+Outputs:
+  VpcId:
+    Description: VPC ID
+    Value: !Ref VPC
+
+  PublicSubnet1:
+    Description: Public Subnet 1
+    Value: !Ref PublicSubnet1
+
+  PublicSubnet2:
+    Description: Public Subnet 2
+    Value: !Ref PublicSubnet2
+
+  GradioServiceUrl:
+    Description: URL for the Gradio service
+    Value: !Sub http://${GradioALB.DNSName}:7860/
+
+  ECSClusterName:
+    Description: Name of the ECS cluster
+    Value: !Ref ECSCluster
+
+  GradioServiceName:
+    Description: Name of the Gradio service
+    Value: !GetAtt GradioService.Name
+  
+  FastAPIServiceName:
+    Description: Name of the FastAPI service
+    Value: !GetAtt FastAPIService.Name
+  
+  FastAPIServiceUrl:
+    Description: URL for the FastAPI service
+    Value: !Sub http://${FastAPIALB.DNSName}:8000/

pyproject.toml

@@ -0,0 +1,19 @@
+[tool.black]
+include = '\.pyi?$'
+exclude = '''
+/(
+    \.git
+  | \.hg
+  | \.mypy_cache
+  | \.tox
+  | \.venv
+  | env
+  |venv
+  | _build
+  | buck-out
+  | build
+  | dist
+  | migrations
+  |alembic
+)/
+'''

rag/rag_pipeline.py

@@ -1,19 +1,15 @@
 # rag/rag_pipeline.py
 import json
 import logging
-from typing import Dict, Any, List
+import re
+from typing import Any, Dict, List, Optional, Tuple
 
-from llama_index.core import Document, VectorStoreIndex
-from llama_index.core.node_parser import SentenceWindowNodeParser, SentenceSplitter
-from llama_index.core import PromptTemplate
+import chromadb
+from llama_index.core import Document, PromptTemplate, VectorStoreIndex
+from llama_index.core.node_parser import SentenceSplitter, SentenceWindowNodeParser
 from llama_index.embeddings.openai import OpenAIEmbedding
 from llama_index.llms.openai import OpenAI
 from llama_index.vector_stores.chroma import ChromaVectorStore
-import chromadb
-from typing import Dict, Any, List, Tuple, Optional
-import re
-import logging
-
 
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
@@ -172,7 +168,6 @@ class RAGPipeline:
             self.extract_page_number_from_query(context) if self.is_pdf else None
         )
 
-
         # This is a hack to index all the documents in the store :)
         n_documents = len(self.index.docstore.docs)
         print(f"n_documents: {n_documents}")

rag/rag_pipeline_backup.py

@@ -1,9 +1,8 @@
 import json
-from typing import Dict, Any
-from llama_index.core import Document, VectorStoreIndex
-from llama_index.core.node_parser import SentenceWindowNodeParser, SentenceSplitter
-from llama_index.core import PromptTemplate
-from typing import List
+from typing import Any, Dict, List
+
+from llama_index.core import Document, PromptTemplate, VectorStoreIndex
+from llama_index.core.node_parser import SentenceSplitter, SentenceWindowNodeParser
 from llama_index.embeddings.openai import OpenAIEmbedding
 from llama_index.llms.openai import OpenAI
 

requirements-dev.txt

@@ -0,0 +1,3 @@
+black==24.10.0
+isort==5.13.2
+flake8==7.1.1

requirements.txt

@@ -1,6 +1,7 @@
-chromadb==0.5.5
-fastapi==0.112.2
+chromadb
+fastapi
 gradio
+gradio_client
 llama-index
 llama-index-vector-stores-chroma
 nest-asyncio==1.6.0
@@ -11,4 +12,6 @@ python-dotenv
 pyzotero
 python-slugify
 PyMuPDF==1.23.8  
-Pillow==10.2.0 
+Pillow==10.2.0
+sqlmodel
+cachetools

study_files.json

@@ -1,8 +1,5 @@
 {
     "Vaccine coverage": "data/vaccine_coverage_zotero_items.json",
     "Ebola Virus": "data/ebola_virus_zotero_items.json",
-    "GeneXpert": "data/gene_xpert_zotero_items.json",
-    "Zotero Collection Pastan": "data/zotero-collection-pastan_zotero_items.json",
-    "EBSCOhost": "data/ebscohost_zotero_items.json",
-    "ExportedRis_file_1_of_1 (1)": "data/exportedris-file-1-of-1-1_zotero_items.json"
+    "GeneXpert": "data/gene_xpert_zotero_items.json"
 }

utils/db.py

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84acae8e51383d6990cd9edb7c1684292e523e7d0af87a71531bd5f9cf2909b5
+size 4907

utils/helpers.py

@@ -1,18 +1,18 @@
 # utils/helpers.py
 
-from typing import Dict, Any
+import json
+from typing import Any, Dict, List
+
+import chromadb
+from chromadb.api.types import Document
 from llama_index.core import Response
-from typing import List
+
 from rag.rag_pipeline import RAGPipeline
 from utils.prompts import (
-    structured_follow_up_prompt,
-    VaccineCoverageVariables,
     StudyCharacteristics,
+    VaccineCoverageVariables,
+    structured_follow_up_prompt,
 )
-import json
-import json
-import chromadb
-from chromadb.api.types import Document
 
 # Initialize ChromaDB client
 chromadb_client = chromadb.Client()

utils/pdf_processor.py

@@ -3,17 +3,17 @@ PDF processing module for ACRES RAG Platform.
 Handles PDF file processing, text extraction, and page rendering.
 """
 
-# utils/pdf_processor.py
-import os
-import fitz
-import logging
-from typing import Dict, List, Optional
 import datetime
-from slugify import slugify
 import json
-from PIL import Image
+import logging
+# utils/pdf_processor.py
+import os
 import re
+from typing import Dict, List, Optional
 
+import fitz
+from PIL import Image
+from slugify import slugify
 
 logger = logging.getLogger(__name__)
 

utils/prompts.py

@@ -1,9 +1,10 @@
 # utils/prompts.py
 
+from typing import List, Optional
+
 from llama_index.core import PromptTemplate
-from typing import Optional, List
-from pydantic import BaseModel, Field
 from llama_index.core.prompts import PromptTemplate
+from pydantic import BaseModel, Field
 
 
 class StudyCharacteristics(BaseModel):