requirements.txt

# Core ML dependencies
torch==2.2.0
transformers==4.37.2
datasets>=2.12.0
pytesseract==0.3.10
Pillow==10.2.0

# API and Web Framework
fastapi==0.109.2
uvicorn==0.27.1
streamlit==1.32.0
python-multipart==0.0.9

# MLOps and Monitoring
wandb>=0.15.0
mlflow>=2.4.0
dvc>=3.0.0
hydra-core>=1.3.2
evidently>=0.2.0
tensorboard>=2.12.0

# Cloud and Deployment
boto3==1.34.34
sagemaker>=2.160.0

# Utilities
numpy==1.26.3
pandas==2.2.0
python-dotenv==1.0.1
pydantic>=2.0.0 
openai
streamlit
plotly==5.18.0
matplotlib
scikit-learn
pdf2image
paddleocr
paddlepaddle
easyocr