# Environment Configuration Example
# Copy this file to .env and modify values as needed

# =================================
# Model Configuration
# =================================
MODEL_TYPE=vader
# Options: vader, distilbert

# For DistilBERT model
DISTILBERT_MODEL=distilbert-base-uncased-finetuned-sst-2-english

# =================================
# Processing Configuration
# =================================
BATCH_SIZE=32
MAX_LENGTH=512
TEXT_COLUMN=Text
LABEL_COLUMN=Score

# =================================
# Performance Settings
# =================================
USE_GPU=false
# Set to true if you have CUDA-capable GPU

NUM_WORKERS=4
# Number of CPU cores to use for processing

CACHE_PREDICTIONS=true
# Cache model predictions for faster repeated processing

# PyTorch optimization
OMP_NUM_THREADS=4
# Optimize CPU usage for PyTorch

# =================================
# API Configuration
# =================================
API_HOST=0.0.0.0
API_PORT=8080
# REST API server configuration

METRICS_PORT=8000
# Prometheus metrics port

# =================================
# Logging Configuration
# =================================
LOG_LEVEL=INFO
# Options: DEBUG, INFO, WARNING, ERROR

PYTHONUNBUFFERED=1
# Ensure logs are flushed immediately

# =================================
# Dataset Configuration
# =================================
# Kaggle dataset download limits (for development)
AMAZON_REVIEWS_LIMIT=50000
SENTIMENT140_LIMIT=100000

# Kaggle API credentials (optional - for dataset downloads)
# KAGGLE_USERNAME=your_username
# KAGGLE_KEY=your_api_key

# =================================
# Development Settings
# =================================
# Flask development mode
FLASK_ENV=production
# Options: development, production

# Debug mode
DEBUG=false
# Set to true for development debugging

# =================================
# Docker Configuration
# =================================
# Container resource limits
MEMORY_LIMIT=2g
CPU_LIMIT=2