src/config/config.yaml

# Model Configuration
model:
  name: "microsoft/layoutlmv3-base"
  device: "cuda"  # or "cpu"
  confidence_threshold: 0.5
  max_length: 512

# Training Configuration
training:
  batch_size: 8
  learning_rate: 2e-5
  num_epochs: 10
  warmup_steps: 100
  weight_decay: 0.01
  gradient_accumulation_steps: 4

# Dataset Configuration
dataset:
  train_path: "data/train"
  val_path: "data/val"
  test_path: "data/test"
  max_samples: null  # Set to null for all samples
  augmentation:
    enabled: true
    rotation_range: 10
    width_shift_range: 0.1
    height_shift_range: 0.1
    zoom_range: 0.1
    fill_mode: "nearest"

# Logging Configuration
logging:
  level: "INFO"
  wandb:
    enabled: true
    project: "formiq"
    entity: null  # Set your W&B username
  tensorboard:
    enabled: true
    log_dir: "logs"

# API Configuration
api:
  host: "0.0.0.0"
  port: 8000
  workers: 4
  timeout: 60

# Frontend Configuration
frontend:
  host: "localhost"
  port: 8501
  debug: false

# MLOps Configuration
mlops:
  dvc:
    remote: "s3://formiq-data"
    cache_dir: ".dvc/cache"
  mlflow:
    tracking_uri: "http://localhost:5000"
    experiment_name: "formiq"
  evidently:
    enabled: true
    drift_threshold: 0.1
    window_size: 1000