MLOps/dvc.yaml
Alexis Bruneteau 9520395ee9 Fix DVC output path overlap in train stage
Changed from tracking entire models/ directory to specific model file
to resolve conflict with models/metrics.json metric tracking.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-10-01 20:51:16 +02:00

33 lines
750 B
YAML

stages:
preprocess:
cmd: python src/data/preprocess.py
deps:
- src/data/preprocess.py
- data/raw
params:
- preprocess.test_size
- preprocess.random_state
outs:
- data/processed/features.csv
- data/processed/train.csv
- data/processed/test.csv
metrics:
- data/processed/data_metrics.json:
cache: false
train:
cmd: python src/models/train_multitask.py
deps:
- src/models/train_multitask.py
- data/processed/train.csv
- data/processed/test.csv
params:
- train.n_estimators
- train.max_depth
- train.random_state
outs:
- models/multitask_model.pkl
metrics:
- models/metrics.json:
cache: false