Hi @kupruser,
The following is an example of what I did.
- create train_split.dvc with train_split directory as output
dvc run -f train_split.dvc -d src/config/train_split.json -d data/raw/train.csv.zip -o data/processed/train_split python src/data/train_split.py
- create build_features_target_enc with train_split directory as input
dvc run -f build_features_target_enc.dvc -d src/config/build_features_target_enc_smooth.json -d data/processed/train_split -d data/raw/test.csv.zip -o data/processed/target_enc_train_features.csv.gz -o data/processed/target_enc_test_features.csv.gz python src/features/build_features_target_enc_smooth.py
- After doing some other work the files inside the train_split directory had changed. I tried to rerun the pipeline expecting it to pick up on those changes and run itself from the build_features_target_enc step.
dvc repro build_features_target_enc.dvc
but it said that there were no changes.