dvc-practice

1. GitHub repo creation
2. Clone the GitHub repo

---

3. Run dvc init
4. Download MNIST dataset from here
5. make MNIST dataset to be half link
6. place MNIST dataset under data/train.csv
7. Run dvc add data/train.csv
8. Run dvc remote add my_stroage -d /tmp/dvc-test
9. Run git add .
10. Run git commit -m "initial commit"
11. Run dvc push

---

12. Create params.yaml
13. Create src/
14. Create src/split.py
15. Run dvc run -n split -p split.ratio -d src/split.py -d data/train.csv -o data/prepared python src/split.py data/train.csv
16. Create src/preprocessing.py
17. Run dvc run -n preprocess -d src/preprocessing.py -d data/prepared -o data/preprocessed python src/preprocessing.py data/prepared
18. Create src/train.py
19. Run dvc run -n train -d src/train.py -d data/preprocessed -o data/model python src/train.py data/preprocessed data/model
20. Create src/evaluate.py [ ] Run `dvc run -n evaluate -d src/evaluate.py -d data/model

---