Adding metrics and plots to dvc.yaml
In this exercise, your task is to complete the contents of dvc.yaml
that defines a model training workflow.
Here preprocess_dataset.py
and train.py
are the files that perform data preprocessing and model training by taking weather.csv
as input in the raw_dataset
folder. As output, the model training code generates a predictions.csv
file that contains the predictions and the ground truth, and metrics.json
file containing structured metrics data. The former would be used to generate a normalized confusion matrix plot for comparing it with previous commits.
This exercise is part of the course
CI/CD for Machine Learning
Hands-on interactive exercise
Turn theory into action with one of our interactive exercises
