Get Started: Metrics, Plots, and Parameters

DVC makes it easy to track metrics, visualize performance with plots, and update parameters. These concepts are introduced below.

All of the above can be combined into experiments to run and compare many iterations of your ML project.

Collecting metrics and plots

First, let's see the mechanism to capture values for these ML attributes. Add a final evaluation stage to our earlier pipeline:

$ dvc stage add -n evaluate \
  -d src/evaluate.py -d model.pkl -d data/features \
  -o eval \
  python src/evaluate.py model.pkl data/features

💡 Expand to get a peek under the hood

dvc stage add will generates this new stage in the dvc.yaml file:

evaluate:
  cmd: python src/evaluate.py model.pkl data/features
  deps:
    - data/features
    - model.pkl
    - src/evaluate.py
  outs:
    - eval

We cache your metrics and plots files with DVC, by making eval directory as a stage output the same way outputs were added in previous stages. This is the easiest way to handle this, and if amount of files and size is growing it doesn't affect your Git history. Alternatively it could be setup in more granular way to track certain metrics files or plots in Git, while other files could still be tracked by DVC.

evaluate.py uses DVCLive to write scalar metrics values (e.g. AUC) and plots data (e.g. ROC curve) to files in the eval directory that DVC can parse to compare and visualize across iterations. By default, DVCLive will configure metrics and plots for you in dvc.yaml, but in this example we customize them by editing dvc.yaml to combine train and test plots.

💡 Expand to see how to customize metrics and plots

To combine train and test data, and to set other custom attributes like titles, we add the following to dvc.yaml:

metrics:
  - eval/metrics.json
plots:
  - ROC:
      template: simple
      x: fpr
      y:
        eval/plots/sklearn/roc/train.json: tpr
        eval/plots/sklearn/roc/test.json: tpr
  - Confusion-Matrix:
      template: confusion
      x: actual
      y:
        eval/plots/sklearn/cm/train.json: predicted
        eval/plots/sklearn/cm/test.json: predicted
  - Precision-Recall:
      template: simple
      x: recall
      y:
        eval/prc/train.json: precision
        eval/prc/test.json: precision
  - eval/importance.png

This flexibility to define your own metrics and plots configuration means that you can even generate your own metrics and plots data without using DVCLive!

Let's run and save these changes:

$ dvc repro
$ git add .gitignore dvc.yaml dvc.lock eval
$ git commit -a -m "Create evaluation stage"

Viewing metrics and plots

You can view metrics and plots from the command line, or you can load your project in VS Code and use the DVC Extension to view metrics, plots, and more.

You can view tracked metrics with dvc metrics show :

$ dvc metrics show
Path                    avg_prec.test    avg_prec.train    roc_auc.test    roc_auc.train
eval/metrics.json  0.94496          0.97723           0.96191         0.98737

You can view plots with dvc plots show (shown below), which generates an HTML file you can open in a browser.

$ dvc plots show
file:///Users/dvc/example-get-started/dvc_plots/index.html

Later we will see how to compare and visualize different pipeline iterations. For now, let's see how to capture another important piece of information which will be useful for comparison: parameters.

Defining stage parameters

It's pretty common for data science pipelines to include configuration files that define adjustable parameters to train a model, do pre-processing, etc. DVC provides a mechanism for stages to depend on the values of specific sections of such a config file (YAML, JSON, TOML, and Python formats are supported).

Luckily, we should already have a stage with parameters in dvc.yaml:

featurize:
  cmd: python src/featurization.py data/prepared data/features
  deps:
    - data/prepared
    - src/featurization.py
  params:
    - featurize.max_features
    - featurize.ngrams
  outs:
    - data/features

⚙️ Expand to recall how it was generated.

The featurize stage was created with this dvc stage add command. Notice the argument sent to the -p option (short for --params):

$ dvc stage add -n featurize \
          -p featurize.max_features,featurize.ngrams \
          -d src/featurization.py -d data/prepared \
          -o data/features \
          python src/featurization.py data/prepared data/features

The params section defines the parameter dependencies of the featurize stage. By default, DVC reads those values (featurize.max_features and featurize.ngrams) from a params.yaml file. But as with metrics and plots, parameter file names and structure can also be user- and case-defined.

Here's the contents of our params.yaml file:

prepare:
  split: 0.20
  seed: 20170428

featurize:
  max_features: 100
  ngrams: 1

train:
  seed: 20170428
  n_est: 50
  min_split: 2

Updating params and iterating

We are definitely not happy with the AUC value we got so far! Let's edit the params.yaml file to use bigrams and increase the number of features:

 featurize:
-  max_features: 100
-  ngrams: 1
+  max_features: 200
+  ngrams: 2

The beauty of dvc.yaml is that all you need to do now is run:

$ dvc repro

It'll analyze the changes, use existing results from the run cache, and execute only the commands needed to produce new results (model, metrics, plots).

The same logic applies to other possible adjustments — edit source code, update datasets — you do the changes, use dvc repro, and DVC runs what needs to be run.

Comparing iterations

Finally, let's see how the updates improved performance. DVC has a few commands to see changes in and visualize metrics, parameters, and plots. These commands can work for one or across multiple pipeline iteration(s). Let's compare the current "bigrams" run with the last committed "baseline" iteration:

$ dvc params diff
Path         Param                   HEAD  workspace
params.yaml  featurize.max_features  100   200
params.yaml  featurize.ngrams        1     2

dvc params diff can show how params in the workspace differ vs. the last commit.

dvc metrics diff does the same for metrics:

$ dvc metrics diff
Path               Metric          HEAD     workspace    Change
eval/metrics.json  avg_prec.test   0.9014   0.925        0.0236
eval/metrics.json  avg_prec.train  0.95704  0.97437      0.01733
eval/metrics.json  roc_auc.test    0.93196  0.94602      0.01406
eval/metrics.json  roc_auc.train   0.97743  0.98667      0.00924

And finally, we can compare all plots with a single command (we show only some of them for simplicity):

$ dvc plots diff
file:///Users/dvc/example-get-started/plots.html

See dvc plots diff for more info on its options.

All these commands also accept Git revisions (commits, tags, branch names) to compare.