plots show

Generate plot from a plots file or plots top-level definition from dvc.yaml.

Synopsis

usage: dvc plots show [-h] [-q | -v] [-t <name_or_path>] [-x <field>]
                      [-y <field>] [--no-header] [--title <text>]
                      [--x-label <text>] [--y-label <text>] [-o <path>]
                      [--show-vega] [--open] [--html-template <path>]
                      [targets [targets ...]]

positional arguments:
  targets               Plots files or plot IDs from `dvc.yaml` to
                        visualize. Shows all plots by default.

Description

This command provides a quick way to visualize certain data such as loss functions, AUC curves, confusion matrices, etc.

All plots defined in dvc.yaml are used by default, but specific plots files or top-level plot IDs can be specified as targets (note that target files don't necessarily have to be defined in dvc.yaml).

The plot style can be customized with plot templates, using the --template option. To learn more about plots file formats and templates, see dvc plots.

The default behavior of this command can be modified per stage plot file with dvc plots modify.

Options

-o <path>, --out <path> - specify a directory to write the HTML file containing the plots. The default is dvc_plots or the value set with the plots.out_dir config option.
-t <name_or_path>, --template <name_or_path> - plot template to be injected with data. The default template is .dvc/plots/default.json. See more details in dvc plots.
-x <field> - field name from which the X axis data comes from. An auto-generated index field is used by default. See Custom templates for more information on this index field. Column names or numbers are expected for tabular plots files.
-y <field> - field name from which the Y axis data comes from. The last field found in the targets is used by default. Column names or numbers are expected for tabular plots files.
--x-label <text> - X axis label. The X field name is the default.
--y-label <text> - Y axis label. The Y field name is the default.
--title <text> - plot title.
--show-vega - produce a Vega-Lite spec file instead of HTML. See dvc plots for more info.
--open - open the HTML generated in a browser automatically. You can enable dvc config plots.auto_open to make this the default behavior.
--no-header - lets DVC know that CSV or TSV targets do not have a header. A 0-based numeric index can be used to identify each column instead of names.
--html-template <path> - path to a custom HTML template.
-h, --help - prints the usage/help message, and exit.
-q, --quiet - do not write anything to standard output. Exit with 0 if no problems arise, otherwise 1.
-v, --verbose - displays detailed tracing information.

Example: Hierarchical data

We'll use hierarchical metrics file train.json for this example:

{
  "train": [
    { "accuracy": 0.96658, "loss": 0.10757 },
    { "accuracy": 0.97641, "loss": 0.07324 },
    { "accuracy": 0.87707, "loss": 0.08136 },
    { "accuracy": 0.87402, "loss": 0.09026 },
    { "accuracy": 0.8795, "loss": 0.0764 },
    { "accuracy": 0.88038, "loss": 0.07608 },
    { "accuracy": 0.89872, "loss": 0.08455 }
  ]
}

Expand for YAML format

Here's a corresponding train.yaml metrics file:

train:
  - accuracy: 0.96658
    loss: 0.10757
  - accuracy: 0.97641
    loss: 0.07324
  - accuracy: 0.87707
    loss: 0.08136
  - accuracy: 0.87402
    loss: 0.09026

DVC identifies and plots JSON objects from the first JSON array found in the file (train):

$ dvc plots show train.json
file:///Users/usr/src/dvc_plots/index.html

Note that only the last field name (loss) is used for the plot by default.

Use the -y option to change the field to plot:

$ dvc plots show -y accuracy train.json
file:///Users/usr/src/dvc_plots/index.html

Example: Tabular data

We'll use tabular metrics file logs.csv for these examples:

epoch,loss,accuracy
1,0.19,0.81
2,0.11,0.89
3,0.07,0.93
4,0.04,0.96

Expand for TSV format

Here's a corresponding train.tsv metrics file:

epoch	loss	accuracy
1	0.19	0.81
2	0.11	0.89
3	0.07	0.93
4	0.04	0.96

By default, this command plots the last column of the table (see -y option):

$ dvc plots show logs.csv
file:///Users/usr/src/dvc_plots/index.html

Use the -y option to change the column to plot:

$ dvc plots show logs.csv -y loss
file:///Users/usr/src/dvc_plots/index.html

Headerless tables

A tabular data file without headers can be plotted with --no-header option. A column can be specified with -y by it's numeric position (starting with 0):

$ dvc plots show --no-header logs.csv -y 2
file:///Users/usr/src/dvc_plots/index.html

Example: Top-level plots

Simple plot definition

Let's work with the following logs.csv data:

epoch,loss,accuracy
1,0.19,0.81
2,0.11,0.89
3,0.07,0.93
4,0.04,0.96

The minimal plot configuration we can put in dvc.yaml is the data source path:

stages:
  train:
    cmd: ...

plots:
  - logs.csv

$ dvc plots show
file:///Users/usr/src/dvc_plots/index.html

We can also customize it:

plots:
  - logs.csv:
      x: epoch
      y: accuracy
      title: Displaying accuracy
      x_label: This is epoch
      y_label: This is accuracy

$ dvc plots show
file:///Users/usr/src/dvc_plots/index.html

Multiple data series plot

Data in training_data.csv:

epoch,train_loss,test_loss
1,0.33,0.4
2,0.3,0.28
3,0.2,0.25
4,0.1,0.23

Plot definition in dvc.yaml:

plots:
  - test_vs_train_loss:
      x: epoch
      y:
        training_data.csv: [test_loss, train_loss]
      title: Compare loss training versus test

$ dvc plots show
file:///Users/usr/src/dvc_plots/index.html

Sourcing data from different files

Lets prepare a comparison for confusion matrix data between the train_classes.csv and a test_classes.csv datasets (below):

actual_class,predicted_class
dog,dog
dog,dog
dog,dog
dog,bird
cat,cat
cat,cat
cat,cat
cat,dog
bird,bird
bird,bird
bird,bird
bird,dog

actual_class,predicted_class
dog,dog
dog,dog
dog,cat
bird,bird
bird,bird
bird,cat
cat,cat
cat,cat
cat,bird

In dvc.yaml:

plots:
  - test_vs_train_confusion:
      x: actual_class
      y:
        train_classes.csv: predicted_class
        test_classes.csv: predicted_class
      title: Compare test vs train confusion matrix
      template: confusion
      x_label: Actual class
      y_label: Predicted class

$ dvc plots show
file:///Users/usr/src/dvc_plots/index.html

Sourcing `x` and `y` from different files

The actual and predicted classes may come from different data sources:

actual_class
dog
dog
dog
bird
bird
bird
cat
cat
cat

predicted_class
dog
dog
cat
bird
bird
cat
cat
cat
bird

In dvc.yaml:

plots:
  - confusion:
      x:
        actual.csv: actual_class
      y:
        preds.csv: predicted_class
      template: confusion

$ dvc plots show
file:///Users/usr/src/dvc_plots/index.html

Example: Vega-Lite specification file

In many automation scenarios (like CI/CD for ML), it is convenient to output the Vega-Lite spec file instead of rendering an HTML plot. For example, to generating another image format like PNG or JPEG, or to include it differently into a web/mobile app. The --show-vega option prevents wrapping this plot spec in HTML, printing the resulting JSON to standard output instead:

$ dvc plots show --show-vega logs.csv -y accuracy

{
    "$schema": "https://vega.github.io/schema/vega-lite/v4.json",
    "data": {
        "values": [
    {
        "accuracy": "0.9418667",
    ...

Custom HTML templates

It's possible to supply an HTML file to dvc plots show and dvc plots diff by using the the --html-template option. This allows you to customize the container where DVC will inject plots it generates.

⚠️ This is a separate feature from custom Vega-Lite templates.

The only requirement for this HTML file is to specify the place to inject plots with a {plot_divs} marker. See an example that uses this feature to render DVC plots without an Internet connection, below.

Example: Offline HTML Template

The plots generated by dvc plots uses Vega-Lite JavaScript libraries, and by default these load online resources. There may be times when you need to produce plots without Internet access, or want to customize the plots output to put some extra content, like banners or extra text. DVC allows to replace the HTML file that contains the final plots.

Download the Vega-Lite libraries into the directory where you'll produce the dvc plots:

$ wget https://cdn.jsdelivr.net/npm/vega@5.20.2 -O my_vega.js
$ wget https://cdn.jsdelivr.net/npm/vega-lite@5.1.0 -O my_vega_lite.js
$ wget https://cdn.jsdelivr.net/npm/vega-embed@6.18.2 -O my_vega_embed.js

Create the following HTML file and save it in .dvc/plots/mypage.html:

<html>
  <head>
    <script src="../path/to/my_vega.js" type="text/javascript"></script>
    <script src="../path/to/my_vega_lite.js" type="text/javascript"></script>
    <script src="../path/to/my_vega_embed.js" type="text/javascript"></script>
  </head>
  <body>
    {plot_divs}
  </body>
</html>

Note that this is a standard HTML file with only {plot_divs} as a placeholder for DVC to inject plots. <script> tags in this file point to the local JavaScript libraries we downloaded above. We can use it like this:

$ dvc plots show --html-template .dvc/plots/mypage.html

You can also make it the default HTML template by setting it as dvc config parameter plots.html_template.

$ dvc config plots.html_template plots/mypage.html

Note that the path supplied to dvc config plots.html_template is relative to .dvc/ directory.

Example: Smooth plot

In some cases we would like to smooth our plot. In this example we will use a noisy plot with 100 data points:

$ dvc plots show data.csv
file:///Users/usr/src/dvc_plots/index.html

We can use the -t (--template) option and smooth template to make it less noisy:

$ dvc plots show -t smooth data.csv
file:///Users/usr/src/dvc_plots/index.html

Example: Confusion matrix

We'll use classes.csv for this example:

actual,predicted
cat,cat
cat,cat
cat,cat
cat,dog
cat,dinosaur
cat,dinosaur
cat,bird
turtle,dog
turtle,cat
...

Let's visualize it:

$ dvc plots show classes.csv --template confusion \
                             -x actual -y predicted
file:///Users/usr/src/dvc_plots/index.html

A confusion matrix template is predefined in DVC.

We can use confusion_normalized template to normalize the results:

$ dvc plots show classes.csv -t confusion_normalized \
                             -x actual -y predicted
file:///Users/usr/src/dvc_plots/index.html

Example: Horizontal bar plot

This is a simple bar plot useful for example in visualizing model feature importances.

We'll use importances.csv for this example:

feature_name,feature_importance
petal_width,0.4
petal_length,0.33
sepal_width,0.24
sepal_length,0.03