Add GS:Visualization and Plots (#3050)

Co-authored-by: Jorge Orpinel <[email protected]>

content/docs/sidebar.json

@@ -65,6 +65,13 @@
         "tutorials": {
           "katacoda": "https://katacoda.com/dvc/courses/get-started/experiments"
         }
+      },
+      {
+        "label": "Visualization with Plots",
+        "slug": "visualization",
+        "tutorials": {
+          "katacoda": "https://katacoda.com/dvc/courses/get-started/visualization"
+        }
       }
     ]
   },

content/docs/start/visualization.md

@@ -0,0 +1,109 @@
+---
+title: 'Get Started: Visualization with Plots'
+---
+
+# Get Started: Visualization with Plots
+
+In this section, we'll add visualization to the [`example-dvc-experiments`][ede]
+project (explored [previously](/doc/start/experiments)). If you would like to
+try these yourself, please refer to the project. [README] about how to install.
+
+[ede]: https://github.com/iterative/example-dvc-experiments
+[readme]:
+  https://github.com/iterative/example-dvc-experiments/blob/main/README.md
+
+## Creating plots from tabular data
+
+A useful plot to show the classification performance is the [confusion matrix].
+In order to produce it, DVC expects a CSV **plots file** in the form:
+
+```csv
+actual,predicted
+0,0
+0,2
+...
+```
+
+> We added a [loop] comparing the results to generate this file from the
+> predictions.
+
+[loop]:
+  https://github.com/iterative/example-dvc-experiments/blob/main/src/train.py#L123
+[confusion matrix]: https://en.wikipedia.org/wiki/Confusion_matrix
+
+Running the experiment with `dvc exp run` will produce `plots/confusion.csv`.
+Use `dvc plots show` to present it as an HTML file, and open it in the browser:
+
+```dvc
+$ dvc plots show plots/confusion.csv --template confusion \
+                                     -x actual -y predicted
+file:///.../example-dvc-experiments/plots/confusion.json.html
+```
+
+![confusion matrix](/img/start_visualization_confusion1.png)
+
+## Displaying user-generated plot images
+
+Let's produce another plot to see misclassified examples from each class. This
+procedure generates the misclassification examples from the validation data and
+arranges them into a _confusion table_ that shows the correct label, and
+misclassification sample. The code to generate an image from a set of training
+images is omitted here but you can find the code in [the example
+project.][misclassified-example-code]
+
+[misclassified-example-code]:
+  https://github.com/iterative/example-dvc-experiments/blob/48b1e5078c957f71674c00f416290eaa3b20b559/src/util.py#L49
+
+```dvc
+$ dvc plots show plots/misclassified.png
+```
+
+![Misclassification table](/img/start_visualization_misclassification.png)
+
+## Autogenerating plots from deep learning code
+
+An important issue for deep learning projects is to observe in which epoch do
+training and validation loss differ. DVC helps in that regard with its Python
+integrations to deep learning libraries via [DVCLive].
+
+The example project uses Keras to train a classifier, and we have a DVCLive
+callback that visualizes the training and validation loss for each epoch. We
+first import the callback from DVCLive.
+
+```python
+from dvclive.keras import DvcLiveCallback
+```
+
+Then we add this callback to the
+[`fit` method](https://keras.io/api/models/model_training_apis/#fit-method)
+call.
+
+```python
+model.fit(
+ ...
+ callbacks=[DvcLiveCallback()],
+ ...)
+```
+
+With these two changes, the model metrics are automatically logged to
+`dvclive.json` and plotted in `training_metrics/index.html`:
+
+![dvclive](/img/start_visualization_dvclive.png)
+
+DVCLive has other capabilities, like saving the model every epoch or modifying
+these default values.
+
+In summary, DVC provides more than one option to use visualization in your
+workflow:
+
+- DVC can generate HTML files that includes interactive plots from data series
+  in JSON, YAML, CSV, or TSV format.
+
+- DVC can keep track of image files produced as [plot outputs] from the
+  training/evaluation scripts.
+
+- [DVCLive] integrations can produce plots automatically during training.
+
+[plot outputs]:
+  /doc/user-guide/project-structure/pipelines-files#metrics-and-plots-outputs
+[dvclive]: /doc/dvclive/dvclive-with-dvc