Web Portal

Web portal is for users to conveniently visualize their NNI experiments, tuning and training progress, detailed metrics, and error logs. Web portal also allows users to control their NNI experiments, trials, such as updating an experiment of its concurrency, duration, rerunning trials.

../../_images/webui.gif

Q&A

There are many trials in the detail table but Default Metric chart is empty

Note

First you should know that Default metric and Hyper parameter chart only show succeeded trials.

What should you do when you think the chart is strange, such as Default metric, Hyper parameter

  • Download the experiment results(experiment config, trial message and intermeidate metrics) from Experiment summary and then upload these results in your issue.

../../_images/summary.png

What should you do when your experiment has error

  • Click the icon in the right of experiment status and screenshot the error message.

  • And then click the learn more to download nni-manager and dispatcher logfile.

  • Please file an issue from the Feedback in the About and upload above message.

../../_images/experiment_error.gif

What should you do when your trial fails

  • Customized trial could be used in here. Just submit the same parameters to the experiment to rerun the trial.

../../_images/customized_trial_button.png ../../_images/customized_trial.png
  • Log model will help you find the error reason. There are three buttons View trial log, View trial error and View trial stdout on local mode. If you run on the OpenPAI or Kubeflow platform, you could see trial stdout and nfs log. If you have any question you could tell us in the issue.

    local mode:

    ../../_images/log_local.png

    OpenPAI, Kubeflow and other mode:

    ../../_images/detail_pai.png

How to use dict intermediate result

The discussion could help you.

Experiments management

Experiments management page could manage many experiments on your machine.

../../_images/nav.png
  • On the All experiments page, you can see all the experiments on your machine.

../../_images/list.png
  • When you want to see more details about an experiment you could click the trial id, look that:

../../_images/to_another_exp.png
  • If has many experiments on the table, you can use the filter button.

../../_images/filter.png

Experiment details

View overview page

  • On the overview tab, you can see the experiment information and status and the performance of top trials.

../../_images/full_oview.png
  • If you want to see experiment search space and config, please click the right button Search space and Config (when you hover on this button).

    Search space file:

    ../../_images/search_space.png

    Config file:

    ../../_images/config.png
  • You can view and download nni-manager/dispatcher log files on here.

../../_images/review_log.png
  • If your experiment has many trials, you can change the refresh interval here.

../../_images/refresh_interval.png
  • You can change some experiment configurations such as maxExecDuration, maxTrialNum and trial concurrency on here.

../../_images/edit_experiment_param.gif

View job default metric

  • Click the tab Default metric to see the point chart of all trials. Hover to see its specific default metric and search space message.

../../_images/default_metric.png
  • Turn on the switch named Optimization curve to see the experiment’s optimization curve.

../../_images/best_curve.png

View hyper parameter

Click the tab Hyper-parameter to see the parallel chart.

  • You can click the add/remove button to add or remove axes.

  • Drag the axes to swap axes on the chart.

  • You can select the percentage to see top trials.

../../_images/hyper_parameter.png

View Trial Duration

Click the tab Trial Duration to see the bar chart.

../../_images/trial_duration.png

View Trial Intermediate Result chart

Click the tab Intermediate Result to see the line chart.

../../_images/trials_intermeidate.png

The trial may have many intermediate results in the training process. In order to see the trend of some trials more clearly, we set a filtering function for the intermediate result chart.

You may find that these trials will get better or worse at an intermediate result. This indicates that it is an important and relevant intermediate result. To take a closer look at the point here, you need to enter its corresponding X-value at #Intermediate. Then input the range of metrics on this intermedia result. In the picture below, we choose the No. 4 intermediate result and set the range of metrics to 0.8-1.

../../_images/filter_intermediate.png

View trials status

Click the tab Trials Detail to see the status of all trials. Specifically:

  • Trial detail: trial’s id, trial’s duration, start time, end time, status, accuracy, and search space file.

../../_images/detail_local.png
  • Support searching for a specific trial by its id, status, Trial No. and trial parameters.

    Trial id:

    ../../_images/search_id.png

    Trial No.:

    ../../_images/search_No.png

    Trial status:

    ../../_images/search_status.png

    Trial parameters:

    parameters whose type is choice:

    ../../_images/search_parameter_choice.png

    parameters whose type is not choice:

    ../../_images/search_parameter_range.png
  • The button named Add column can select which column to show on the table. If you run an experiment whose final result is a dict, you can see other keys in the table. You can choose the column Intermediate count to watch the trial’s progress.

../../_images/add_column.png
  • If you want to compare some trials, you can select them and then click Compare to see the results.

../../_images/select_trial.png ../../_images/compare.png
  • You can use the button named Copy as python to copy the trial’s parameters.

../../_images/copy_parameter.png
  • Intermediate Result chart: you can see the default metric in this chart by clicking the intermediate button.

../../_images/intermediate.png
  • Kill: you can kill a job that status is running.

../../_images/kill_running.png