Skip to content
Snippets Groups Projects
Select Git revision
  • 2a5865c555d7ff30605860db6d4064229315d89e
  • master default protected
  • emoUS
  • add_default_vectorizer_and_pretrained_loading
  • clean_code
  • readme
  • issue127
  • generalized_action_dicts
  • ppo_num_dialogues
  • crossowoz_ddpt
  • issue_114
  • robust_masking_feature
  • scgpt_exp
  • e2e-soloist
  • convlab_exp
  • change_system_act_in_env
  • pre-training
  • nlg-scgpt
  • remapping_actions
  • soloist
20 results

plot_results

  • Open with
  • Download source code
  • Download directory
  • Your workspaces

      A workspace is a virtual sandbox environment for your code in GitLab.

      No agents available to create workspaces. Please consult Workspaces documentation for troubleshooting.

  • user avatar
    Carel van Niekerk authored
    04024645
    History
    Code owners
    Assign users and groups as approvers for specific file changes. Learn more.
    Name Last commit Last update
    ..
    README.md
    example_map.json
    plot.py

    Plot training curve

    Plot training curve directly from tensorboard event files.

    The file structure of tb_file is like this:

    .
    ├── dir                  
        └── map["dir"]
            └── experiment_*
                └── tb_dir
                    └── events.* 

    The structure of map file is like this:

    [
      {
        "dir": "Algorithm1_results",
        "legend": "Algorithm1"
      },
      {
        "dir": "Algorithm2_results",
        "legend": "Algorithm2"
      }
    ]

    The value of legend will be the depicted in the plot legend.

    If you want to truncate the figure to a certain number of training dialogues on the x-axis, use the argument --max-dialogues.