Commit History

  1. fix: Put ALL_LANGUAGES and ALL_TASKS at the end Running

    317dd0a
  2. fix: Use absolute imports

    c8f17da
  3. fix: Add missing datasets

    5451e45
  4. chore: Add makefile

    f2e50bb
  5. fix: Remove hashes from requirements.txt

    a013ca7
  6. fix: Update sdk_version in readme

    a0453dc
  7. feat: Add requirements.txt, update datasets

    d288ea3
  8. fix: Remove breakpoint

    83ff0f7
  9. feat: Update languages and datasets, add compat with eee

    da3cf26
  10. feat: Change BERTScore to ChrF3++

    180adc7
  11. feat: Update languages, and switch to using pyproject.toml

    5053774

  1. fix: Change Gradio version to 4.39.0

    Verified signature
    a25e7f4

  1. fix: Fetch data from tar.gz file

    a27fbcc

  1. fix: Separate zero-shot performance from few-shot

    376f461

  1. feat: Add Spanish and Finnish

    60d6a88

  1. style: Re-order

    4f27e41
  2. fix: Do not update last_fetch if nothing in dropdowns

    0495bc2

  1. fix: Tooltip

    6fd0cfa
  2. feat: Use actual ranks on scale

    c97530c
  3. feat: Update datasets

    6bdb37f

  1. feat: Add support for Italian

    cdd9094
  2. style: Rename ScandEval to EuroEval

    9fa29df

  1. style: Rename "reasoning" to "common-sense reasoning"

    e5acaa3

  1. fix: Use all datasets from a task, use ranks instead of log_ranks

    637c71d
  2. feat: Add update colours buttom

    4a4e1c1
  3. fix: Do nothing in update_colour_mapping if model_ids is empty

    86e510b
  4. feat: Add hotter-and-colder-sentiment

    ddc1d20
  5. style: Do not upper case language codes

    bef135e
  6. fix: Use language codes if more than 5 languages are selected

    4907a8b
  7. feat: Optimise colour mapping for visible models only

    1d11c02

  1. fix: Fetch new results properly

    64071e4
  2. fix: Do not use "test"

    a1248d7
  3. fix: Use new URL for results

    5e55e05

  1. Update app.py

    Verified signature
    fd7fab5
  2. Update app.py

    Verified signature
    5c9ed9a
  3. Update README.md (#1)⁠

    Verified signature
    c24aee4

  1. fix: Lower case model sorting

    6e9ab8e
  2. feat: Change UPDATE_FREQUENCY_MINUTES to 5

    9e3c3cd

  1. chore: Small update

    3a84b3b
  2. feat: Update app with log rank scores

    5f70754
  3. fix: Update win ratios to take ranks into account

    734648f
  4. feat: Add update colours button

    c34e772

  1. feat: Update datasets used in ScandEval

    8157f53

  1. feat: Sorting is case-independent

    f04c64c
  2. fix: Sort models and languages in the beginning

    995f0f4
  3. feat: Sort dropdown list of model IDs

    27bc6fa

  1. feat: Change order of tasks, to avoid hiding INFORMATION_EXTRACTION

    437ac86
  2. feat: Fix colour for each model (up to retakes), reduce logging

    a73e53c

  1. chore: Revert last change

    576340d
  2. feat: Use experimental nested t-test to determine statistical significance

    ada1f6c