Skip to content

Releases: kiudee/chess-tuning-tools

chess-tuning-tools 0.9.5 (19 July 2023)

19 Jul 15:48
9e27d2e
Compare
Choose a tag to compare

0.9.5 (2023-07-19)

  • No functional changes. Bump dependency range for scikit-learn. Support python 3.9.

chess-tuning-tools 0.9.4 (19 July 2023)

19 Jul 11:34
Compare
Choose a tag to compare

0.9.4 (2023-07-19)

  • No functional changes. Fixes dependency ranges. Now only compatible with python 3.8.

chess-tuning-tools 0.9.3 (1 April 2022)

02 Apr 11:41
Compare
Choose a tag to compare

0.9.3 (2022-04-01)

Local tuner

  • Add support for two-sided resign adjudication. If set to true, both engines
    have to report an absolute score higher than the resign_score threshold
    (#95).
  • Add support for engineX_restart (default "auto") which allows one to set
    the restart mode used by cutechess (#95).
  • Add depth-based time control using engineX_depth (#95).
  • Log the estimated draw rate of the current match (#197).

chess-tuning-tools 0.9.2 (13 March 2022)

13 Mar 15:18
Compare
Choose a tag to compare

0.9.2 (2022-03-13)

Local tuner

  • Add CLI flag --evaluate-points -p CSVFILE which allows the user to supply
    a CSV file with points the tuner should try before resuming normal operation.
    An optional integer column can be specified, which will be used to indicate
    the number of rounds to run each point for (#91).
  • Log version of the tuner in the log output (#191).

chess-tuning-tools 0.9.1 (06 March 2022)

09 Mar 20:39
6e150aa
Compare
Choose a tag to compare

0.9.1 (2022-03-06)

Local tuner

  • Fix a bug where the Elo and optima plots were showing incorrect iterations,
    if the user reduced parameter ranges (#183).
  • Only use scientific notation in the optima plots, if the parameter is
    optimized on a log scale (#182).

chess-tuning-tools 0.9.0 (11 February 2022)

11 Feb 21:02
Compare
Choose a tag to compare

0.9.0 (2022-02-11)

Local tuner:

  • Add a plot which shows the optima predicted by the tuner across the
    iterations (#172). This can be useful to gauge convergence.
  • Add a plot which shows the estimated Elo (+ confidence interval) of the
    predicted optima (#176).
  • Tuner saves optima and their Elo performance (including standard deviation)
    to disk now (#171).

chess-tuning-tools 0.8.3 (30 January 2022)

30 Jan 17:24
Compare
Choose a tag to compare

0.8.3 (2022-01-30)

  • Fix compatibility with Python 3.7 (#150).

chess-tuning-tools 0.8.2 (26 January 2022)

26 Jan 21:03
Compare
Choose a tag to compare

0.8.2 (2022-01-26)

Local tuner


* Add support for plotting one-dimensional optimization landscapes (#34).

chess-tuning-tools 0.8.1 (11 September 2021)

11 Sep 18:29
4dddb2e
Compare
Choose a tag to compare

0.8.1 (2021-09-11)

Local tuner

  • Emit errors and warnings for common cutechess-cli problems:

    • Unknown UCI parameters
    • An engine loses on time
    • The connection to an engine stalls (usually a crash)
    • Any other error captured by cutechess-cli
  • Add support for specifying nested parameters.

chess-tuning-tools 0.8.0 (15 August 2021)

15 Aug 19:05
2f48924
Compare
Choose a tag to compare

0.8.0 (2021-08-15)

Local tuner

  • Replace default lengthscale priors by inverse-gamma distributions.

  • Add the following command line flags, which allow the user to override the
    prior parameters:

    • --gp-signal-prior-scale for the scale of the signal prior.
    • --gp-noise-prior-scale for the scale of the noise prior.
    • --gp-lengthscale-prior-lb for the lower bound of the lengthscale prior.
    • --gp-lengthscale-prior-ub for the upper bound of the lengthscale prior.