Released versions of Bear
-
April XX, 2024:
Version 0.5
(XXX KB; MD5: )
-
“Bear Forest”: A complete overhaul of Bear, borrowing
liberally from the philosophy underlying random forests.
-
March 15, 2024:
Version 0.4.9
(365 KB; MD5: 7695088fada346176d6e8a518f52eb29)
-
Fixed a bug that caused Bear to give up in bootstrapping before finding its
first nontrivial model.
-
October 19, 2023:
Version 0.4.8
(365 KB; MD5: 894ba4844ddb40fdda0a16007803d2d7)
-
Improved the completeness modeling and the
messaging when the first run exhausts its time budget before
minimizing the loss.
-
October 11, 2023:
Version 0.4.7
(364 KB; MD5: d67ebafcf19e91ba5401833db9610055)
-
“Python Bear”: Added compare_data to confirm that
bear_predict.py gives the same
predictions as bear_predict.
-
October 9, 2023:
Version 0.4.6
(363 KB; MD5: f5db72e2e8ba163a96e35ca963fc091a)
-
“Bear Cull”: removes features not used in the final model and
does a second run with just the useful features.
-
October 4, 2023:
Version 0.4.5
(420 KB; MD5: 5daf00e9259044280f5ef7db216f3c4b)
-
Added support for the MAE (mean absolute error) loss function.
-
October 1, 2023:
Version 0.4.4
(413 KB; MD5: c5d68ee308f3a1e23d15db446e9f74ea)
-
“Auto Bear”: an exponentially increasing time budget
attempts to find the minimum-loss model.
-
September 25, 2023:
Version 0.4.3
(410 KB; MD5: c2d947eac1a1e245c80b485c34f75deb)
-
Bug fixes.
-
September 24, 2023:
Version 0.4.2
(410 KB; MD5: 426c4923b6d4b5941c1f80459272b2f4)
-
Bug fixes and proper handling of the balanced log loss function.
-
August 21, 2023:
Version 0.4.1
(400 KB; MD5: 7c441075e04ff3b21ccaf8a7dcaf57d2)
-
Bug fixes.
-
August 18, 2023:
Version 0.4
(399 KB; MD5: 4bbc10283179f6121d953f6719004197)
-
“Bear Bones”: returned the core philosophy of Bear back to its
2012 world-beating simplicity.
-
May 31, 2023:
Version 0.3.5
(413 KB; MD5: 07cf73e73f37c9d3296a221110a4cfc0)
-
Improved the model-building process and
removed my initial attempt
at Convolutional Bear.
-
May 9, 2023:
Version 0.3.4
(422 KB; MD5: 55fba9d8a28d57c1544bd883d567f516)
-
Hotfix for mistakes for features with missing values
when there is more than one feature.
-
May 6, 2023:
Version 0.3.3
(419 KB; MD5: 538d554d2d6694912dd8e83981830ce6)
-
Fixed two minor bugs that caused crashes in two of the trivial tutorial
examples.
Progress reporting improved.
-
May 5, 2023:
Version 0.3.2
(419 KB; MD5: e061944a5d65910c34d05188a9d1f8a6)
-
Improved the model-growing algorithm.
-
May 3, 2023:
Version 0.3.1
(418 KB; MD5: 622448acd3496e4e8213ea8e709e5f8f)
-
Performance improvements, including the introduction of the
paw 16-bit floating point format for Bear’s
internal use.
-
April 28, 2023:
Version 0.3
(433 KB; MD5: 2d08fd4bcf6cb35becc740e80840cb58)
-
Training and streamed deployment can be separated with
memory_bear
and bear_predict.
-
April 15, 2023:
Version 0.2.1
(422 KB; MD5: aed69d2596dba97b15e7b410f378546ff)
-
Performance improvements.
-
April 9, 2023:
Version 0.2
(424 KB; MD5: 16780167c4d76927121c906579d1a210)
-
Updated to automatically swap memory out to storage as required.
Input data should still fit in memory.
-
March 24, 2023:
Version 0.1.1
(409 KB; MD5: 1fc63af9abb4d8753bfe5fa0c4a9e8bf)
-
Updated to build cleanly on Ubuntu.
Instructions included for building on AWS EC2 Ubuntu instances.
-
March 21, 2023:
Version 0.1
(408 KB; MD5: 8e6e5a86ae51ea824e360a84da195f0f)
-
Initial release.
Fully in-memory beta implementation.
© 2023–2024 John Costella