-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Verify univariate Gaussian KL divergence calculation #15
Comments
The code was calculating KL(q||p) rather than KL(p||q). Fixed. |
bmmalone
added a commit
that referenced
this issue
Apr 8, 2019
Squashed commit of the following: commit 3aa5b0173741fed399055a17b289b83c2d359e49 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:14:51 2019 +0200 MNT bump for versions 1.0.1 commit 6410d3f91c0bd07973ce43b51cb02bb1cdbac586 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:06:36 2019 +0200 DOC domain-specific docs page, initial physionet docs commit 84cd9c64639caa54b5fc849c0ca61a306e560456 Merge: e5cdf63 fd81736 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:03:17 2019 +0200 Merge branch 'dev' of github.com:bmmalone/pyllars into dev commit fd81736a773939c11ca1c2810bb1a0126d7f1bfd Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:02:54 2019 +0200 FIX missing imports from hp utils commit e5cdf63d819fd659c36d326bceadebb9400843a6 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:02:11 2019 +0200 UPD mimic file utils commit 33232694b28ce620aff2ccb7a82ee733a028ab79 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Mar 28 21:09:33 2019 +0100 FIX typos, missing import for multiclass metrics commit 7dd21fd18fceb706be083a2ee8be1163dbf61adc Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Mar 19 23:16:54 2019 +0100 ADD xgboost utilities commit 95ef3307a42aeb6a1bb869be1f9a131e1775030f Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 22:55:42 2019 +0100 DOC matrix_utils commit 68946578c27a8a90daedd60c3c4a8a7d72498276 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 21:55:08 2019 +0100 ADD nlp, stats tests commit 65c685b3f597b1ad330009eeea061d0059607ee6 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 21:54:20 2019 +0100 FIX univariate gaussian kl divergence This addresses Issue #15. commit 44b1570e4cab2a70da1a8c859286e49672e36a44 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 18:32:47 2019 +0100 FIX docs, based on version of sphinx on read-the-docs commit 23daf4a8c6627f567d241bb8799e48f3c106e077 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 16:40:20 2019 +0100 DOC nlp_utils commit faa903a55f6d503bce7e6beedf7ca8d3f2f76d3a Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 16:13:07 2019 +0100 DOC mpl_utils commit e366e61e54f62f5c1538dd9836fec9aea330ec4b Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 18 17:23:11 2019 +0100 DOC [WIP] mpl_utils documentation This commit also cleans up some of the mpl_utils code. commit f798a1deec8197046cecbc44c0908f614cf7a71b Merge: c4027f7 0d030d6 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 17 20:05:10 2019 +0100 Merge branch 'dev' of github.com:bmmalone/pyllars into dev commit 0d030d66a41f66814fe36b6b582bb2e355f7b0d7 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 13 11:29:12 2019 +0100 FIX missing dependency, requests commit d50db25ef5c03196f2959f7a6a57f711ca456a6c Merge: 85885c5 844ada2 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 13 10:25:37 2019 +0100 Merge branch 'dev' of github.com:bmmalone/pyllars into dev commit 85885c513184f8e2661ea9820bcf2e55d4dd1340 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 13 10:24:40 2019 +0100 ADD helper to convert sparse matrix to list of sparse row vectors The "sparse row vectors" are really just sparse matrices with shape (1,num_cols). commit c4027f71fbff9280774e8b8f6f6f1df54299b386 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 11 21:06:15 2019 +0100 DOC installation instructions in readme commit 844ada20fa5fc0f8dd79a70158b2175474123389 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 11 20:52:05 2019 +0100 MNT update to version 1.0.0 commit 8259d344acb5a27f355d20cb3514d538ee66cb94 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 11 17:21:50 2019 +0100 MNT ignore dist directory in git commit 8ad60cc281fae84a2ecbfbee7ea79af0ddc7c12e Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 11 17:16:20 2019 +0100 UPD travis settings to include both dev and master branches commit 676c2c00f1c40f9426b0e93e6d57a4c43fdd7077 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 20:59:57 2019 +0100 UPD branches in readme commit 8da79b33b2329451d22fe13b8164dd52b975cf23 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 20:20:22 2019 +0100 MNT merge for 0.99.1 Squashed commit of the following: commit 22ab008d6758f5a24d2457dffe891acf02613498 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 19:58:29 2019 +0100 MNT merge dev into master for 0.3.0 Squashed commit of the following: commit 0f13237fdc08a4b4454d107b2caeadec6c4d0241 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 18:58:45 2019 +0100 FIX typos in hp_utils commit d0f93840157639d8035a3d197b7eb1b436715d98 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 18:37:07 2019 +0100 DOC api for stats_utils commit e229fef10bbbe47c9a89af9f5df890554f60a20d Merge: 4037575 6dd4700 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 9 09:33:27 2019 +0100 Merge branch 'dev' of github.com:bmmalone/pyllars into dev commit 403757574accb353a732686e06ef2a9f571d0c00 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 9 09:31:49 2019 +0100 FIX missing import in matrix_utils commit 6dd47009f6dcd8b0887f0badeb4f22abe4ae8100 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Feb 1 00:08:39 2019 +0100 ADD [WIP] high-level hyperparameter helpers commit a57202c4acbcf1568c9e73de9fd59292fad79090 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 29 12:37:32 2019 +0100 UPD ml_utils.eval_hps to allow predict_proba commit 78d5bb761c97810bfe7798f55d0638686a80c345 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 29 12:37:10 2019 +0100 FIX missing imports in matrix_utils commit e3e11d4ba8d3fce695853dfa1deabaf8c6f24c24 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 27 18:10:56 2019 +0100 ADD [WIP] ml notebook commit 1ad13bef6e3486a5d2f3dab1aa2ecd367b63aa31 Merge: 6fbb4ac b794be5 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 27 18:09:51 2019 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 6fbb4ac131c4b3ebfd326e4e9f19ec9827b24cfd Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 27 18:07:13 2019 +0100 ADD ml training helper commit b794be536d5422c0f2b517ab3079300315d82523 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 24 22:48:09 2019 +0100 UPD mygene_utils to new module structure commit c8cba699e44801465071e2c467d5411a6ea376f2 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 24 22:47:33 2019 +0100 FIX renamed parameter for validating sequences commit ec389f516d4105a31d30862c293efff41c7a2340 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 21 20:14:40 2019 +0100 ADD data frame filtering helper commit 4cf48ee0eaf107f221f9a98eaa7f6ac98fa8f676 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:41:56 2019 +0100 DOC coll_utils.wrap_in_set commit dc70396f6940bb205abf7f8b43c8325b925da9b4 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:34:57 2019 +0100 UPD ml_utils.get_fold_data to allow fields_to_ignore commit f810d18fed7a3e1651e06b35fa70861da59c117d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:34:15 2019 +0100 FIX missing import in validation_utils commit c5fbc1dec5a360eada64b74ba6404795904883c9 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:33:26 2019 +0100 FIX coll_utils.wrap_in_set to handle strings commit ab4cd00967dce800064bbbc7847ec479cd116e54 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:17:38 2019 +0100 UPD ml_utils.get_cv_folds to handle non-set inputs In particular, it automatically wraps the train, validation, and testing inputs if they are not compatible with `isin`. commit 278c563edef2af56b45d9624a1f41f69f0376ab9 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:01:45 2019 +0100 ADD helpers to wrap objects in a set commit 36d3c77d9d94e78e8ece7279e823af876e88693a Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 1 16:13:16 2019 +0100 ADD [WIP] tests for ml_utils commit 4d04ab0adf4a10420158e68c495bee6d2d65cc43 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 1 15:29:52 2019 +0100 DOC ml_utils module commit 3f69b4974975159aa3821b304e52f4b7e90ddef2 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 18:11:53 2018 +0100 ADD [WIP] pandas_utils tests commit 36f995e8d99cb8e0d0b622eae52af6c21f0231a1 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 17:50:07 2018 +0100 DOC pandas_utils commit 9aa9f5567995cbfbd73f6256c7df13416792b12e Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 12:46:54 2018 +0100 MNT [WIP] renamed import from misc to pyllars commit 1e3c58a0854960bdda00d958c47132ccf1d8ddf3 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 11:59:58 2018 +0100 MNT badges links in readme, test dependency in setup commit 07208c68c5be787d372fd7f0ecbb8fdeb0d773ce Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 11:25:20 2018 +0100 DEL slurm utilities commit 76fa8131e7168446bd9407b5f83dbb6041988a17 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 30 10:18:15 2018 +0100 DOC small content changes commit 12defa2ae0618d680e4246668b8afe2e75f27c33 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 30 10:17:52 2018 +0100 MNT bump to version 1.0.0 commit 3c77921defa2bffbc4623bed3cd4a23110e3d5be Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 29 15:34:37 2018 +0100 MNT rename package to pyllars commit 7e46364ff8670b3b58c61a85c4889c59cf45f594 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 11:30:35 2018 +0100 ADD read-the-docs config file [skip ci] commit 58eb05659578cbd18818d596bd674843912fcd88 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 11:19:18 2018 +0100 FIX [WIP] setup.py to work with read-the-docs commit b4d6485f819b65225b963494101ae080bd5314cb Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 09:58:45 2018 +0100 UPD [WIP] test coverage configuration commit 9b052db0aadc90c6bd11dbc55791488e0eab17b1 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 01:10:21 2018 +0100 DOC code coverage on readme [scip ci] commit 545473b07033f282ff9570b05f0498a98a625497 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 01:02:30 2018 +0100 FIX [WIP] setup.cfg for test coverage commit 1ce1705debed8db0a6158765350b1bdcfb5cb549 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 00:40:01 2018 +0100 FIX deprecated section in setup.cfg commit dc7457b6da524b8b11a3af9ec9daa8a8cbf8a581 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 00:25:01 2018 +0100 ADD coverage configuration commit 7abd6383f0975fbdf1c8d1d369d6600c1c615153 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 23:31:12 2018 +0100 ADD travis ci badge to readme commit 7546c14f7efa036a21a20757b0738207ce53996a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 23:24:54 2018 +0100 ADD travis ci config commit cc2c9c63367bec58da5a4bd5fa86cb489dd21153 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 23:19:34 2018 +0100 ADD simple testing infrastructure commit 3b26892f5b19f41bf426fea51c57dca40febb3f1 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 20:46:45 2018 +0100 DOC create basic docs structure commit 9d6bd43b325c494ec0930207ef3c158914074a25 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 20:45:53 2018 +0100 DOC prepare collection, dask utils for sphinx commit 8b71175760859c1e485150a6482cc30cd9ad4410 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 20:05:54 2018 +0100 ADD [WIP] sphinx docs commit bdc87d7c6e488d8e56bc94bcf9a4a3f54e0caecd Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 20:02:26 2018 +0100 DOC update modules to work with sphinx commit 7da1c0d2bd4e7c263979d9b83db18295bf68ac20 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 18:22:24 2018 +0100 MNT remove missingdata module for transformers commit c04b1555c1ba7ee28f5bf145c614162bb77a1d12 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 17:16:32 2018 +0100 FIX typo from setup commit 0a8043ca94d585aad83c09ec68845733aaf10af5 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 16:39:24 2018 +0100 MNT bump internal versions to 0.3.0 commit 1e7ba087b6415f8a577cf728250065e6ee2f8304 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 12:16:31 2018 +0100 MNT split math_utils into more specific modules commit 772b010161db0bda4aad30af2ed3fccb5af04d3e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 12:14:21 2018 +0100 MNT fastparquet is optional dependency It seems to sometimes have version dependency problems, so it is now only installed when specifically asked. commit 806c66beefaa93993b4210ba8b8c404f02874246 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 25 22:30:20 2018 +0100 MNT split general `utils` into more specific modules commit 174e929c516d49ed41283495b9577aba62ba3674 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 25 13:07:59 2018 +0100 UPD [WIP] readme with new folder structure commit 63f14dc4e6f22cea8e18b49657d192093b98b04d Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 25 12:57:07 2018 +0100 MNT __init__.py files for sklearn transformers and slurm_utils commit d8e4937e672fd7a8a306217433981cedf3c565e6 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 25 12:52:58 2018 +0100 MNT file structure changes commit 9167db5d820343884586a120575221357f88334f Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 22 12:18:41 2018 +0100 MNT update version info for 0.2.11 commit 56248d292ea5bba56ee6ea58223a8df6832b5c28 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 22 12:09:10 2018 +0100 ADD separate stats module commit 28df573e5ac5cf54f50016fede86d119065c5118 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 22 12:07:43 2018 +0100 ADD simple equal-aspect scatter plot helper commit b9ee654964581b48e8bb123239d19cd3de44e0da Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 22 12:00:49 2018 +0100 ADD scaler creation from means, stds commit 8cf38aeae6c32911c80bf6ad382f3282f6ddad18 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 17 08:10:53 2018 +0100 ADD separate module for ml utilities commit 19f504cfc02c47551cd8810fe1b0728a1ee3d7ae Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 17 08:09:11 2018 +0100 ADD helpers for extracting GO hierarchies commit 121d6b2706c08b3824ea8cef5d1ce5f385a1c634 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 2 12:34:10 2018 +0100 ADD gene ontology helpers commit 220b5ae94feadc693d150c836ffe8535d0381877 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 2 10:12:08 2018 +0100 ADD helper for finding many pairwise set intersections commit 51a7b4d6543176b66afd2f9e09afbaaf3e1f8e02 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Nov 30 00:58:15 2018 +0100 ADD mygene helpers commit eef08baa038512acdaa26cd4fa70c49b0af2c089 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Nov 30 00:56:58 2018 +0100 UPD utils to include wrap objects in a list commit e3b01e9e7724b9ea1069c207ecb2c28002c3a682 Merge: 167fd85 4a6993e Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Nov 11 23:24:48 2018 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 167fd85c292fabc21c4c4f9174be7534979c8896 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Nov 11 23:24:13 2018 +0100 UPD hide_tick_labels with axis parameter commit 4a6993e8f8696f8e77e862699ad40fedd4eb84fd Merge: 96161c5 eede55b Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Nov 5 16:53:36 2018 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 96161c5e446330371405672348bfb89957db855a Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Nov 5 16:52:04 2018 +0100 UPD plot sorted helper to optionally use cumulative density commit eede55b7a89067b2c5d68eda352a6891d649d9ab Merge: 9ca01aa a7c355c Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 19 12:05:40 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 9ca01aa37fc24221343ba1b1dbd5f401ce73fa92 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 19 12:05:07 2018 +0200 UPD transparent file opening for compressed files commit a7c355cdc0078840a518359d69bd867f3d685cfa Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 17 23:21:27 2018 +0200 DOC df_to_dict docstring commit 4feb49ba2afbb64fe22d3d60048e951f6bc06500 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 6 19:34:13 2018 +0200 ADD simple scatter plot helper commit 2a163f930ad95d16408f3225c8be2cf2d7332f2b Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 18:11:23 2018 +0200 FIX circular imports in utils commit 3ef4356aadc5aaef528d591d061897f0723e775e Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:47:20 2018 +0200 DEL load_config from utils This function needs to use some functionality from validation_utils, so it cannot be included in the base utils.py module. commit 41a9347455234d86988408ca08bbdc6f13856ea6 Merge: 0a86f5d dcf2ffb Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:22:40 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 0a86f5d210a29dc951bfc8f9b9ffb2781284cf6c Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:20:38 2018 +0200 FIX missing imports from utils commit 2823876f2d754193f0e05a603a54951a9125f9c4 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:19:57 2018 +0200 UPD creating bar charts in mpl_utils commit 06ab3ce7527747bfef380a34d04cc3a3b6195c34 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:17:21 2018 +0200 UPD yaml config loader commit dcf2ffb0c37c2c0d78ae8de3761f08c20eab389f Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 16 11:17:52 2018 +0200 UPD `compress` keyword for writing data frames commit c1ec0e4301bdca4483768ca37a62f67def3c64dd Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 15 13:07:00 2018 +0200 UPD dask_utils to accept priorities for jobs commit 7afa9d8cc95913b74fd6f2470fbb7b19cc722a6e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 15 13:06:26 2018 +0200 UPD specify bar chart tick offsets in mpl_utils commit f198ef2cfe1934d3979ef8522df358eb3d566144 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 15 13:05:50 2018 +0200 FIX numpy import in function body commit b4d1c740f78b5a495f2665148ab16a33c3f4b6e8 Merge: 1eda060 56ff5a2 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 9 12:39:27 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 1eda060e336dce01c8b16088d5d2111227a3c1d4 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 9 12:39:13 2018 +0200 UPD scip output parsing commit 56ff5a29eb275a6dc4863dd28b6bf82fc8e6cd97 Merge: 2ca6d92 694ed19 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jul 19 18:37:55 2018 +0200 UPD mpl_utils to plot simple sorted lists Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 2ca6d92ffa57bfe6ffbe82ba11cfb12c6d45a98e Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jul 19 18:23:22 2018 +0200 FIX typos in num_bow_union commit 694ed199678cb392efa44d728cea8fc23d564b74 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jul 9 21:58:48 2018 +0200 FIX typos and missing import in SimpleNumBowUnion commit 72c005f7f60d955a200131a47bcfce045d92fc59 Merge: 157fdba 41b0c2d Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jul 8 18:15:40 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 157fdba4f9daee4b296415bf5ce30243ef6dc952 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jul 8 18:15:15 2018 +0200 ADD [WIP] simple handler for BoW and numeric data commit 41b0c2dfea03b520bca72d00ad4024e42ca8ef3e Merge: 903dbc6 812fae8 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jul 5 22:00:27 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 903dbc65baf61a58383c192cce34774aa641947a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jul 5 21:30:42 2018 +0200 FIX missing import in utils commit 812fae88d69224e2fba9670d8951604a9182a5ba Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jun 28 12:01:27 2018 +0200 UPD cancel helper in dask_utils commit ee51abb9cfe21cd8c499b7a47f4804157141c654 Merge: 0b2d9a3 241cabb Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jun 28 10:34:57 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev Conflicts: misc/validation_utils.py commit 0b2d9a324e9033339d8623b89bbaf5b7834094bc Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jun 28 10:29:42 2018 +0200 UPD k-fold splitter to include validation set commit baadb50deab68f9b5c7639e81a26b630e8bec295 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 14:45:55 2018 +0200 DOC changelog commit fc2f1e818e79bcc963bc249e820ee93686e28c3e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 14:44:52 2018 +0200 UPD cinc gender and icu mappings to include reverse commit f3fcecfd7efea8dee23ef671e46a2541d041172f Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 14:44:32 2018 +0200 FIX nan_ohe to handle sparse matrices commit 4c70d655cd856c5924072790a565bfe202a1b65b Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 13:31:35 2018 +0200 UPD changelog commit d53bb3f43d20703e007205af19d73e56bc227e1e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 13:30:42 2018 +0200 UPD validation_utils to include non-pydata helpers commit d5c6aef891b0a35259cd4654983fbaf80d6f64af Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 12:41:17 2018 +0200 UPD utils to include reverse_dict helper commit 82d405a5a28ac3dd9e3f537602e0d8c21ab7124a Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 12:40:55 2018 +0200 UPD cinc-2012 field constants commit 241cabb75b42d534d8fee148c49612906e715a6a Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jun 19 13:42:42 2018 +0200 UPD validation_utils to include more general validators commit 1fb81fa9864d2406d195e691c2c5366469bd4017 Merge: 95d75c4 d47b4fd Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jun 18 16:07:04 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 95d75c421b53a70e614c859a7c5cd8774c718a38 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jun 18 16:06:39 2018 +0200 MNT merge from master commit d47b4fdd9a5dd07289c51a67b9bfc5de7ce60a26 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jun 8 00:33:12 2018 +0200 UPD helper to collect dask futures commit c6978f8b26f4a609f524295a852e7a839bc5a414 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 6 12:23:16 2018 +0200 MNT update version to 0.2.10 commit 6c9bf7c4758e2aa82b1a3b2804e15a123b0c47b8 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jun 1 17:01:35 2018 +0200 UPD a few numeric validation utils to handle sparse matrices commit debaf0ab2822ed8f526ef64628f837b5a0e3d7c7 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 30 19:45:22 2018 +0200 DOC updated changelog commit a5c2502aa66955a889e552a3d87cf1478ceac4ca Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 30 19:43:47 2018 +0200 UPD physionet utils to work with mimic waveforms commit 3f787f20ea278a069488c7fdd23c023db1eee230 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 30 19:43:18 2018 +0200 UPD check_status helper for dask futures lists commit 8e5722f7a87585a7190d83db03ab0fd4fa1bb553 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 30 19:42:34 2018 +0200 FIX parse_scip_output when scip crashed commit 8f56e03203809478a9d2e83872241c1eacd15fe6 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat May 12 15:27:45 2018 +0200 ADD followup table construction for mimic commit a3eb613edcbeca1a8e5f7da505a10346fcbbafd6 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu May 10 00:27:30 2018 +0200 ADD cinc-2012 time series names to physionet_utils These are fixed and will not change since the dataset cannot be linked back to MIMIC. So they can be treated as constants. commit e4b7e73d4096a0b4b83561db1c4aa83213dd8f63 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu May 10 00:27:07 2018 +0200 FIX typos in deprecated message commit c9ff1f48a4b52632857bc6ac662061cd31402cae Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 9 13:21:22 2018 +0200 MNT notes in changelog commit 19a18cbee58a677d2d83b20d0410c78f652b424b Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 9 13:20:18 2018 +0200 UPD ds_man to optionally encode target variable commit a32cac7569d285606abda11005dfa5047c4ecb36 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 9 13:19:23 2018 +0200 FIX scip_utils to use dicttoolz for merging This removes a deprecated warning message. commit ea3b707d0384e4d1bfebba4e81db44f89f3ae22a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Apr 19 19:33:44 2018 +0200 UPD debug output for nan_le commit 7e67f6dff5e847a6a276ac38377f87a72c12b038 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Apr 19 03:56:39 2018 +0200 UPD validation to handle ragged arrays in int check commit 72a850471983332f81d6a907bfe781b8b51cd0b4 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Apr 19 01:46:50 2018 +0200 FIX missing indices in ds_man.get_fold commit 9f55a5688571be0bf3615f7ffaee41fcaf2edef5 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 18 22:20:38 2018 +0200 MNT bump version info to 0.2.9 commit d2f8a10935e78125d33af6df53c96f3a540e0fae Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 18 16:24:21 2018 +0200 FIX missing sklearn_pandas dependency commit 8e533dd267266ef1c1cf71af4823a2f73d856074 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 18 16:17:46 2018 +0200 FIX clone command for users without github accounts commit 38b959b0e7b25d123e7648de4f8722df21d92811 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 16 22:31:26 2018 +0200 UPD nan_le to avoid printing very long debug messages commit 6c3ec25491f1a1978488548b5555ac18e9576c43 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 16 17:42:49 2018 +0200 UPD get_kth_fold to include indices of training, testing sets commit 3980beab682ca8e2a08a367e3d72c514a43bbdf6 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 16 03:11:40 2018 +0200 UPD sklearn metric calculators to handle multiclass, regression commit 77f0493ecc7eee71314b6e03fe08a00c8bdfe14c Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 19:29:05 2018 +0200 FIX missing imports in validation_utils commit 5c818ec4dc8eaf45fd4e48eaf624aea602bb1ab7 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 19:27:12 2018 +0200 ADD validation_utils module commit 25c33ceff0ce75d070eb478e7605f639e5561c51 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 17:46:49 2018 +0200 DEP dict helpers They now suggest using toolz.dicttolz functions. commit 09959b9f211e8daa7ef771fcd500ed840ce7762c Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 15:54:41 2018 +0200 UPD pd_utils to include apply_group helper commit 480f598d8ec6b9cac24e79c416732b3dc4567842 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 14:34:29 2018 +0200 UPD split_df to accept chunk_size instead of num_groups commit 6366d7a9fd0d120beaefdd6ca16f6b63b811ac27 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Apr 14 13:08:41 2018 +0200 UPD replace_nans to treat np.inf as nan commit ae414752ad23b63ff59a84d011d4922d0b4dd661 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Apr 14 13:08:11 2018 +0200 FIX missing import for pd_utils commit 9ace0ec393f05a0d0d3b9a60f4a82f96caed0354 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Apr 14 13:07:05 2018 +0200 UPD cat_mle to handle base-1 observations commit 177f9655863079848bac561f5dd94f770ccf6b4c Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Apr 14 13:06:24 2018 +0200 UPD ds_manager to handle inf's commit c596f907f94b8d565942b59902ae5d3aa63a0673 Merge: b521145 d76faef Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 4 22:49:31 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit b5211452ca6d6894674e603e4db0049bdb6cfd97 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 4 22:49:00 2018 +0200 UPD math_utils.check_range to return whether the value was in the range commit d76faef41cfb513b9d0739784b50e9aff88388ac Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Apr 3 18:19:17 2018 +0200 ADD check_is_fitted helper commit 5a1ad0de8bd63196b24f7d35dcc3c8212cbff5aa Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Mar 21 11:12:25 2018 +0100 ADD utilities for parsing, etc., SCIP output files commit ecbc2efad6bfbe1667dee7e2d00819ac4a39ae2c Merge: a542cac a02b8a2 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 19 18:34:41 2018 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit a542cacd250aa4078476d49e4094633093afc81b Merge: cc5ffe6 f47f0da Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 19 18:34:02 2018 +0100 UPD md_utils to handle ds_mgr without categoricals Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev Conflicts: setup.py commit a02b8a2dbc2a1c96a774b9115d4ee32dbfdc3da5 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Mar 18 21:39:18 2018 +0100 ADD helper for categorical MLE calculation commit cc5ffe6d2767ec9148556132da6ff5585ae1fd9c Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Mar 14 01:43:40 2018 +0100 UPD simply apply helper for pandas data frames commit f0565bac888acfdc2c63ff455530a7f88e189caf Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Mar 14 01:42:43 2018 +0100 FIX missing sklearn_pandas dependency commit f47f0da44bda88ce434e735defe3849b403dbf40 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Mar 13 00:05:30 2018 +0100 MNT download small nltk resources during setup commit 28af5572169e9b64764e5505562841c39f0428a1 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 22:35:30 2018 +0100 DOC use docstrings for all module descriptions commit c2433fc300a136c93e8b4434c1a5b99ef1a19407 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 22:15:35 2018 +0100 DEL spurious automlutils import in missing_data_utils commit 6c1ad9781dea19e7167bbfe36a327c35d21be3a6 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 22:14:12 2018 +0100 DEL partial class The use for this class is extremely niche. Also, the partial class cannot be pickled, so that typically does not make this a very useful thing to do. commit 78e79b39f530caf9e45818be6d25515a8fa32d7a Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 21:53:02 2018 +0100 DEL automl_utils They are now in their own package, so no need to add the dependencies here. commit cf7a39b9d587cbcf3d379435d4cde038562cbe33 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 21:52:29 2018 +0100 DOC added unsupervised cv to changelog commit e64fce50acb34e0f876e426fd51ed2df7dfe746c Merge: 7779ee4 cf7a39b Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 21:44:11 2018 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 7779ee41cb810e5e7a2674668936ef03f216e267 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 21:41:28 2018 +0100 DEL mysql helpers These helpers add another large dependency that can be difficult to install correctly. Thus, they are likely to cause problems and the logic is generally handled by higher-level libraries (say, flask) anyway. commit 92b22f05f1dafe253b781426ebd2e676e4fb2995 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 17:45:13 2018 +0100 UPD k^th fold util to handle unsupervised data commit b0b0bc71bfd8c0cffc1785859fc4fd28f05ebc08 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Mar 8 20:38:03 2018 +0100 DEL pickle-stan and pystan dependencies commit dde8764fbf17f5784e08bdc441ac1cbc829bd900 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Mar 3 13:01:22 2018 +0100 FMT changed logger level for `ensure_path_to_file_exists` commit f28347d7a53577758a1695b5bbe665804283e8e3 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Mar 3 12:08:19 2018 +0100 UPD pd_utils with helper to create chunks of groups commit 0b7945edaef5a65b55983d7a8bfdb361886fd123 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Mar 2 19:48:42 2018 +0100 MNT remove executable permissions from files commit 6e93b222e2c81d22c7524abd8cd4073884df4339 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Mar 2 19:13:57 2018 +0100 MNT bump version references to 0.2.6 commit 718ff9e2c7a68b1d575acfd3bb6f1866b7798104 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 28 13:48:52 2018 +0100 FIX dataset_manager to handle missing target and dropping commit e5cb6e7ea108fa42555739a4e4d2aa5138f23aab Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 7 19:52:10 2018 +0100 ADD incremental_count_vectorizer commit 6a1c4590cf3799e24175b1729052540b5171eee2 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Feb 2 13:27:33 2018 +0100 UPD utils with merge_sets commit a0d869b5dff10c4867dc60900db0978d553452a0 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Feb 1 19:07:20 2018 +0100 MNT updated changelog for nlp_utils commit 7b1b76dbdc75de8bb2952cdcb42983a6d311db0a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Feb 1 19:06:27 2018 +0100 ADD nlp_utils commit 3c3a951da1f8c6d8a8140bd181de2c53b8ebb71a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Feb 1 17:51:35 2018 +0100 FIX progress bar in dask_utils.apply_df commit f1950f2ff8f787df12149cdbdb1b46ca05585dca Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 30 15:41:00 2018 +0100 UPD utils.open to accept args, kwargs commit 90ec5a4389386d989c5b8cde21233d516f1aa2a1 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 29 15:23:54 2018 +0100 ADD dask helper for groupby results commit ce6a5183fd5a738eb6634dbbdb065c02227e752f Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jan 24 00:26:22 2018 +0100 ADD dask_pipeline helper for submitting sklearn pipelines to dask commit 93cb8285a2fcad1a6d7c2dc9b0fcad418ec04b5b Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 23 12:43:13 2018 +0100 UPD optional fields_to_ignore for dataset_manager commit f00dcff577029c66ff22fa49bf57362686ad9f48 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 23 12:42:37 2018 +0100 ADD helper to create missing data preprocessing pipeline for sklearn commit e1506ac8a6d4bb53a0c3e4d9910e7b0208d737e3 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 18 17:06:04 2018 +0100 FIX nan_le incorrectly converting floats to strings In particular, this was a problem in the `fit` function, and the `classes_` dictionary would not have the correct type of keys. commit 85b3dc6c3c2959521a464004dbba5616de85e801 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 14 22:13:53 2018 +0100 UPD physionet utils to read more tables commit 3a5ce7899eedb80be29a7e15cbe63a7ffa2bb469 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 14 22:13:12 2018 +0100 UPD nan_encoder to use dictionary for encoding commit f32d2d5f4641500ab643f64cf0bdf49cb1b3199f Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 14 22:12:17 2018 +0100 UPD `dask_utils` to have `apply` helper for iterators commit 10dce10879b1ef4e2dcc2e945af054c2107736e2 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 14 14:50:22 2018 +0100 ADD `apply` helpers in `dask_utils` commit 93ae10eb80e9030bc42b73eaaf4a1d930824f716 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Jan 13 22:07:27 2018 +0100 UPD physionet helpers commit 033e9498ea2a1f91ee4fb8b7cc38f03bad761755 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Jan 13 16:13:51 2018 +0100 FIX missing logger in physionet_utils commit eb7881838643052e6a29e5b809c5b58bf80339db Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 11 00:06:29 2018 +0100 FIX nan_encoder to handle object dtypes commit 320f463e6f875ea33458282983588920c45c941e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jan 10 16:11:13 2018 +0100 UPD nan_encoder to handle unknown labels In particular, if asked to transform an unknown label and the `treat_unknown_as_missing` flag is `True`, then the encoder will replace the unknown label with `np.nan`. commit f8dafc2ab78eba3a1db2db2027725c80adf5c297 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 9 18:17:47 2018 +0100 UPD dataset_manager to handle non-numeric targets commit d517217503e595780f30deb59e1efc19e05f1b5e Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 9 10:51:54 2018 +0100 FMT remove print from nan_scaler commit 0400b80f674c70e58dbff68d6d5974802392dd19 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 8 19:15:28 2018 +0100 FIX nan_scaler to work with 1-d np.ndarrays Previously, it would work with np.arrays (which are inherently 1-d) but not ndarrays which happen to only have 1-d. commit 47b097eb884b6a0ea548f55d9db1cf0181139bd1 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 8 17:15:15 2018 +0100 UPD nan_scaler to handle 1-d arrays in `fit` commit 28290f767508e7c8744702686791b10c2dd47005 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 8 17:08:47 2018 +0100 FMT logging output for nan_label_encoder commit 413ce18ee05233988db3f5e705cffd62a07f4fe3 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jan 5 04:05:12 2018 +0100 FIX type in nan_le triggered by corner case commit ea7d85222917e1c9918ce062bddf7d97bd1c853b Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jan 5 03:36:44 2018 +0100 UPD nan_ohe to handle 1-d np.arrays commit fb7eebb1fd80f48442a7923e42af91fbb5913f57 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jan 5 03:36:15 2018 +0100 FIX indices of encoded labels for nan_le commit fccc33095dee93254cce1d9e7b6426870ab91930 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 13 18:55:24 2017 +0100 FIX nan_nn to point to new missingdata submodule commit da99f0f1ce960fec905350cd355a3bb037849569 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 12 22:35:48 2017 +0100 FIX nan_one_hot_enc to handle no categorical variables In this case, the one-hot encoder simply returns the original array unchanged. This is consistent with the basic sklearn implementation. commit 9311e56e742deaca60a81322c44b3c62c7ed78c9 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 7 03:15:11 2017 +0100 FIX typo in import for new missingdata subpackage commit cafe94a8b11715574e47eaada3d5256dadf370e1 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 7 03:14:41 2017 +0100 UPD category counts in dataset_manager commit a7f6d5ee1bd71b5bce4423807908895f28289662 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 6 22:38:32 2017 +0100 MNT moved missing data utilities to subpackage commit 8e945259c79311fe51e6e4f751300e5c94b0f7b7 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 6 22:20:15 2017 +0100 UPD one-hot encoder for missing data This implementation is now complete. commit 8932b4c62fefa18807584a0be586be3b70b1615b Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 6 20:27:18 2017 +0100 ADD [WIP] one-hot encoder that handles missing data commit deacf9b69f75820d6a77afa5af51fadd335ee9c1 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 6 02:29:23 2017 +0100 UPD ds_manager with various helpers These mostly ease extracting information about the categories of the categorical fields. commit d7d5330ceeb550e742f67e10c90e25091613c4fb Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 5 23:54:17 2017 +0100 FIX unnecessary dependency in multicolumn_imputer commit e2f9c526eae9f18913491834196efe069efc31b6 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 4 01:19:04 2017 +0100 ADD helpers to handle encoding labels with nans commit c64e197323a31be5e1abcb9cd30c3c6fc857aaed Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 18:20:54 2017 +0100 UPD several changes to ease working with mixed datasets commit a1c67e8b4ef345cadf8920349296d93d2ac7a1d0 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 15:16:37 2017 +0100 ADD multicolumn categorical imputer This class replaces missing values from categorical columns with the mode from the respective column. commit 123cc0fbf372adf0018ad1fa9124e794c64bdd7e Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 15:16:15 2017 +0100 DOC add column docs for multicolumn_label_encoder commit 9c6ead261bd53c8f31b8f614da0cb1e55c3764f9 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 15:15:18 2017 +0100 UPD ds manager to include field type helpers commit 3cb22298837aba8b5cd198a13a6fa26a98d9f527 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 05:26:10 2017 +0100 FIX typo in get k^th fold commit 6dd308599d6b1df2371a72770751a03da0b5f780 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 04:50:42 2017 +0100 ADD dataset manager commit 0c819ca34a8509ca59b4fa9e0b59d2ab76c6e07c Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 01:38:19 2017 +0100 UPD cv fold helper to include some error checking commit dc6a42bfa3e644be925b4e62340791d1cf4b3cb4 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 01:36:08 2017 +0100 ADD k-fold cv helper in math_utils commit 43b4f904872b2352b6c6882976ae4c2d41b12f35 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 01:35:33 2017 +0100 ADD class to wrap constructors with fixed arguments commit af197f8d0042edb05e04e22d08a71e105ac8cebe Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Nov 1 14:02:53 2017 +0100 ADD class to suppress pystan output commit ff9aa2c3bd2f329d1f765925331bfa59b1d7d4e1 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 27 00:25:01 2017 +0200 MNT update change log with fastparquet patch commit 278a3fcc25883c63f744695a9387fef49c76cec3 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 27 00:22:24 2017 +0200 FIX fastparquet imports in pd_utils This is a patch-fix for Issue #4. commit c4b40ba75794b554a3c058955e2e9814bd16be4f Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 26 22:06:51 2017 +0200 DEL old mimic_utils file commit fbc8a050f22b4c78d6a8898662a0bd6ce1827e7d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 26 22:06:23 2017 +0200 MNT merge with master before version bump commit 742bbd78b54375d89f0f88b920b030fcce395e4d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 26 21:56:37 2017 +0200 MNT versions in setup for networkx, pystan commit 10f4a5063d39eac6b693e0dfdbded4d66a558614 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 26 21:23:31 2017 +0200 MNT prepare for version 0.2.5 commit 06b71564093bf7dd2765e64a8d1bda3b03068ad5 Merge: 84c7fa9 627b84e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 25 14:02:08 2017 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 84c7fa9b5edb7d6a196e78f6f00d5e879a45f1f0 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 25 14:01:50 2017 +0200 FIX missing sklearn dependency for math_utils commit 627b84ee74c91b2f8bc312480510c2498042b38c Merge: b2c3751 facd5af Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 19 19:39:06 2017 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit b2c37517dddc468ca5af409a1d81e4c84de0f172 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 19 19:38:18 2017 +0200 FIX ensure classes always ints in multiclass auc functions in math_utils commit facd5af1fdbc1876344d3117f942afddbcf9e63e Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 19 02:42:31 2017 +0200 FIX missing more_itertools prereq commit 00180b7af4f02b2ec518467e86dfeedfefa42e14 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 19 00:53:43 2017 +0200 UPD added identity column options for cv splits commit 0de080703ed936cb2aa32134464850c2d0129bec Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 18 23:53:25 2017 +0200 FIX typo in nan scaler commit 15b98b070f7bd1548d5d7b1063a2f3c63a76c97f Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 18 20:07:04 2017 +0200 UPD utils to load more mimic tables commit 159287060c6fb6e1e102c7991a3dbbd8a3bdf36a Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 18 20:06:29 2017 +0200 UPD nan_scaler to work with pandas data frames Parts of the code look rather brittle, though. commit 53ccd8b59170886dae561047d2c5d29ef225b28d Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Oct 15 17:55:59 2017 +0200 FIX typo in missing data training helper commit a24338762a2280cf61dc3fa0c53a4416244d7b94 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Oct 15 14:43:37 2017 +0200 ADD helper for training and predicting with missing data commit f03209c1e42314db7eca477bcd8525af46b3f530 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Oct 14 15:37:38 2017 +0200 UPD readme to be in-sync with the current contents of the repo commit 2c07202da9883f74cfbd230ea64c55b31410edeb Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Oct 14 15:31:14 2017 +0200 MNT deprecated automl_utils commit 27d52375bed7a4c70a40c18f6c7b6eff40b245c0 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Oct 14 15:19:01 2017 +0200 ADD simple knn wrapper which handles np.nans commit e5eba04f9ab45d1fc7d183e838a1670d50e94182 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Oct 14 15:18:17 2017 +0200 ADD utils to remove data according to different missingness mechanisms commit 80650a8e93cadc4ace6d3304316803b776ca4c8b Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 13 20:41:29 2017 +0200 ADD nearest_neighbors which handles missing values commit 12cb576b8b946218de5fa5c228d8d9ef5f42d4ad Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 13 20:40:16 2017 +0200 UPD nan_scaler to handle 1D input for transform commit 8a32c85037ec4329ecba4560c73c97b232ca0e5d Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 13 20:39:29 2017 +0200 ADD distance metric helper for vectors with missing values commit 2680184b97fa0fce03716cdfa2f9a90c67e0d19e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 11 19:54:01 2017 +0200 ADD math_utils helper to check for nans commit 765f2cbdbe267e33672f01e4fb5b4f6bd23444a5 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Oct 10 18:35:56 2017 +0200 MNT add new updates to changelog commit d166e29dc341fe9f323498171e8f0ed30fe6c00f Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Oct 10 12:42:46 2017 +0200 UPD mpl_utils.plot_confusion_matrix to work on axis objects commit 359b2d654e5c417a7e2156ebe5192c328f20ede9 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Oct 10 12:40:11 2017 +0200 ADD helper to randomly remove values from np.arrays In particular, math_utils.mask_random_values uses an MCAR removal strategy to add missing values to a data matrix. commit ff88c23d4e047b30baf726b6775de44440bb7488 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 5 16:31:50 2017 +0200 ADD multicolumn label encoder commit 53ac2c1ec0913f1899fccbdf994958d5e07037a0 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 15 18:00:43 2017 +0200 ADD join df list helper in pd_utils commit 138636bbb62d48db09422dd117b10db57ea41e32 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 14 15:17:12 2017 +0200 UPD order of params for binary classification metrics to match sklearn commit b0f59b53d4ba556a01ad25db6606f50c1f7a52b6 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 17:13:47 2017 +0200 ADD multiclass auc from [Provost and Domingos, 2000] commit 57c3207979655e451cdf86b349269dd1cb67bce4 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 15:02:48 2017 +0200 ADD multiclass auc calculation from [Hand and Till, 2001] commit 88ee39a7523666386d6025a01569ab56aa843ef3 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 15:01:37 2017 +0200 ADD convenient font sizes for mpl_utils commit 36eaf0a2948e0b08ed7b23b1b9b87a484b019ac6 Merge: 8d4de7a 7df5e0f Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 11:14:44 2017 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 8d4de7a56f9f3bdcfa518778a21a7fb3fa8a6a45 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 11:13:25 2017 +0200 UPD classifier type-pretty name map commit 7df5e0f7b5f1c3fc67b4e1c9ab706a620a3bcf4a Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Sep 9 17:28:06 2017 +0200 ADD helper for drawing rectangles in mpl_utils commit f6721f006606fde808acd5ad451f055e24ab1153 Merge: c58c61a fc0f9b4 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Sep 9 14:05:39 2017 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit c58c61aca1ad46b3a0dbea22560a27b06e23f0e3 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Sep 9 14:05:20 2017 +0200 UPD sklearn type-name map commit fc0f9b4cd47859458199dc647f09a384a7fbaac4 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 17:40:01 2017 +0200 DEL kwargs from asl_wrapper member variables commit 4af7681eb318c8c8d832c54fd05651d74aa725b6 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 14:30:47 2017 +0200 UPD asl_wrapper to accept metric in constructor This addresses Issue #3. commit 843c6278f2b23fef75fa6ce0a518c2d55d1930f2 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 14:18:10 2017 +0200 FIX asl_wrapper to save its label encoder This addresses Issue #2. commit 7d19c061935836d82df129179a062beec5ba4ce0 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 14:08:39 2017 +0200 ADD asl_wrapper ensemble model summary helper commit 799617686a20ed5ae024470841b73a6b3a011a66 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 14:08:17 2017 +0200 FIX utils.get_types to handle unknown types commit f3d4ad0bfd51be4a69ffe99f2f4b3627307e379d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 18:50:17 2017 +0200 UPD cinc-2012 to use HADM_ID as the id column commit d71b77d42e9e7a7638c4046fd126bf24bfc83a64 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 16:43:53 2017 +0200 FIX parameter list in automl_utils predict method commit a961bf12b5616469e03f4859c1b07eca7be9ff31 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 16:43:08 2017 +0200 UPD loading CinC 2012 records to handle missing gender commit 611f3d4999d122501f532bf063dbf2ef1aa87227 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 16:40:49 2017 +0200 UPD index parameter to control writing parquet files In particular, fastparquet.write looks for "write_index", while pd.write_csv looks for "index". The pd_utils.write_df function now converts "index=False" to "write_index=False" for parquet files. commit 80b1298feb9a5fa4b3cce4d2d02fb3541ff8e234 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 16:40:22 2017 +0200 UPD dask utils to optionally restart client commit b9bff1db66b65c7ce1f601b536680a11eebcb474 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 13:15:21 2017 +0200 ADD joblib backend helper in dask_utils commit f22dcb9149dcc93f8a38a0c3617b0a8667c34b88 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 6 18:34:40 2017 +0200 UPD physionet tools to handle missing descriptors commit 15fe28698a30c8e389b57a86a547e08163218f8a Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 6 17:16:20 2017 +0200 ADD CinC 2012 helpers commit d6545a6862480c193f0615124d245d7cb43dd582 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 31 16:40:53 2017 +0200 FIX date for 0.2.4 in change log commit b8c0afef7e3d2c71f7943042c328a20f78732bf9 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 31 16:30:25 2017 +0200 MNT bump dev to version 0.2.4 commit 02a7246af1cc20765c271f2dd2be2301499b2134 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 31 16:28:14 2017 +0200 UPD readme to include descriptions of all modules and scripts commit 411b9263a3a5d98d6db8eea17387cfc64fc5560e Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 31 16:26:46 2017 +0200 DEL the visualize-roc script commit d6c0fcac24a8de25d4e904902299f3a87db53571 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 31 15:41:22 2017 +0200 DEL external_sparse_pickle_lis commit 79fcd5918534c6e1c32b20419afec2beb1b9987d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 31 15:40:52 2017 +0200 MNT deprecated column_selector commit b033199a9d399a5e638c7a198a5f1f1aa8c9be4e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 30 22:14:19 2017 +0200 FIX typo in pd_utils.write_df when ensuring directory exists commit ce9885d704171cfbd9cec2bceacc89ea044f73f6 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 30 21:34:22 2017 +0200 UPD subdirectory utils commit cf49dd024a788b40a767ad3dac4d810b86adc328 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 30 17:54:33 2017 +0200 UPD automl to handle sklearn and asl pipelines commit e01a325bb295f42ee7024a08bee0f9ef0880865a Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 30 16:11:01 2017 +0200 ADD retraining for asl_wrapper ensembles commit 6f33ae48e44200bcd29819ec0c3a2dbf240e33fa Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 30 05:33:06 2017 +0200 UPD blas command line options string helper commit f322767d4cf270cb7e415d740420e5ddcf3708f5 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Aug 29 18:13:18 2017 +0200 UPD grid search-like iterator utility commit f6e81b442922ba4122257c637919d86a37f5fefc Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Aug 29 17:08:45 2017 +0200 UPD option to load aslib scenario without name commit 6b54c852f99a5ac6bc42261935c9a64acd3738d9 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Aug 29 17:08:31 2017 +0200 UPD logging in nan_standard_scaler commit 36a9ead5a27315d377f2984878214052c4ab5b18 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Aug 29 02:00:44 2017 +0200 UPD aslib dependency helpers commit 7f5fc8537703bca9e790e010533a61eee91a5226 Author: Brandon Malone <b…
bmmalone
added a commit
that referenced
this issue
Jun 25, 2019
Squashed commit of the following: commit b8da88f759344532d285a1cfdc9879a3aa41bbec Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jun 25 18:01:54 2019 +0200 MNT bump for 1.0.2 commit 973feac95c165754b7f6f5c3c52157312869611b Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jun 25 17:57:18 2019 +0200 ADD helper to plot mean roc curve commit 7e9d718025491ecddea7054055356d23d8a8fc5c Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jun 25 17:55:08 2019 +0200 FIX missing import in validation_utils commit 53fce85c93448f43c36bec2afecea703c8036f55 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jun 25 17:50:44 2019 +0200 FIX logging_utils to properly quote extracted command line options commit b3ce83760bc29be9ec6fc6ba142b572192d09a62 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jun 25 17:49:24 2019 +0200 UPD binary classification metrics to include roc points commit e0b852676ee12f873e40d9c949b66966d897fa4f Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jun 25 17:48:14 2019 +0200 ADD command line options extraction for dask_utils The options are extracted as an array and can be passed to functions like subprocess.call. commit 5411bb54665ea22dac56e7ebfd3d4bd09b2351af Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jun 23 10:52:36 2019 +0200 DOC logging_utils commit fb9d602b8306bda03058bb7ea88927ee5e6b54a4 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Jun 22 15:03:33 2019 +0200 ADD binary prediction score plotting helper commit 33a94ecb2bc14c947b9fd2b7fc28da96d89fb736 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 19 16:13:51 2019 +0200 FIX tests to account for different seed commit a6eee65a45dfaecb0a85fdf8ca9531a7087b904b Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 19 15:51:11 2019 +0200 FIX typo in dask_utils commit eb5812001a6e5097e43b46c582e46d91e05b8166 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jun 16 21:36:30 2019 +0200 UPD ml tests commit 531f7f460dd91ddea710d178ce960fc7f85a6da1 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jun 16 21:07:10 2019 +0200 UPD typo in test commit 3d6b917a68cb4c1b9ab48921698c7a6364b5e9ca Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jun 16 20:59:50 2019 +0200 UPD a typo commit aca5097a4f76ae5502352d9e8e9077a7bc1a53dd Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jun 16 20:55:11 2019 +0200 UPD tests for cross validation commit c682c46de97f14f777c38ed3b1264a8e97d77abb Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jun 7 22:55:42 2019 +0200 ADD precision_at_k binary classification metric commit b353903fde018e6fab6a8a02e7a8fb984cbec18c Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 29 20:43:23 2019 +0200 FIX typo in fold_data fields commit dbf3e097cc44ecb88b0640e8a572cd5617a2b115 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu May 23 16:45:42 2019 +0200 FIX xgb validation kept a reference to best booster As a result, the reference would continue to be updated as more training rounds were performed. Thus, the best booster would be lost. A copy is now kept instead. commit 371fd1b424ec0b7cd79d381ebf444aef792549fd Author: Brandon Malone <bmmalone@gmail.com> Date: Thu May 23 16:43:54 2019 +0200 UPD plot_roc_curve to work with arrays of colors Previously, it used colormaps to determine the color for different points, lines, etc. In practcie, this was quite unwieldy. Also, the function now optionally draws only the lines. commit 3aa5b0173741fed399055a17b289b83c2d359e49 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:14:51 2019 +0200 MNT bump for versions 1.0.1 commit 6410d3f91c0bd07973ce43b51cb02bb1cdbac586 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:06:36 2019 +0200 DOC domain-specific docs page, initial physionet docs commit 84cd9c64639caa54b5fc849c0ca61a306e560456 Merge: e5cdf63 fd81736 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:03:17 2019 +0200 Merge branch 'dev' of github.com:bmmalone/pyllars into dev commit fd81736a773939c11ca1c2810bb1a0126d7f1bfd Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:02:54 2019 +0200 FIX missing imports from hp utils commit e5cdf63d819fd659c36d326bceadebb9400843a6 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 8 17:02:11 2019 +0200 UPD mimic file utils commit 33232694b28ce620aff2ccb7a82ee733a028ab79 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Mar 28 21:09:33 2019 +0100 FIX typos, missing import for multiclass metrics commit 7dd21fd18fceb706be083a2ee8be1163dbf61adc Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Mar 19 23:16:54 2019 +0100 ADD xgboost utilities commit 95ef3307a42aeb6a1bb869be1f9a131e1775030f Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 22:55:42 2019 +0100 DOC matrix_utils commit 68946578c27a8a90daedd60c3c4a8a7d72498276 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 21:55:08 2019 +0100 ADD nlp, stats tests commit 65c685b3f597b1ad330009eeea061d0059607ee6 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 21:54:20 2019 +0100 FIX univariate gaussian kl divergence This addresses Issue #15. commit 44b1570e4cab2a70da1a8c859286e49672e36a44 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 18:32:47 2019 +0100 FIX docs, based on version of sphinx on read-the-docs commit 23daf4a8c6627f567d241bb8799e48f3c106e077 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 16:40:20 2019 +0100 DOC nlp_utils commit faa903a55f6d503bce7e6beedf7ca8d3f2f76d3a Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 23 16:13:07 2019 +0100 DOC mpl_utils commit e366e61e54f62f5c1538dd9836fec9aea330ec4b Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 18 17:23:11 2019 +0100 DOC [WIP] mpl_utils documentation This commit also cleans up some of the mpl_utils code. commit f798a1deec8197046cecbc44c0908f614cf7a71b Merge: c4027f7 0d030d6 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 17 20:05:10 2019 +0100 Merge branch 'dev' of github.com:bmmalone/pyllars into dev commit 0d030d66a41f66814fe36b6b582bb2e355f7b0d7 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 13 11:29:12 2019 +0100 FIX missing dependency, requests commit d50db25ef5c03196f2959f7a6a57f711ca456a6c Merge: 85885c5 844ada2 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 13 10:25:37 2019 +0100 Merge branch 'dev' of github.com:bmmalone/pyllars into dev commit 85885c513184f8e2661ea9820bcf2e55d4dd1340 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 13 10:24:40 2019 +0100 ADD helper to convert sparse matrix to list of sparse row vectors The "sparse row vectors" are really just sparse matrices with shape (1,num_cols). commit c4027f71fbff9280774e8b8f6f6f1df54299b386 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 11 21:06:15 2019 +0100 DOC installation instructions in readme commit 844ada20fa5fc0f8dd79a70158b2175474123389 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 11 20:52:05 2019 +0100 MNT update to version 1.0.0 commit 8259d344acb5a27f355d20cb3514d538ee66cb94 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 11 17:21:50 2019 +0100 MNT ignore dist directory in git commit 8ad60cc281fae84a2ecbfbee7ea79af0ddc7c12e Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Feb 11 17:16:20 2019 +0100 UPD travis settings to include both dev and master branches commit 676c2c00f1c40f9426b0e93e6d57a4c43fdd7077 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 20:59:57 2019 +0100 UPD branches in readme commit 8da79b33b2329451d22fe13b8164dd52b975cf23 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 20:20:22 2019 +0100 MNT merge for 0.99.1 Squashed commit of the following: commit 22ab008d6758f5a24d2457dffe891acf02613498 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 19:58:29 2019 +0100 MNT merge dev into master for 0.3.0 Squashed commit of the following: commit 0f13237fdc08a4b4454d107b2caeadec6c4d0241 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 18:58:45 2019 +0100 FIX typos in hp_utils commit d0f93840157639d8035a3d197b7eb1b436715d98 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Feb 10 18:37:07 2019 +0100 DOC api for stats_utils commit e229fef10bbbe47c9a89af9f5df890554f60a20d Merge: 4037575 6dd4700 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 9 09:33:27 2019 +0100 Merge branch 'dev' of github.com:bmmalone/pyllars into dev commit 403757574accb353a732686e06ef2a9f571d0c00 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Feb 9 09:31:49 2019 +0100 FIX missing import in matrix_utils commit 6dd47009f6dcd8b0887f0badeb4f22abe4ae8100 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Feb 1 00:08:39 2019 +0100 ADD [WIP] high-level hyperparameter helpers commit a57202c4acbcf1568c9e73de9fd59292fad79090 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 29 12:37:32 2019 +0100 UPD ml_utils.eval_hps to allow predict_proba commit 78d5bb761c97810bfe7798f55d0638686a80c345 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 29 12:37:10 2019 +0100 FIX missing imports in matrix_utils commit e3e11d4ba8d3fce695853dfa1deabaf8c6f24c24 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 27 18:10:56 2019 +0100 ADD [WIP] ml notebook commit 1ad13bef6e3486a5d2f3dab1aa2ecd367b63aa31 Merge: 6fbb4ac b794be5 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 27 18:09:51 2019 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 6fbb4ac131c4b3ebfd326e4e9f19ec9827b24cfd Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 27 18:07:13 2019 +0100 ADD ml training helper commit b794be536d5422c0f2b517ab3079300315d82523 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 24 22:48:09 2019 +0100 UPD mygene_utils to new module structure commit c8cba699e44801465071e2c467d5411a6ea376f2 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 24 22:47:33 2019 +0100 FIX renamed parameter for validating sequences commit ec389f516d4105a31d30862c293efff41c7a2340 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 21 20:14:40 2019 +0100 ADD data frame filtering helper commit 4cf48ee0eaf107f221f9a98eaa7f6ac98fa8f676 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:41:56 2019 +0100 DOC coll_utils.wrap_in_set commit dc70396f6940bb205abf7f8b43c8325b925da9b4 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:34:57 2019 +0100 UPD ml_utils.get_fold_data to allow fields_to_ignore commit f810d18fed7a3e1651e06b35fa70861da59c117d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:34:15 2019 +0100 FIX missing import in validation_utils commit c5fbc1dec5a360eada64b74ba6404795904883c9 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:33:26 2019 +0100 FIX coll_utils.wrap_in_set to handle strings commit ab4cd00967dce800064bbbc7847ec479cd116e54 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:17:38 2019 +0100 UPD ml_utils.get_cv_folds to handle non-set inputs In particular, it automatically wraps the train, validation, and testing inputs if they are not compatible with `isin`. commit 278c563edef2af56b45d9624a1f41f69f0376ab9 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 10 00:01:45 2019 +0100 ADD helpers to wrap objects in a set commit 36d3c77d9d94e78e8ece7279e823af876e88693a Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 1 16:13:16 2019 +0100 ADD [WIP] tests for ml_utils commit 4d04ab0adf4a10420158e68c495bee6d2d65cc43 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 1 15:29:52 2019 +0100 DOC ml_utils module commit 3f69b4974975159aa3821b304e52f4b7e90ddef2 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 18:11:53 2018 +0100 ADD [WIP] pandas_utils tests commit 36f995e8d99cb8e0d0b622eae52af6c21f0231a1 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 17:50:07 2018 +0100 DOC pandas_utils commit 9aa9f5567995cbfbd73f6256c7df13416792b12e Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 12:46:54 2018 +0100 MNT [WIP] renamed import from misc to pyllars commit 1e3c58a0854960bdda00d958c47132ccf1d8ddf3 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 11:59:58 2018 +0100 MNT badges links in readme, test dependency in setup commit 07208c68c5be787d372fd7f0ecbb8fdeb0d773ce Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 31 11:25:20 2018 +0100 DEL slurm utilities commit 76fa8131e7168446bd9407b5f83dbb6041988a17 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 30 10:18:15 2018 +0100 DOC small content changes commit 12defa2ae0618d680e4246668b8afe2e75f27c33 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 30 10:17:52 2018 +0100 MNT bump to version 1.0.0 commit 3c77921defa2bffbc4623bed3cd4a23110e3d5be Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 29 15:34:37 2018 +0100 MNT rename package to pyllars commit 7e46364ff8670b3b58c61a85c4889c59cf45f594 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 11:30:35 2018 +0100 ADD read-the-docs config file [skip ci] commit 58eb05659578cbd18818d596bd674843912fcd88 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 11:19:18 2018 +0100 FIX [WIP] setup.py to work with read-the-docs commit b4d6485f819b65225b963494101ae080bd5314cb Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 09:58:45 2018 +0100 UPD [WIP] test coverage configuration commit 9b052db0aadc90c6bd11dbc55791488e0eab17b1 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 01:10:21 2018 +0100 DOC code coverage on readme [scip ci] commit 545473b07033f282ff9570b05f0498a98a625497 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 01:02:30 2018 +0100 FIX [WIP] setup.cfg for test coverage commit 1ce1705debed8db0a6158765350b1bdcfb5cb549 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 00:40:01 2018 +0100 FIX deprecated section in setup.cfg commit dc7457b6da524b8b11a3af9ec9daa8a8cbf8a581 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Dec 28 00:25:01 2018 +0100 ADD coverage configuration commit 7abd6383f0975fbdf1c8d1d369d6600c1c615153 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 23:31:12 2018 +0100 ADD travis ci badge to readme commit 7546c14f7efa036a21a20757b0738207ce53996a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 23:24:54 2018 +0100 ADD travis ci config commit cc2c9c63367bec58da5a4bd5fa86cb489dd21153 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 23:19:34 2018 +0100 ADD simple testing infrastructure commit 3b26892f5b19f41bf426fea51c57dca40febb3f1 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 20:46:45 2018 +0100 DOC create basic docs structure commit 9d6bd43b325c494ec0930207ef3c158914074a25 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 27 20:45:53 2018 +0100 DOC prepare collection, dask utils for sphinx commit 8b71175760859c1e485150a6482cc30cd9ad4410 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 20:05:54 2018 +0100 ADD [WIP] sphinx docs commit bdc87d7c6e488d8e56bc94bcf9a4a3f54e0caecd Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 20:02:26 2018 +0100 DOC update modules to work with sphinx commit 7da1c0d2bd4e7c263979d9b83db18295bf68ac20 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 18:22:24 2018 +0100 MNT remove missingdata module for transformers commit c04b1555c1ba7ee28f5bf145c614162bb77a1d12 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 17:16:32 2018 +0100 FIX typo from setup commit 0a8043ca94d585aad83c09ec68845733aaf10af5 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 16:39:24 2018 +0100 MNT bump internal versions to 0.3.0 commit 1e7ba087b6415f8a577cf728250065e6ee2f8304 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 12:16:31 2018 +0100 MNT split math_utils into more specific modules commit 772b010161db0bda4aad30af2ed3fccb5af04d3e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 26 12:14:21 2018 +0100 MNT fastparquet is optional dependency It seems to sometimes have version dependency problems, so it is now only installed when specifically asked. commit 806c66beefaa93993b4210ba8b8c404f02874246 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 25 22:30:20 2018 +0100 MNT split general `utils` into more specific modules commit 174e929c516d49ed41283495b9577aba62ba3674 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 25 13:07:59 2018 +0100 UPD [WIP] readme with new folder structure commit 63f14dc4e6f22cea8e18b49657d192093b98b04d Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 25 12:57:07 2018 +0100 MNT __init__.py files for sklearn transformers and slurm_utils commit d8e4937e672fd7a8a306217433981cedf3c565e6 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 25 12:52:58 2018 +0100 MNT file structure changes commit 9167db5d820343884586a120575221357f88334f Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 22 12:18:41 2018 +0100 MNT update version info for 0.2.11 commit 56248d292ea5bba56ee6ea58223a8df6832b5c28 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 22 12:09:10 2018 +0100 ADD separate stats module commit 28df573e5ac5cf54f50016fede86d119065c5118 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 22 12:07:43 2018 +0100 ADD simple equal-aspect scatter plot helper commit b9ee654964581b48e8bb123239d19cd3de44e0da Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Dec 22 12:00:49 2018 +0100 ADD scaler creation from means, stds commit 8cf38aeae6c32911c80bf6ad382f3282f6ddad18 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 17 08:10:53 2018 +0100 ADD separate module for ml utilities commit 19f504cfc02c47551cd8810fe1b0728a1ee3d7ae Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 17 08:09:11 2018 +0100 ADD helpers for extracting GO hierarchies commit 121d6b2706c08b3824ea8cef5d1ce5f385a1c634 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 2 12:34:10 2018 +0100 ADD gene ontology helpers commit 220b5ae94feadc693d150c836ffe8535d0381877 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 2 10:12:08 2018 +0100 ADD helper for finding many pairwise set intersections commit 51a7b4d6543176b66afd2f9e09afbaaf3e1f8e02 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Nov 30 00:58:15 2018 +0100 ADD mygene helpers commit eef08baa038512acdaa26cd4fa70c49b0af2c089 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Nov 30 00:56:58 2018 +0100 UPD utils to include wrap objects in a list commit e3b01e9e7724b9ea1069c207ecb2c28002c3a682 Merge: 167fd85 4a6993e Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Nov 11 23:24:48 2018 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 167fd85c292fabc21c4c4f9174be7534979c8896 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Nov 11 23:24:13 2018 +0100 UPD hide_tick_labels with axis parameter commit 4a6993e8f8696f8e77e862699ad40fedd4eb84fd Merge: 96161c5 eede55b Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Nov 5 16:53:36 2018 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 96161c5e446330371405672348bfb89957db855a Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Nov 5 16:52:04 2018 +0100 UPD plot sorted helper to optionally use cumulative density commit eede55b7a89067b2c5d68eda352a6891d649d9ab Merge: 9ca01aa a7c355c Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 19 12:05:40 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 9ca01aa37fc24221343ba1b1dbd5f401ce73fa92 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 19 12:05:07 2018 +0200 UPD transparent file opening for compressed files commit a7c355cdc0078840a518359d69bd867f3d685cfa Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 17 23:21:27 2018 +0200 DOC df_to_dict docstring commit 4feb49ba2afbb64fe22d3d60048e951f6bc06500 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 6 19:34:13 2018 +0200 ADD simple scatter plot helper commit 2a163f930ad95d16408f3225c8be2cf2d7332f2b Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 18:11:23 2018 +0200 FIX circular imports in utils commit 3ef4356aadc5aaef528d591d061897f0723e775e Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:47:20 2018 +0200 DEL load_config from utils This function needs to use some functionality from validation_utils, so it cannot be included in the base utils.py module. commit 41a9347455234d86988408ca08bbdc6f13856ea6 Merge: 0a86f5d dcf2ffb Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:22:40 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 0a86f5d210a29dc951bfc8f9b9ffb2781284cf6c Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:20:38 2018 +0200 FIX missing imports from utils commit 2823876f2d754193f0e05a603a54951a9125f9c4 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:19:57 2018 +0200 UPD creating bar charts in mpl_utils commit 06ab3ce7527747bfef380a34d04cc3a3b6195c34 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Sep 2 17:17:21 2018 +0200 UPD yaml config loader commit dcf2ffb0c37c2c0d78ae8de3761f08c20eab389f Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 16 11:17:52 2018 +0200 UPD `compress` keyword for writing data frames commit c1ec0e4301bdca4483768ca37a62f67def3c64dd Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 15 13:07:00 2018 +0200 UPD dask_utils to accept priorities for jobs commit 7afa9d8cc95913b74fd6f2470fbb7b19cc722a6e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 15 13:06:26 2018 +0200 UPD specify bar chart tick offsets in mpl_utils commit f198ef2cfe1934d3979ef8522df358eb3d566144 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Aug 15 13:05:50 2018 +0200 FIX numpy import in function body commit b4d1c740f78b5a495f2665148ab16a33c3f4b6e8 Merge: 1eda060 56ff5a2 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 9 12:39:27 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 1eda060e336dce01c8b16088d5d2111227a3c1d4 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Aug 9 12:39:13 2018 +0200 UPD scip output parsing commit 56ff5a29eb275a6dc4863dd28b6bf82fc8e6cd97 Merge: 2ca6d92 694ed19 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jul 19 18:37:55 2018 +0200 UPD mpl_utils to plot simple sorted lists Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 2ca6d92ffa57bfe6ffbe82ba11cfb12c6d45a98e Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jul 19 18:23:22 2018 +0200 FIX typos in num_bow_union commit 694ed199678cb392efa44d728cea8fc23d564b74 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jul 9 21:58:48 2018 +0200 FIX typos and missing import in SimpleNumBowUnion commit 72c005f7f60d955a200131a47bcfce045d92fc59 Merge: 157fdba 41b0c2d Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jul 8 18:15:40 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 157fdba4f9daee4b296415bf5ce30243ef6dc952 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jul 8 18:15:15 2018 +0200 ADD [WIP] simple handler for BoW and numeric data commit 41b0c2dfea03b520bca72d00ad4024e42ca8ef3e Merge: 903dbc6 812fae8 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jul 5 22:00:27 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 903dbc65baf61a58383c192cce34774aa641947a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jul 5 21:30:42 2018 +0200 FIX missing import in utils commit 812fae88d69224e2fba9670d8951604a9182a5ba Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jun 28 12:01:27 2018 +0200 UPD cancel helper in dask_utils commit ee51abb9cfe21cd8c499b7a47f4804157141c654 Merge: 0b2d9a3 241cabb Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jun 28 10:34:57 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev Conflicts: misc/validation_utils.py commit 0b2d9a324e9033339d8623b89bbaf5b7834094bc Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jun 28 10:29:42 2018 +0200 UPD k-fold splitter to include validation set commit baadb50deab68f9b5c7639e81a26b630e8bec295 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 14:45:55 2018 +0200 DOC changelog commit fc2f1e818e79bcc963bc249e820ee93686e28c3e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 14:44:52 2018 +0200 UPD cinc gender and icu mappings to include reverse commit f3fcecfd7efea8dee23ef671e46a2541d041172f Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 14:44:32 2018 +0200 FIX nan_ohe to handle sparse matrices commit 4c70d655cd856c5924072790a565bfe202a1b65b Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 13:31:35 2018 +0200 UPD changelog commit d53bb3f43d20703e007205af19d73e56bc227e1e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 13:30:42 2018 +0200 UPD validation_utils to include non-pydata helpers commit d5c6aef891b0a35259cd4654983fbaf80d6f64af Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 12:41:17 2018 +0200 UPD utils to include reverse_dict helper commit 82d405a5a28ac3dd9e3f537602e0d8c21ab7124a Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 27 12:40:55 2018 +0200 UPD cinc-2012 field constants commit 241cabb75b42d534d8fee148c49612906e715a6a Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jun 19 13:42:42 2018 +0200 UPD validation_utils to include more general validators commit 1fb81fa9864d2406d195e691c2c5366469bd4017 Merge: 95d75c4 d47b4fd Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jun 18 16:07:04 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 95d75c421b53a70e614c859a7c5cd8774c718a38 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jun 18 16:06:39 2018 +0200 MNT merge from master commit d47b4fdd9a5dd07289c51a67b9bfc5de7ce60a26 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jun 8 00:33:12 2018 +0200 UPD helper to collect dask futures commit c6978f8b26f4a609f524295a852e7a839bc5a414 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jun 6 12:23:16 2018 +0200 MNT update version to 0.2.10 commit 6c9bf7c4758e2aa82b1a3b2804e15a123b0c47b8 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jun 1 17:01:35 2018 +0200 UPD a few numeric validation utils to handle sparse matrices commit debaf0ab2822ed8f526ef64628f837b5a0e3d7c7 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 30 19:45:22 2018 +0200 DOC updated changelog commit a5c2502aa66955a889e552a3d87cf1478ceac4ca Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 30 19:43:47 2018 +0200 UPD physionet utils to work with mimic waveforms commit 3f787f20ea278a069488c7fdd23c023db1eee230 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 30 19:43:18 2018 +0200 UPD check_status helper for dask futures lists commit 8e5722f7a87585a7190d83db03ab0fd4fa1bb553 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 30 19:42:34 2018 +0200 FIX parse_scip_output when scip crashed commit 8f56e03203809478a9d2e83872241c1eacd15fe6 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat May 12 15:27:45 2018 +0200 ADD followup table construction for mimic commit a3eb613edcbeca1a8e5f7da505a10346fcbbafd6 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu May 10 00:27:30 2018 +0200 ADD cinc-2012 time series names to physionet_utils These are fixed and will not change since the dataset cannot be linked back to MIMIC. So they can be treated as constants. commit e4b7e73d4096a0b4b83561db1c4aa83213dd8f63 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu May 10 00:27:07 2018 +0200 FIX typos in deprecated message commit c9ff1f48a4b52632857bc6ac662061cd31402cae Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 9 13:21:22 2018 +0200 MNT notes in changelog commit 19a18cbee58a677d2d83b20d0410c78f652b424b Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 9 13:20:18 2018 +0200 UPD ds_man to optionally encode target variable commit a32cac7569d285606abda11005dfa5047c4ecb36 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed May 9 13:19:23 2018 +0200 FIX scip_utils to use dicttoolz for merging This removes a deprecated warning message. commit ea3b707d0384e4d1bfebba4e81db44f89f3ae22a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Apr 19 19:33:44 2018 +0200 UPD debug output for nan_le commit 7e67f6dff5e847a6a276ac38377f87a72c12b038 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Apr 19 03:56:39 2018 +0200 UPD validation to handle ragged arrays in int check commit 72a850471983332f81d6a907bfe781b8b51cd0b4 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Apr 19 01:46:50 2018 +0200 FIX missing indices in ds_man.get_fold commit 9f55a5688571be0bf3615f7ffaee41fcaf2edef5 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 18 22:20:38 2018 +0200 MNT bump version info to 0.2.9 commit d2f8a10935e78125d33af6df53c96f3a540e0fae Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 18 16:24:21 2018 +0200 FIX missing sklearn_pandas dependency commit 8e533dd267266ef1c1cf71af4823a2f73d856074 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 18 16:17:46 2018 +0200 FIX clone command for users without github accounts commit 38b959b0e7b25d123e7648de4f8722df21d92811 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 16 22:31:26 2018 +0200 UPD nan_le to avoid printing very long debug messages commit 6c3ec25491f1a1978488548b5555ac18e9576c43 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 16 17:42:49 2018 +0200 UPD get_kth_fold to include indices of training, testing sets commit 3980beab682ca8e2a08a367e3d72c514a43bbdf6 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Apr 16 03:11:40 2018 +0200 UPD sklearn metric calculators to handle multiclass, regression commit 77f0493ecc7eee71314b6e03fe08a00c8bdfe14c Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 19:29:05 2018 +0200 FIX missing imports in validation_utils commit 5c818ec4dc8eaf45fd4e48eaf624aea602bb1ab7 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 19:27:12 2018 +0200 ADD validation_utils module commit 25c33ceff0ce75d070eb478e7605f639e5561c51 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 17:46:49 2018 +0200 DEP dict helpers They now suggest using toolz.dicttolz functions. commit 09959b9f211e8daa7ef771fcd500ed840ce7762c Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 15:54:41 2018 +0200 UPD pd_utils to include apply_group helper commit 480f598d8ec6b9cac24e79c416732b3dc4567842 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Apr 15 14:34:29 2018 +0200 UPD split_df to accept chunk_size instead of num_groups commit 6366d7a9fd0d120beaefdd6ca16f6b63b811ac27 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Apr 14 13:08:41 2018 +0200 UPD replace_nans to treat np.inf as nan commit ae414752ad23b63ff59a84d011d4922d0b4dd661 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Apr 14 13:08:11 2018 +0200 FIX missing import for pd_utils commit 9ace0ec393f05a0d0d3b9a60f4a82f96caed0354 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Apr 14 13:07:05 2018 +0200 UPD cat_mle to handle base-1 observations commit 177f9655863079848bac561f5dd94f770ccf6b4c Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Apr 14 13:06:24 2018 +0200 UPD ds_manager to handle inf's commit c596f907f94b8d565942b59902ae5d3aa63a0673 Merge: b521145 d76faef Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 4 22:49:31 2018 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit b5211452ca6d6894674e603e4db0049bdb6cfd97 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Apr 4 22:49:00 2018 +0200 UPD math_utils.check_range to return whether the value was in the range commit d76faef41cfb513b9d0739784b50e9aff88388ac Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Apr 3 18:19:17 2018 +0200 ADD check_is_fitted helper commit 5a1ad0de8bd63196b24f7d35dcc3c8212cbff5aa Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Mar 21 11:12:25 2018 +0100 ADD utilities for parsing, etc., SCIP output files commit ecbc2efad6bfbe1667dee7e2d00819ac4a39ae2c Merge: a542cac a02b8a2 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 19 18:34:41 2018 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit a542cacd250aa4078476d49e4094633093afc81b Merge: cc5ffe6 f47f0da Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 19 18:34:02 2018 +0100 UPD md_utils to handle ds_mgr without categoricals Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev Conflicts: setup.py commit a02b8a2dbc2a1c96a774b9115d4ee32dbfdc3da5 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Mar 18 21:39:18 2018 +0100 ADD helper for categorical MLE calculation commit cc5ffe6d2767ec9148556132da6ff5585ae1fd9c Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Mar 14 01:43:40 2018 +0100 UPD simply apply helper for pandas data frames commit f0565bac888acfdc2c63ff455530a7f88e189caf Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Mar 14 01:42:43 2018 +0100 FIX missing sklearn_pandas dependency commit f47f0da44bda88ce434e735defe3849b403dbf40 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Mar 13 00:05:30 2018 +0100 MNT download small nltk resources during setup commit 28af5572169e9b64764e5505562841c39f0428a1 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 22:35:30 2018 +0100 DOC use docstrings for all module descriptions commit c2433fc300a136c93e8b4434c1a5b99ef1a19407 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 22:15:35 2018 +0100 DEL spurious automlutils import in missing_data_utils commit 6c1ad9781dea19e7167bbfe36a327c35d21be3a6 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 22:14:12 2018 +0100 DEL partial class The use for this class is extremely niche. Also, the partial class cannot be pickled, so that typically does not make this a very useful thing to do. commit 78e79b39f530caf9e45818be6d25515a8fa32d7a Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 21:53:02 2018 +0100 DEL automl_utils They are now in their own package, so no need to add the dependencies here. commit cf7a39b9d587cbcf3d379435d4cde038562cbe33 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 21:52:29 2018 +0100 DOC added unsupervised cv to changelog commit e64fce50acb34e0f876e426fd51ed2df7dfe746c Merge: 7779ee4 cf7a39b Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 21:44:11 2018 +0100 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 7779ee41cb810e5e7a2674668936ef03f216e267 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 21:41:28 2018 +0100 DEL mysql helpers These helpers add another large dependency that can be difficult to install correctly. Thus, they are likely to cause problems and the logic is generally handled by higher-level libraries (say, flask) anyway. commit 92b22f05f1dafe253b781426ebd2e676e4fb2995 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Mar 12 17:45:13 2018 +0100 UPD k^th fold util to handle unsupervised data commit b0b0bc71bfd8c0cffc1785859fc4fd28f05ebc08 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Mar 8 20:38:03 2018 +0100 DEL pickle-stan and pystan dependencies commit dde8764fbf17f5784e08bdc441ac1cbc829bd900 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Mar 3 13:01:22 2018 +0100 FMT changed logger level for `ensure_path_to_file_exists` commit f28347d7a53577758a1695b5bbe665804283e8e3 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Mar 3 12:08:19 2018 +0100 UPD pd_utils with helper to create chunks of groups commit 0b7945edaef5a65b55983d7a8bfdb361886fd123 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Mar 2 19:48:42 2018 +0100 MNT remove executable permissions from files commit 6e93b222e2c81d22c7524abd8cd4073884df4339 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Mar 2 19:13:57 2018 +0100 MNT bump version references to 0.2.6 commit 718ff9e2c7a68b1d575acfd3bb6f1866b7798104 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 28 13:48:52 2018 +0100 FIX dataset_manager to handle missing target and dropping commit e5cb6e7ea108fa42555739a4e4d2aa5138f23aab Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Feb 7 19:52:10 2018 +0100 ADD incremental_count_vectorizer commit 6a1c4590cf3799e24175b1729052540b5171eee2 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Feb 2 13:27:33 2018 +0100 UPD utils with merge_sets commit a0d869b5dff10c4867dc60900db0978d553452a0 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Feb 1 19:07:20 2018 +0100 MNT updated changelog for nlp_utils commit 7b1b76dbdc75de8bb2952cdcb42983a6d311db0a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Feb 1 19:06:27 2018 +0100 ADD nlp_utils commit 3c3a951da1f8c6d8a8140bd181de2c53b8ebb71a Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Feb 1 17:51:35 2018 +0100 FIX progress bar in dask_utils.apply_df commit f1950f2ff8f787df12149cdbdb1b46ca05585dca Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 30 15:41:00 2018 +0100 UPD utils.open to accept args, kwargs commit 90ec5a4389386d989c5b8cde21233d516f1aa2a1 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 29 15:23:54 2018 +0100 ADD dask helper for groupby results commit ce6a5183fd5a738eb6634dbbdb065c02227e752f Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jan 24 00:26:22 2018 +0100 ADD dask_pipeline helper for submitting sklearn pipelines to dask commit 93cb8285a2fcad1a6d7c2dc9b0fcad418ec04b5b Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 23 12:43:13 2018 +0100 UPD optional fields_to_ignore for dataset_manager commit f00dcff577029c66ff22fa49bf57362686ad9f48 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 23 12:42:37 2018 +0100 ADD helper to create missing data preprocessing pipeline for sklearn commit e1506ac8a6d4bb53a0c3e4d9910e7b0208d737e3 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 18 17:06:04 2018 +0100 FIX nan_le incorrectly converting floats to strings In particular, this was a problem in the `fit` function, and the `classes_` dictionary would not have the correct type of keys. commit 85b3dc6c3c2959521a464004dbba5616de85e801 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 14 22:13:53 2018 +0100 UPD physionet utils to read more tables commit 3a5ce7899eedb80be29a7e15cbe63a7ffa2bb469 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 14 22:13:12 2018 +0100 UPD nan_encoder to use dictionary for encoding commit f32d2d5f4641500ab643f64cf0bdf49cb1b3199f Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 14 22:12:17 2018 +0100 UPD `dask_utils` to have `apply` helper for iterators commit 10dce10879b1ef4e2dcc2e945af054c2107736e2 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Jan 14 14:50:22 2018 +0100 ADD `apply` helpers in `dask_utils` commit 93ae10eb80e9030bc42b73eaaf4a1d930824f716 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Jan 13 22:07:27 2018 +0100 UPD physionet helpers commit 033e9498ea2a1f91ee4fb8b7cc38f03bad761755 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Jan 13 16:13:51 2018 +0100 FIX missing logger in physionet_utils commit eb7881838643052e6a29e5b809c5b58bf80339db Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Jan 11 00:06:29 2018 +0100 FIX nan_encoder to handle object dtypes commit 320f463e6f875ea33458282983588920c45c941e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Jan 10 16:11:13 2018 +0100 UPD nan_encoder to handle unknown labels In particular, if asked to transform an unknown label and the `treat_unknown_as_missing` flag is `True`, then the encoder will replace the unknown label with `np.nan`. commit f8dafc2ab78eba3a1db2db2027725c80adf5c297 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 9 18:17:47 2018 +0100 UPD dataset_manager to handle non-numeric targets commit d517217503e595780f30deb59e1efc19e05f1b5e Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Jan 9 10:51:54 2018 +0100 FMT remove print from nan_scaler commit 0400b80f674c70e58dbff68d6d5974802392dd19 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 8 19:15:28 2018 +0100 FIX nan_scaler to work with 1-d np.ndarrays Previously, it would work with np.arrays (which are inherently 1-d) but not ndarrays which happen to only have 1-d. commit 47b097eb884b6a0ea548f55d9db1cf0181139bd1 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 8 17:15:15 2018 +0100 UPD nan_scaler to handle 1-d arrays in `fit` commit 28290f767508e7c8744702686791b10c2dd47005 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Jan 8 17:08:47 2018 +0100 FMT logging output for nan_label_encoder commit 413ce18ee05233988db3f5e705cffd62a07f4fe3 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jan 5 04:05:12 2018 +0100 FIX type in nan_le triggered by corner case commit ea7d85222917e1c9918ce062bddf7d97bd1c853b Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jan 5 03:36:44 2018 +0100 UPD nan_ohe to handle 1-d np.arrays commit fb7eebb1fd80f48442a7923e42af91fbb5913f57 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Jan 5 03:36:15 2018 +0100 FIX indices of encoded labels for nan_le commit fccc33095dee93254cce1d9e7b6426870ab91930 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 13 18:55:24 2017 +0100 FIX nan_nn to point to new missingdata submodule commit da99f0f1ce960fec905350cd355a3bb037849569 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 12 22:35:48 2017 +0100 FIX nan_one_hot_enc to handle no categorical variables In this case, the one-hot encoder simply returns the original array unchanged. This is consistent with the basic sklearn implementation. commit 9311e56e742deaca60a81322c44b3c62c7ed78c9 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 7 03:15:11 2017 +0100 FIX typo in import for new missingdata subpackage commit cafe94a8b11715574e47eaada3d5256dadf370e1 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Dec 7 03:14:41 2017 +0100 UPD category counts in dataset_manager commit a7f6d5ee1bd71b5bce4423807908895f28289662 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 6 22:38:32 2017 +0100 MNT moved missing data utilities to subpackage commit 8e945259c79311fe51e6e4f751300e5c94b0f7b7 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 6 22:20:15 2017 +0100 UPD one-hot encoder for missing data This implementation is now complete. commit 8932b4c62fefa18807584a0be586be3b70b1615b Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 6 20:27:18 2017 +0100 ADD [WIP] one-hot encoder that handles missing data commit deacf9b69f75820d6a77afa5af51fadd335ee9c1 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Dec 6 02:29:23 2017 +0100 UPD ds_manager with various helpers These mostly ease extracting information about the categories of the categorical fields. commit d7d5330ceeb550e742f67e10c90e25091613c4fb Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Dec 5 23:54:17 2017 +0100 FIX unnecessary dependency in multicolumn_imputer commit e2f9c526eae9f18913491834196efe069efc31b6 Author: Brandon Malone <bmmalone@gmail.com> Date: Mon Dec 4 01:19:04 2017 +0100 ADD helpers to handle encoding labels with nans commit c64e197323a31be5e1abcb9cd30c3c6fc857aaed Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 18:20:54 2017 +0100 UPD several changes to ease working with mixed datasets commit a1c67e8b4ef345cadf8920349296d93d2ac7a1d0 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 15:16:37 2017 +0100 ADD multicolumn categorical imputer This class replaces missing values from categorical columns with the mode from the respective column. commit 123cc0fbf372adf0018ad1fa9124e794c64bdd7e Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 15:16:15 2017 +0100 DOC add column docs for multicolumn_label_encoder commit 9c6ead261bd53c8f31b8f614da0cb1e55c3764f9 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 15:15:18 2017 +0100 UPD ds manager to include field type helpers commit 3cb22298837aba8b5cd198a13a6fa26a98d9f527 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 05:26:10 2017 +0100 FIX typo in get k^th fold commit 6dd308599d6b1df2371a72770751a03da0b5f780 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 04:50:42 2017 +0100 ADD dataset manager commit 0c819ca34a8509ca59b4fa9e0b59d2ab76c6e07c Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 01:38:19 2017 +0100 UPD cv fold helper to include some error checking commit dc6a42bfa3e644be925b4e62340791d1cf4b3cb4 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 01:36:08 2017 +0100 ADD k-fold cv helper in math_utils commit 43b4f904872b2352b6c6882976ae4c2d41b12f35 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Dec 3 01:35:33 2017 +0100 ADD class to wrap constructors with fixed arguments commit af197f8d0042edb05e04e22d08a71e105ac8cebe Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Nov 1 14:02:53 2017 +0100 ADD class to suppress pystan output commit ff9aa2c3bd2f329d1f765925331bfa59b1d7d4e1 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 27 00:25:01 2017 +0200 MNT update change log with fastparquet patch commit 278a3fcc25883c63f744695a9387fef49c76cec3 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 27 00:22:24 2017 +0200 FIX fastparquet imports in pd_utils This is a patch-fix for Issue #4. commit c4b40ba75794b554a3c058955e2e9814bd16be4f Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 26 22:06:51 2017 +0200 DEL old mimic_utils file commit fbc8a050f22b4c78d6a8898662a0bd6ce1827e7d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 26 22:06:23 2017 +0200 MNT merge with master before version bump commit 742bbd78b54375d89f0f88b920b030fcce395e4d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 26 21:56:37 2017 +0200 MNT versions in setup for networkx, pystan commit 10f4a5063d39eac6b693e0dfdbded4d66a558614 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 26 21:23:31 2017 +0200 MNT prepare for version 0.2.5 commit 06b71564093bf7dd2765e64a8d1bda3b03068ad5 Merge: 84c7fa9 627b84e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 25 14:02:08 2017 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 84c7fa9b5edb7d6a196e78f6f00d5e879a45f1f0 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 25 14:01:50 2017 +0200 FIX missing sklearn dependency for math_utils commit 627b84ee74c91b2f8bc312480510c2498042b38c Merge: b2c3751 facd5af Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 19 19:39:06 2017 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit b2c37517dddc468ca5af409a1d81e4c84de0f172 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 19 19:38:18 2017 +0200 FIX ensure classes always ints in multiclass auc functions in math_utils commit facd5af1fdbc1876344d3117f942afddbcf9e63e Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 19 02:42:31 2017 +0200 FIX missing more_itertools prereq commit 00180b7af4f02b2ec518467e86dfeedfefa42e14 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 19 00:53:43 2017 +0200 UPD added identity column options for cv splits commit 0de080703ed936cb2aa32134464850c2d0129bec Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 18 23:53:25 2017 +0200 FIX typo in nan scaler commit 15b98b070f7bd1548d5d7b1063a2f3c63a76c97f Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 18 20:07:04 2017 +0200 UPD utils to load more mimic tables commit 159287060c6fb6e1e102c7991a3dbbd8a3bdf36a Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 18 20:06:29 2017 +0200 UPD nan_scaler to work with pandas data frames Parts of the code look rather brittle, though. commit 53ccd8b59170886dae561047d2c5d29ef225b28d Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Oct 15 17:55:59 2017 +0200 FIX typo in missing data training helper commit a24338762a2280cf61dc3fa0c53a4416244d7b94 Author: Brandon Malone <bmmalone@gmail.com> Date: Sun Oct 15 14:43:37 2017 +0200 ADD helper for training and predicting with missing data commit f03209c1e42314db7eca477bcd8525af46b3f530 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Oct 14 15:37:38 2017 +0200 UPD readme to be in-sync with the current contents of the repo commit 2c07202da9883f74cfbd230ea64c55b31410edeb Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Oct 14 15:31:14 2017 +0200 MNT deprecated automl_utils commit 27d52375bed7a4c70a40c18f6c7b6eff40b245c0 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Oct 14 15:19:01 2017 +0200 ADD simple knn wrapper which handles np.nans commit e5eba04f9ab45d1fc7d183e838a1670d50e94182 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Oct 14 15:18:17 2017 +0200 ADD utils to remove data according to different missingness mechanisms commit 80650a8e93cadc4ace6d3304316803b776ca4c8b Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 13 20:41:29 2017 +0200 ADD nearest_neighbors which handles missing values commit 12cb576b8b946218de5fa5c228d8d9ef5f42d4ad Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 13 20:40:16 2017 +0200 UPD nan_scaler to handle 1D input for transform commit 8a32c85037ec4329ecba4560c73c97b232ca0e5d Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Oct 13 20:39:29 2017 +0200 ADD distance metric helper for vectors with missing values commit 2680184b97fa0fce03716cdfa2f9a90c67e0d19e Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Oct 11 19:54:01 2017 +0200 ADD math_utils helper to check for nans commit 765f2cbdbe267e33672f01e4fb5b4f6bd23444a5 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Oct 10 18:35:56 2017 +0200 MNT add new updates to changelog commit d166e29dc341fe9f323498171e8f0ed30fe6c00f Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Oct 10 12:42:46 2017 +0200 UPD mpl_utils.plot_confusion_matrix to work on axis objects commit 359b2d654e5c417a7e2156ebe5192c328f20ede9 Author: Brandon Malone <bmmalone@gmail.com> Date: Tue Oct 10 12:40:11 2017 +0200 ADD helper to randomly remove values from np.arrays In particular, math_utils.mask_random_values uses an MCAR removal strategy to add missing values to a data matrix. commit ff88c23d4e047b30baf726b6775de44440bb7488 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Oct 5 16:31:50 2017 +0200 ADD multicolumn label encoder commit 53ac2c1ec0913f1899fccbdf994958d5e07037a0 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 15 18:00:43 2017 +0200 ADD join df list helper in pd_utils commit 138636bbb62d48db09422dd117b10db57ea41e32 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 14 15:17:12 2017 +0200 UPD order of params for binary classification metrics to match sklearn commit b0f59b53d4ba556a01ad25db6606f50c1f7a52b6 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 17:13:47 2017 +0200 ADD multiclass auc from [Provost and Domingos, 2000] commit 57c3207979655e451cdf86b349269dd1cb67bce4 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 15:02:48 2017 +0200 ADD multiclass auc calculation from [Hand and Till, 2001] commit 88ee39a7523666386d6025a01569ab56aa843ef3 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 15:01:37 2017 +0200 ADD convenient font sizes for mpl_utils commit 36eaf0a2948e0b08ed7b23b1b9b87a484b019ac6 Merge: 8d4de7a 7df5e0f Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 11:14:44 2017 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit 8d4de7a56f9f3bdcfa518778a21a7fb3fa8a6a45 Author: Brandon Malone <bmmalone@gmail.com> Date: Wed Sep 13 11:13:25 2017 +0200 UPD classifier type-pretty name map commit 7df5e0f7b5f1c3fc67b4e1c9ab706a620a3bcf4a Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Sep 9 17:28:06 2017 +0200 ADD helper for drawing rectangles in mpl_utils commit f6721f006606fde808acd5ad451f055e24ab1153 Merge: c58c61a fc0f9b4 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Sep 9 14:05:39 2017 +0200 Merge branch 'dev' of github.com:bmmalone/pymisc-utils into dev commit c58c61aca1ad46b3a0dbea22560a27b06e23f0e3 Author: Brandon Malone <bmmalone@gmail.com> Date: Sat Sep 9 14:05:20 2017 +0200 UPD sklearn type-name map commit fc0f9b4cd47859458199dc647f09a384a7fbaac4 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 17:40:01 2017 +0200 DEL kwargs from asl_wrapper member variables commit 4af7681eb318c8c8d832c54fd05651d74aa725b6 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 14:30:47 2017 +0200 UPD asl_wrapper to accept metric in constructor This addresses Issue #3. commit 843c6278f2b23fef75fa6ce0a518c2d55d1930f2 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 14:18:10 2017 +0200 FIX asl_wrapper to save its label encoder This addresses Issue #2. commit 7d19c061935836d82df129179a062beec5ba4ce0 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 14:08:39 2017 +0200 ADD asl_wrapper ensemble model summary helper commit 799617686a20ed5ae024470841b73a6b3a011a66 Author: Brandon Malone <bmmalone@gmail.com> Date: Fri Sep 8 14:08:17 2017 +0200 FIX utils.get_types to handle unknown types commit f3d4ad0bfd51be4a69ffe99f2f4b3627307e379d Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 18:50:17 2017 +0200 UPD cinc-2012 to use HADM_ID as the id column commit d71b77d42e9e7a7638c4046fd126bf24bfc83a64 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 16:43:53 2017 +0200 FIX parameter list in automl_utils predict method commit a961bf12b5616469e03f4859c1b07eca7be9ff31 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 16:43:08 2017 +0200 UPD loading CinC 2012 records to handle missing gender commit 611f3d4999d122501f532bf063dbf2ef1aa87227 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 16:40:49 2017 +0200 UPD index parameter to control writing parquet files In particular, fastparquet.write looks for "write_index", while pd.write_csv looks for "index". The pd_utils.write_df function now converts "index=False" to "write_index=False" for parquet files. commit 80b1298feb9a5fa4b3cce4d2d02fb3541ff8e234 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 16:40:22 2017 +0200 UPD dask utils to optionally restart client commit b9bff1db66b65c7ce1f601b536680a11eebcb474 Author: Brandon Malone <bmmalone@gmail.com> Date: Thu Sep 7 13:15:21 2017 +0200 ADD joblib backend helper in dask_utils commit f22dcb9149dcc93f8a38a0c…
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The
stats_utils.calculate_univariate_gaussian_kl
function aims to be numerically stable by performing most calculations in logspace. However, it is not clear that the equations are correct.The text was updated successfully, but these errors were encountered: