• a0a117aab8687ca3dffa98e4de790b4ce21599ab [python] make dump_text() private (#2434)
• 2bf9a0f10c60e7f62cfc6eaa9fce78ae119a4eb2 remove the duplicate data path (#2432)
• fe9f292cb3403861f0f755688d1e94fd981ce207 added new early_stopping param alias (#2431)
• be206a96e7be553b0918805f6641e91285ce89d1 refined order of includes (#2436)
• 2e9f255e91016f0b4133d4436d2ea49ad9e1d0be [docs] refer to RTD site directly (#2415)
• e47d9414282ab93bd4e1e582cb17632a711643a6 [ci] temp fix for missing dependency on Windows (#2440)
• e77658205d3b1bc5c19e4ba3821f4222f5be8c05 Update setup.sh (#2421)
• f1fbc2f1eed48318aeb83117d9606c1e3d145052 added JAVA codeowner (#2430)
• 4fee5821eeec82521a1e58ca275982b3abcd68b9 fix codeowners (#2429)
• f1a1486929f8e981c44cba45132f5b4929453f8a fix many cpp lint errors (#2426)
• 4f89cc10785318f051602f5fa7fa9d61bc805980 [ci] added build/include_subdir check to filter for cpplint (#2427)
• 306c6dbfd507c2d89b0d593cc2a42cd72a14e777 [ci] added cpplint check (#2416)
• a3a353d6832b3653f222886f912293998ff8ae69 At least 2 features are chosen in subcolumn (#2409)
• a119639a5bfaa78887a4867da611da977fc072f2 fix the objective init issues in distributed mode (#2420)
• 023749239ebf390b5d56e08f227a013cbd972eb0 added editorconfig (#2403)
• 84754399a6b5575906628d457559ff6fb5aca2c2 [python] Bug fix for first_metric_only on earlystopping. (#2209) [ #2273 ]
• 939a02482bda2ff5e15887549a8ddfbe9865b087 fixed C API SetField description (#2406)
• 15566421bcc96d46bac353dee17b981a12827604 [examples] Update example notebook (#2391)
• ad8e8ccc9c9a5f8acb192b28715827ebc28dcc07 update feature_fraction_bynode (#2381)
• a1a4103797fd320f5379f9ad399affd34e50c6a5 [examples] really use weights in example (#2378)
• b310fb417e4078cc30b153a5ef8f60c06d9ca9cf [ci] add if-else job to Travis (#2389)
• 9f6e4413bddb7395f9dd1b2678df4820bcb65483 [python] keep consistent state for Dataset fields (#2390)
• f52be9be5091d300078f0b4fa3641a165ed3f905 [python] Improved python tree plots (#2304)
• b6d4ad8361379a902091e3a584491c39fa89ef2d added JS file (#2387)
• 33d0378ffe1c7d51adb9ac20ea137b94841d0677 avoid nan and inf in weight/label/init_score (#2377)
• de1f3cb3e2e71dd83913447fe70b52b59ad8aaea [python] removed excess condition (#2385)
• 22cd39e89a9139f7fb74aec7b612cef57ddf760c [python] fixed typo (#2386)
• 7509ec8ade06b13bd10aff41ece78c02a7993535 [python] fix group type in lgb.cv (#2384)
• 29525ffefba94ea994e28d0db1d8029a6041e3ae [docs] added reference to leaves Go package (#2379)
• 4c8418ea12056fb364d6e0b4357a1bddad890971 [docs] R docs cleanup (#2375)
• df26b65d786532b88b13d13cf8e4b7ac9fa8d8cd [ci][tests] install joblib for test directly (#2374)
• bbbad73db3006ad75db21d0947b50a97610d4d90 sub-features for node level (#2330)
• 288ba0ad8fd9815b4778a15af2b2ff9da09e94bd Added link to Ruby gem [skip ci] (#2367)
• e7c6e67aaa975fb515f39c6938671d1a12bef64f [tests] simplified test and added dumped data to gitignore (#2372)
• 0f9720bffe996183382209ee6171069c3cfee550 [docs] fix version type in the R-package tooltip (#2370)
• faba6cafb7ea74203102d1b8752e53ada6e7c0cc [docs][R] added R-package docs generation routines (#2176)
• 317b1bfbbfdd8f8a4f5af16b970526bb9486f588 [java][mmlspark] Fix cached predictor causing bad values for predicted probabilities (#2356)
• 254a8699c9b869fde4547cf87289f455d690f7f9 [java][mmlspark] fix lightgbm on windows (#2364)
• 0551f77517d971ce9bc32e2633768245957b410a change network retry delay strategy (#2354)
• 5f5dcff60471853d27b1c0ff82122199e9196080 [python] correctly handle full path to compiler in CC and CXX env vars (#2345)
• 4fc0416fabee5ba8f062f6d7dd966e1b207b7f4f [docs] Clarify LightGBM on Spark Reference (#2358)
• 76e57c5dc0c45e75f1c3f8ba7ca246952228357a informative error for custom obj in RF (#2355)
• 0dfda82607633132e10a693eba9666ed75585ac8 normalize the lambdas in lambdamart objective (#2331)
• 5e60224426f99a1a58fa85f279d5e3fad01e1b09 fixed compiler check regex in install.libs.R (#2346)
• 15d861cec3dfde1fa31c777978cc74f59cedf777 [R-package] fixed overly-strict compiler check in install.libs.R (fixes #2259) (#2339)
• 20f94c52524ace54a8f05df3b3ad07d1f91ff617 fix the bug in bin with small values (#2342)
• 86c6a2d0884521e971c8cf17947438f1be64661d [R-package] trimmed whitespace in R code (#2338)
• 8415924c8fac4c1192cad209e38193ad0059e0d3 [R-package] corrected typo in R unit tests (fixes #2336) (#2337)
• c0883f174cab469a3a59f7796807081fe31d6497 Fixed keep-empty-dirs flag in build_r.R (fixes #2332) (#2333)
• 686b0cc2f9a1639a65a3b9ae4b7ed11bdd142082 [docs][python] allow usage and compilation of 32-bit library (#2188)
• aee92f63ba124e1f6a3168eb2864d032567cbf9e sigmoid_ in grad and hess for rank objective (#2322)
• c421f89851dd460ce50c8dc8f4b638f9b5748501 Bug fix: small values of max_bin cause program to crash (#2299)
• 9558417a12dec04b2ee9ba1e3f8be40521b33642 fix nan in tree model (#2303)
• 578a8c8a996c74c5d408e1ca5dff27bfe81f79cb updated docker folder (#2316)
• b0b45911c0bc87f18cfc09fba6cc62e0d0f90d01 updated R-package authors (#2321)
• 8f446be749cb7a03fc4c65cd47e595a4e5c94d1f [python] add sparsity support for new version of pandas and check Series for bad dtypes (#2318)
• 9cf6b8283d6257bf5b044197de7c77c709c85def add Dockerfile for R (#2308)
• 5cff4e8e2fb2280ed302ed73c08bd95d035c7889 [python] Deep copy params in _update_params of DataSet (#2310)
• a8d8b07d3afd3b2386c64fce3fd2d8f302372ab2 [docs] remove aliases from options in params and remove separators in FAQ (#2296)
• 6b94c8e0f2b841785b77cad9f69736419c6d2518 [docs] generate contents in Installation guide automatically (#2298)
• 28dc333d3f5eccf0a904aff4fb2404a57850f421 [docs] fixed params description (#2306)
• 0197dba0613c46316518ef022e18a110025a1c59 [ci][docs] update the list of ignoring sites for linkchecker (#2307)
• f68e1b50fb87f157b36cda16136a326e5a29db09 [docs] limit number of files to scan for autosummary generation and reorganize conf.py (#2297)
• 55656f27e13e35a634eb01a11493b10ca1c65c1c delete the variable which is never used (#2294)
• c17a5f5b8402908da1b1dfdc74a6e40530a6ffe0 Refine the description of is_unbalance
• 6c210209fbe4694af613bca405fa5de72f10d72a change the priority of init_score and init_model. (#2291)
• 04a56010ec94124da9e8c37006278a5a4b34b842 [docs] 📝 FAQ overhaul for linking to individual questions (#2293) [ #2268 ]
• 11c24cf83584bb771b42520071e9c239700b0e86 [docs] added FAQ on early stopping (fixes #2270) (#2272)
• 274f340ae8232c707428d1747a857f4fc2511146 allow passing empty value in parameters (#2289)
• 207bb3ef329b38fc255ef36b1d7ca765924363ea [docs] 🎨 Sphinx Autosummary for generating Python-API documentation (#2286)
• 5dc1507e1ce0e21db37d2132ed373f8c8a708351 [docs] updated js file attaching mechanism (#2287)
• 5d3a3ea47eb8693c924ffb891387e93534f92896 fix metric alias (#2273)
• 716fe4d015c2333f9a261f853ff559ef385add17 fixed cpplint errors about spaces and indents (#2282)
• ee28ea36a456a064bc1d6ea900a0880f9a3c711f [ci] remove temp fix and fix activation for conda (#2280)
• e1d7a7b9e332851063ebb066d395937afefe76f1 add weight in tree model output (#2269)
• 86a957830b7caf2cedb7aaa23e824df6b38f7c7f fix MissingType::Zero in categorical features. (#2275)
• 7189743d224a405bb48050834460d107fb7c1c0b Update README.md
• 0d59859c670b9de37bffa8a6e536497c88d9f25d throw error when meet non ascii (#2229)
• 0d02499e4c00fba279b4e3aa15f314800d1d4461 [ci] added CODEOWNERS (fixes #2194) (#2196)
• 7360cff94b65562509989802183223017ed3bcc0 fix init_model with subset (#2252)
• ebc831bc53b746c5cc079955364008f591877015 fix bug when using dart with init_model (#2251)
• 291752de3e647f645689f180a0a2ba0efdaf80b9 Max bin by feature (#2190)
• 1bd15b9447e0925e6704cfce6caa3ba4fa9b7758 fix refit decay rate in python (#2254)
• bf78008b47a7b1c0ca7c6174481e21b85d02c1aa switch name and alias of rmse metric (#2257)
• 4ca85fbce7db160414141db76b60691b65eaa229 turn on release notes (#2256)
• 16751b3803e847592dba3bab876c4f052507e45d [ci] downgrade conda (#2245)
• 62935f52621549bd273dc3e11d92c6b444435aa7 updated boost (#2218)
• 86269ee38da20785097cd35aa7eb0b946864fe76 [tests] use numpy.testing.assert_allclose (#2207)
• 5b24834102528d56521ba8e44bbddf039319ebcf avoid the bad_alloc when overflow.
• cdba71475f3751e26b62beb3e6bfec9a3f6f6962 balanced bagging (#2214)
• 7da11ffeacf79be81839431b1cbc7c293ba7d9bd [docs] add "download" badge (#2224)
• 21e356d02443e675a9696aaa44cbef4ab10316cb [doc] updated links in README (#2231)
• c4cd65ff05f4594be926305ff6fad4dfb4cfdd67 [R-package] mojave install r openmp fix (#2164)
• 8ecae562582a2b76062a81b9e70888e18ccfa031 [ci] add OpenMPI link to ignore list (#2227)
• dace0508cbe94e06294fed590aa1c422d1f20fb0 consider max_depth for histogram_pool_size (#2216)
• b6f6578368dfeb605a66df9cce753886d76f71f3 [python] fix class_weight (#2199)
• 7d03ced3880cb56af68734d447ebfdd18a9dcc11 [python] removed unused import and variable (#2213)
• de70c7d7e7b44b591336ab814a373b30f32f4ba0 [Python] Fix typo in engine.py (#2211)
• 520247ccb34d5e576fa762d70d665873abbae62a fix unsorted eval_at (#2206)
• b1e5a843ab57c64b01c8896ef3cda6fdf907d417 [docs] fixed and enhanced format of C API (#2195)
• f70a05328c87f3f9b40ab941648d15675fa9d109 [R-package] Fix best_iter and best_score (#2159)
• 2459362acd9a6ce946af8cb6d319d0919a9a4730 [python] fixed picklability of sklearn models with custom obj and updated docstings for custom obj (#2191)
• e5b6e50e30efbdf2fa1202666c54a8b6600bea5f [docs] reorg the doc in main page (#2192)
• b3db9e924a8d48a0b4d67097cf7dcc79bc50b9d9 Top k multi error (#2178)
• 19de2be0deb388a0cafc0aef90de0d1c9da49812 [ci][docs] refine C API docs (#2076)
• fc789e044a9bb2559c413905569677ec3e2ad07e fixed cast error (#2186)
• 6a1a538f456b36f05a70d6956c47be7f0dc2b7df [docs] remove duplicated param in Python-Intro.rst (#2181)
• a66c34407dcb3f94f6d29b0a5098d7ee0e1dd59b [R] fix map metric is higher better (#2183)
• a7885b60dd398fc89d797049a31c7b85713c966b [ci] removed temp hotfix for Yahoo URL (#2162)
• f01b2aca13e787e2945eba7c2bd85e31c65923f8 first metric only in earlystopping for cli (#2172)
• c9d681ac0d8c7e242b2fff217db268c11c83c745 [ci] migrate to new Dockerfile (#2174)
• 6f3fae51fe98c71a5c54e0fe27d7a05d5195c3f6 [doc] minor doc fix for gamma param (#2180)
• 3d8770aff7c4cb4c5bc4774802b68fde0500c381 [docs] fixing max_depth param description (#2155) [ #1879 ]
• b3c1ffbf9f0e7fd2c98b40b8b6ffb86308d8f408 [ci] create GitHub release automatically (#2171)
• f91e5644a3ab5104a942e2fb23f6f9239ef08a54 [python] added ability to pass first_metric_only in params (#2175)
• 2a3691703c1f5aa5109881455cfc909f80c7f50f [ci] added swig compilation on macOS and Windows with artifacts (#2170)
• 76170788c4bd6239334408f88cf51183ec791fa2 [ci][docs] updated fix for OpenMP conflicts (#2166)
• e09e4caa4ded45a074e05ab247e8a4f11ed12644 fixed minor issues with R package (#2167)
• 1a1c2fd752ee2ea27312663a6c3ccfdb6574b298 use stable_sort for splits (#2169)
• abbbbd7545e6a0549c17a182b7a59d7468cfb58e [ci] CI fixes (#2161)
• 5d6513ef20f9b51368b4171b0d19a5b0530cee18 sort after get top K splits (#2154)
• 1c27a15e42f0076492fcc966b9dbcf9da6042823 removed external dependenciesin build_r.R (fixes #2122) (#2123)
• 94fbe5bb9fbb5f6067821f0be9cd145f0b7d7d94 [docs] updated Microsoft GitHub URL (#2152)
• f46f8b2a049f0745c0b8eedfd662289649faccf3 fix warning
• 46d2147684a86d7b4219c83386c469d34c1b075f fix a bug when bagging with reset_config (#2149)
• 2c41d15e06a54b3757ac0b3cae53390fb4db52c2 [ci] nose is not needed anymore (#2151)
• a47782f540c9afe149ed608cd68efea53e4746ed added estimator's tags (#2150)
• cfcc020e5857b152cf4664af834961f950fea0a1 [ci][docs] generate docs for C API (#2059)
• 804c5cdfa7c187f3197d4fa69452bf4ad48ca060 [doc] OS name in bold (#2147)
• b40d0dc3d65e84fca7eec1689d238c9943d37db5 [ci][SWIG] fixed SWIG -Wstrict-aliasing warnings (#2139)
• 7156c9b09edcbd61ea82ed6c73e73c46d6de23c9 [SWIG] fix swig build warnings/errors on windows (#2141)
• 40e3048f6185bb8f3f50bd9fe7275cf514b03b16 [ci] removed graphviz from docker (#2142)
• 93a899ac7c80b7e71ea2ccc31b4f15ba3fabb7dd [ci] hotfix for check the docs(#2145)
• 611cf5d414a8273dfdd261651e2f127322a3db89 [python] added plot_split_value_histogram function (#2043)
• 65c7779d18aed2c5d6d5457eb5ca2a32bf1a7dd1 [ci] point to latest CMake modules (#2140)
• 5ece53b5e04821e70c0ac81706531c1733b91f58 [docs] updated the part about Boost in installation guide (#2082)
• 57ed8794e40a77e2910a19e67635f0b53a57a696 Fix build of GPU debugging code (#2071)
• a295f6b0679d116c7f6388f420da355099e7f9b1 Removed legacy code (#2137)
• 739c2635f4084e07f5b57666bd7f2fdd929bf30e [ci] added job for swig compilation on Linux with artifacts (#2125)
• 8d2ec69f4f685b0ab1c4624d59ee2d3287bb3147 [python][R][docs][ci] better compatibility with Visual Studio 2019 (#2083)
• de82ed585145414390cc8f9f6aa061566f732d97 [SWIG] added typemap for string arrays in SWIG interface (#2118)
• c45ec4f1535539caf099a883f0505a5856593c9d dump_model() bug with num_leaves=1
• 24ad35f748724a341c3318f136948dbd5c7263ff fixed minor typos (#2119)
• 044a39d0ee7b3d3d91414f3ada281f271a90587f [ci] hotfixes (#2132)
• 4be53a5a1f18892692138dfd3085190124bc1a3e [python] disable default pandas cat features if cat features were explicitly provided (#2121)
• d115769c2a2ddffadc76c7b84739a47937114c77 [python] ignore pandas ordered categorical columns by default (#2115)
• 89f2021a4d971284b1fb2dea5a94b8ec4725933b [docs] Update doc string for pred_contrib (#2116) [ #1969 ]
• beb35d567de899b140bd61e174ef3b9ef5fd0769 [docs] added note about the spoiled probabilities (#2113)
• 29617593375c06dba9ad7ca270484a6a1e88068f minor fixes in CMakeLists (#2099)
• 975d2641c59a53bd8d98d64797f989f6c7fbe5a6 Fixed typo in regression_metric.hpp (#2111)
• ca85b6795002cc0c74d8e97d19fee88ca6ecc98e [python] add flag of displaying train loss for lgb.cv() (#2089)
• 8ffd8d80e47e40ed6439123de85ab8bacffc1bb6 fix overflow in WeightedPercentile
• 5b5b98235e4fa8c1eed67f57bf5c409bbe1a09e0 [python] make possibility to create Booster from string official (#2098)
• 0a4a7a86f5a1d3146c36c7d8c082154a193d4893 fixed cpplint errors about spaces and newlines (#2102)
• 32ef7603aba858b1e3dbf98e23d99a4cb2ee99b8 added copyright message in files (#2101)
• 50ce01b55e6bbac1dd00e2f9b1b639dcc37f2660 reworked includes in source files (#2066)
• c56412a859d4968f2b720514306be3404552b385 [docs] updated HDFS guide (#1890)
• fb9b2b8af61806343586aaf74be814cc8c9e977d add IntelliJ IDEA to .gitignore (#2093)
• 0e025e9c223023e49f90b1d026b6eb287d125972 [ci] update macOS on Travis to Mojave (#2086)
• b3c31c40151bcefe825a7ed5ac1ae28ff737a57e [docs] Python wrapper doesn't support params in form of list of pairs (#2078)
• 691b842832f511a72b4aaab715b3cb6b00de8f90 [ci] update CI stuff (#2079)
• 76102284d1997ef8b876cb40ae7c3e44b6ffbb66 Add Cost Effective Gradient Boosting (#2014)
• fe115bbb7215d3cf966ee406a017f208b5610900 [docs] Fix typo in Python-Intro.rst (#2074)
• 2027f6b485b1b707b774ca6cfbf6142aa71e8665 addressed cpplint error about C-style cast (#2064)
• 82f803b3df65f840b757b023e661e484ad9ed789 updated gitignore to ignore files created by local python installation (#2061)
• 572ae4003897c7435999cbc68d0a2c08e503694c [docs] Small aesthetic improvements to RTD docs (#2060)
• 3c999be3b7bd2bf8e487957cce8c8f9a1ae19b6a fixed cpplint error about spaces and newlines (#2068)
• 823fc03cc820758fceae91eb33851e1534b65360 Add API method LGBM_BoosterPredictForMats (#2008)
• 011cc90a770482e835e613fdbcb2ceb9f3adcef2 [python] Use first_metric_only flag for early_stopping function. (#2049)
• 548fc91e59a21254b4ec6cb59fa4ee006476190f remove warnings
• e5e9fbea5e8170d93422ed32a34aff8b31851998 fixed broken link (#2065)
• 7a89d0054bcf0db9f96af3c53a078b6aa3152efe removed Python 3.4 support (#2051)
• babe8b313aad747899a3199fba7bc3eae10e582a removed hotfix (#2054)
• beeb6e0fd5812065578be6969f3c82055e0fc957 Added additional APIs to better support JNI on Spark (#2032)
• 95246cdaec4feb739f3725e3d0fbf2efeb37bc85 lightgbm SWIG Java wrapper changes needed to add early stopping in mmlspark (#2047)
• b020a25d3dbfcd9f1128352ca1582abdf67ff87b [ci] compatibility hotfix for notebook execution (#2048)
• 74ce2cfe9418b0def315c5e17f6244eefac64a60 added examples for multiple custom metrics (#2021)
• ffb134cc31cc6ec7a456257fcdae9054ded54559 [python] disabled split value histogram for categorical features (#2045)
• 7ab94e6bba68d4f75c864503fdb269a70def59e6 [examples] updated tree index with categorical feature (#2044)
• 8d6666e0ffb7165096753dc392044af18ef2eae6 [python] added get_split_value_histogram method (#2041)
• 9de526f8acc36f493252c635880bca043e643ae3 Fix typo in GatherInfoForThresholdCategorical (#2040)
• 5801460a355faaf96ae16d152df7dd8c61202e4e [python] Minor fix: close open file pointer (#2039)
• 8aa08c4af22705338adba9ffbd61a73b8bc7baad [tests] fixed and refactored some tests (#2035)
• 219c943dd7983ee006772cf2cff52a1207bc8002 Add ability to move features from one data set to another in memory (#2006)
• f9ab5f588d6bbe145b0aacc28c4aa8711f116caf [docs] added notes about params usage when data is provided via path and removed unused param (#2024)
• c5cfe3e3e4579a3490989cfac9e6460394fdafee [python] update DataTable handling (#2020)
• 414bb609c0d8c97091627a77ab49a1647825f1a9 added LightGBM SWIG wrappers for macOS and updated docs (#2002)
• 7ebf80f8cf8afe2d2dfc76acf422084a8ad594c8 Fix wording (#2015)
• a777aedd9d2885d1adbe3ca6c623198dd916aaaa Change variable name test_data > validation_data (#2018)
• b447a7bc341d07602f39d664d22f9c09dd09a941 fixed conda (#2016)
• 5256cda69300d6b83b18180da2992a1e50a6b392 Fix #1988 missing file issue (#2000)
• 462612b42c6fe7bb1be1ccd8b72d77021b9f2d38 fixed modifiers indent (#1997)
• 8e286b38a3f597bc7fbe2905b8d71ea20340fe71 update version number at master branch (#1996)

This list of changes was auto generated.