ftnext
commented
3 years ago datasetはv1.8使用。lgbmのパラメタだけ変える [CV: 1.0309] [LB: 1.0064]
$ python training.py data/preprocessed/v1.8/ data/datasets/train.csv submissions/ --y_max 7.5
12026it [00:52, 229.15it/s]
12008it [00:49, 241.13it/s]
train: (12026, 1306)
test: (12008, 1306)
[LightGBM] [Warning] bagging_freq is set=1, subsample_freq=3 will be ignored. Current value: bagging_freq=1
[LightGBM] [Warning] boosting is set=gbdt, boosting_type=gbdt will be ignored. Current value: boosting=gbdt
[LightGBM] [Warning] bagging_fraction is set=0.8, subsample=0.9 will be ignored. Current value: bagging_fraction=0.8
[LightGBM] [Warning] min_data_in_leaf is set=64, min_child_samples=10 will be ignored. Current value: min_data_in_leaf=64
[LightGBM] [Warning] num_threads is set=6, n_jobs=-1 will be ignored. Current value: num_threads=6
Training until validation scores don't improve for 100 rounds
[500] valid_0's rmse: 1.05699
Early stopping, best iteration is:
[401] valid_0's rmse: 1.05523
Fold 0 RMSLE: 1.0552
[LightGBM] [Warning] bagging_freq is set=1, subsample_freq=3 will be ignored. Current value: bagging_freq=1
[LightGBM] [Warning] boosting is set=gbdt, boosting_type=gbdt will be ignored. Current value: boosting=gbdt
[LightGBM] [Warning] bagging_fraction is set=0.8, subsample=0.9 will be ignored. Current value: bagging_fraction=0.8
[LightGBM] [Warning] min_data_in_leaf is set=64, min_child_samples=10 will be ignored. Current value: min_data_in_leaf=64
[LightGBM] [Warning] num_threads is set=6, n_jobs=-1 will be ignored. Current value: num_threads=6
Training until validation scores don't improve for 100 rounds
[500] valid_0's rmse: 1.03432
Early stopping, best iteration is:
[583] valid_0's rmse: 1.03322
Fold 1 RMSLE: 1.0332
[LightGBM] [Warning] bagging_freq is set=1, subsample_freq=3 will be ignored. Current value: bagging_freq=1
[LightGBM] [Warning] boosting is set=gbdt, boosting_type=gbdt will be ignored. Current value: boosting=gbdt
[LightGBM] [Warning] bagging_fraction is set=0.8, subsample=0.9 will be ignored. Current value: bagging_fraction=0.8
[LightGBM] [Warning] min_data_in_leaf is set=64, min_child_samples=10 will be ignored. Current value: min_data_in_leaf=64
[LightGBM] [Warning] num_threads is set=6, n_jobs=-1 will be ignored. Current value: num_threads=6
Training until validation scores don't improve for 100 rounds
[500] valid_0's rmse: 1.02079
Early stopping, best iteration is:
[433] valid_0's rmse: 1.01943
Fold 2 RMSLE: 1.0194
[LightGBM] [Warning] bagging_freq is set=1, subsample_freq=3 will be ignored. Current value: bagging_freq=1
[LightGBM] [Warning] boosting is set=gbdt, boosting_type=gbdt will be ignored. Current value: boosting=gbdt
[LightGBM] [Warning] bagging_fraction is set=0.8, subsample=0.9 will be ignored. Current value: bagging_fraction=0.8
[LightGBM] [Warning] min_data_in_leaf is set=64, min_child_samples=10 will be ignored. Current value: min_data_in_leaf=64
[LightGBM] [Warning] num_threads is set=6, n_jobs=-1 will be ignored. Current value: num_threads=6
Training until validation scores don't improve for 100 rounds
Early stopping, best iteration is:
[381] valid_0's rmse: 1.03247
Fold 3 RMSLE: 1.0325
[LightGBM] [Warning] bagging_freq is set=1, subsample_freq=3 will be ignored. Current value: bagging_freq=1
[LightGBM] [Warning] boosting is set=gbdt, boosting_type=gbdt will be ignored. Current value: boosting=gbdt
[LightGBM] [Warning] bagging_fraction is set=0.8, subsample=0.9 will be ignored. Current value: bagging_fraction=0.8
[LightGBM] [Warning] min_data_in_leaf is set=64, min_child_samples=10 will be ignored. Current value: min_data_in_leaf=64
[LightGBM] [Warning] num_threads is set=6, n_jobs=-1 will be ignored. Current value: num_threads=6
Training until validation scores don't improve for 100 rounds
Early stopping, best iteration is:
[377] valid_0's rmse: 1.01359
Fold 4 RMSLE: 1.0136
--------------------------------------------------
FINISHED | Whole RMSLE: 1.0309
features count: 1306
material__object_collection_vector_6 221840.04009914398
material_vector_1 186767.52309298515
material__object_collection_vector_8 96709.62258911133
material__object_collection_vector_17 49002.177185058594
size_h 22639.34058737755
material__object_collection_vector_0 21404.48133277893
CE__acquisition_date 16847.672721266747
material_vector_14 16058.157525897026
size_w 15930.997326374054
StringLength__more_title 12731.860483169556
dating_year_early 11593.119365692139
dating_year_late 11371.020359694958
StringLength__title 9280.47452813387
description_tfidf_5 8717.001561760902
description_tfidf_16 8351.599660754204
StringLength__description 7823.619671046734
CE__principal_maker 7816.960752725601
description_tfidf_47 7605.430411398411
description_tfidf_1 7490.088755488396
CE__dating_period 7465.452939867973
StringLength__long_title 7199.282612085342
CE__dating_year_late 6992.813401222229
CE__dating_presenting_date 6899.057689130306
CE__dating_sorting_date 6852.647631645203
title__lang=__label__en 5684.905223488808
description_tfidf_27 5631.3304438591
description_tfidf_9 5512.515014529228
description_tfidf_31 5493.111938238144
description_tfidf_22 5475.718765079975
description_tfidf_38 5465.606556653976
StringLength__sub_title 5442.212498664856
description_tfidf_10 5240.814885079861
description_tfidf_36 5171.609034180641
description_tfidf_39 5044.674207150936
description_tfidf_41 4914.697444200516
description_tfidf_33 4889.5087404847145
description_tfidf_29 4856.150610268116
description_tfidf_28 4845.901842772961
description_tfidf_15 4813.795782387257
description_tfidf_7 4758.510672569275
description_tfidf_46 4656.812683224678
CE__principal_or_first_maker 4651.718207001686
description_tfidf_18 4541.441883683205
description_tfidf_45 4498.1753224134445
description_tfidf_49 4422.603837788105
CE__acquisition_method 4383.243691205978
description_tfidf_42 4222.123824596405
description_tfidf_35 4172.149320423603
description_tfidf_48 4026.278638601303
description_tfidf_21 3892.9054334163666
https://www.guruguru.science/competitions/16/discussions/8d476062-3058-45a3-8a8c-d2d4973862b5/ の設定で上書きして駆け込んだ