Share
Embed
学習・推論パイプラインを構築する上で大切にしていること
kaggleやSIGNATEなどの分析コンペで使用する学習・推論パイプラインを構築する上で大切にしていること
Takanobu Nozawa
November 30, 2019
Transcript
81 slides
-
ΞδΣϯμ ࣗݾհ ύΠϓϥΠϯ͕ͳ͔ͬͨ࣌ͷπϥϛ େʹ͍ͯ͠Δ͜ͱ ‣ ֶशͷ࠶ݱੑ
‣ 1%$"ͷߴԽ ‣ ҙࣝ͠ͳͯ͘ཧͰ͖ΔΈ ࣮ৄࡉ -
ࣗݾհ ໊લɿᖒরʢ/P[BXB5BLBOPCVʣ ॴଐɿίωώτגࣜձࣾ ɹɹɿ͔ͨͺ͍!UBLBQZ w ʙίωώτʹ.-ΤϯδχΞͱͯ͠+0*/ w ػցֶशʢ/-1ɺਪનγεςϜʣΛϝΠϯʹΓͭͭΠϯϑϥʢ"84ʣษڧத w
,BHHMFͨ͠ΓɺϒϩάʢIUUQTXXXUBLBQZXPSLʣॻ͍ͨΓɺٿͨ͠Γɺ ϥʔϝϯ৯ͨΓ͍ͯ͠·͢ w ΦεεϝͷυϥϜࣜચ୕ػ͋ͬͨΒڭ͍͑ͯͩ͘͞ -
ࣗݾհ ˞ʮӾཡʯʮར༻ऀʯϝσΟΞͱΞϓϦͷ߹ܭʢ݄݄ͷฏۉʣ ˞ʮϚϚ͚/PΞϓϦʯ݄Πϯςʔδௐɹௐࠪରɿ৷தʙ̎ࡀ̌ϲ݄ͷࢠڙΛ࣋ͭঁੑ O Λநग़ ˞*OTUBHSBNͷϑΥϩϫʔɺ'BDFCPPLͷ͍͍Ͷɺ-*/&ͷͱͩͪͷ߹ܭ ݄࣌
ϚϚϦ ΞϓϦɾ8FC 4/4 *OTUBHSBNɾ-*/&ɾ'BDFCPPL هࣄ ϚϚಉ࢜ͰΈΛ૬ஊ͠߹͏2"ίϛϡχςΟΛத৺ʹ ϢʔβʔΛ֦େ͍ͯ͠·͢ ʮϚϚϦʯͰϢʔβʔಉ͕࢜ ͲΜͲΜܨ͕͍ͬͯ·͢ ϚϚͷੜ׆ʹཱͭهࣄΛ ෯͍δϟϯϧͰ৴͍ͯ͠·͢ ϚϚ͚/P̍ΞϓϦʹબग़ ਓͷϚϚ͕બͿʮݱࡏ͍ͬͯΔΞϓϦʯʹ ͯɺ߲ ଞͷϚϚʹΦεεϝ͍ͨ͠ɺೝɺ ར༻ɺརศੑɺײ Ͱ̍ҐΛ֫ಘ͠·ͨ͠ هࣄ 6,000 هࣄҎ্ ྦྷܭϑΝϯ 85 ສਓ ˞ ݄ؒӾཡ 1.5ԯճ ˞ ݄ؒར༻ऀ 650ສਓ ˞ ˞ lΈzͱzڞײzΛ࣠ʹϚϚʹدΓఴ͍ ΞϓϦɾ8FCɾ4/4ͱଟ֯తʹαʔϏεΛల։͍ͯ͠·͢ -
ࣗݾհ 0 450,000 900,000 1,350,000 1,800,000 2014/4 2014/5 2014/6
2014/7 2014/8 2014/9 2014/10 2014/11 2014/12 2015/1 2015/2 2015/3 2015/4 2015/5 2015/6 2015/7 2015/8 2015/9 2015/10 2015/11 2015/12 2016/1 2016/2 2016/3 2016/4 2016/5 2016/6 2016/7 2016/8 2016/9 2016/10 2016/11 2016/12 2017/1 2017/2 2017/3 2017/4 2017/5 2017/6 2017/7 2017/8 2017/9 2017/10 2017/11 2017/12 2018/1 2018/2 2018/3 2018/4 2018/5 2018/6 2018/7 2018/8 ݄ؒߘ 150ສ݅ िʹҎ্ىಈ͢Δ ΞΫςΟϒϢʔβʔ 50 ਓʹਓ 57$. ์ө ΞϓϦ૯%-ສ ਓʹਓ ਓʹਓ ਓʹਓ ਓʹਓ ˞ ˞ʮϚϚϦʯͷग़࢈༧ఆΛઃఆͨ͠Ϣʔβʔͱɺްੜ࿑ಇলൃදʮਓޱಈଶ౷ܭʯͷग़ੜ͔Βࢉग़ ˞िʹճҎ্ىಈ͢ΔϢʔβʔ ˞ ʹग़࢈ͨ͠ϚϚͷʮਓʹਓʯ͕ϚϚϦΛར༻த ຊ࠷େڃنΛތΔϒϥϯυͱ͍ͯ͠·͢ ˞ -
ύΠϓϥΠϯ͕ͳ͔ͬͨ࣌ͷπϥϛ <> import numpy as np import pandas as
pd OPUFCPPLͷத ɾ ɾ ɾ <> hogehoge <> hogehoge <> hogehoge ɾ ɾ ɾ -
ύΠϓϥΠϯ͕ͳ͔ͬͨ࣌ͷπϥϛ <> import numpy as np import pandas as
pd OPUFCPPLͷத ɾ ɾ ɾ <> hogehoge <> hogehoge <> hogehoge ɾ ɾ ɾ ηϧͷ࣮ߦॱ͕ͪ͝Όͪ͝Ό ˣ ࠶ݱੑ͕ͳ͍ -
ύΠϓϥΠϯ͕ͳ͔ͬͨ࣌ͷπϥϛ <> import numpy as np import pandas as
pd OPUFCPPLͷத ɾ ɾ ɾ <> submission.to_csv('submission.csv', index=False) -
ύΠϓϥΠϯ͕ͳ͔ͬͨ࣌ͷπϥϛ <> import numpy as np import pandas as
pd OPUFCPPLͷத ɾ ɾ ɾ <> submission.to_csv('submission.csv', index=False) ηϧ͕ଟ͘ɺಉ͡ܭࢉΛෳ࣮ߦ͠ͳ͚ΕͳΒͳ͍ ˣ 1%$"͕͘ͳΓ͕ͪ -
ύΠϓϥΠϯ͕ͳ͔ͬͨ࣌ͷπϥϛ dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC
dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC ʜʜʜʜʜ -
ύΠϓϥΠϯ͕ͳ͔ͬͨ࣌ͷπϥϛ dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC dOPUFCPPL-JHIU#(.@TDPSF@$PQZJQZOC
dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC dOPUFCPPL-JHIU#(.@TDPSF@JQZOC ʜʜʜʜʜ OPUFCPPL͕ࡍݶͳ͘૿৩͠ ཧ͕ΊͪΌͪ͘Όࡶʹ -
େʹ͍ͯ͠Δ͜ͱ w ͭͷQZUIPOεΫϦϓτͰ%0//"ΛঠѲ ༻͢Δಛྔ ༻͢Δύϥϝʔλ ༻͢Δ$7
ʙߴͳ1%$"Λࢦͯ͠ʙ -
େʹ͍ͯ͠Δ͜ͱ w ͭͷQZUIPOεΫϦϓτͰ%0//"ΛঠѲ ༻͢Δಛྔ ༻͢Δύϥϝʔλ ༻͢Δ$7
w GFBUVSFJNQPSUBODFͷݟ͑ΔԽ ࣍ͷ࣮ݧͷצॴΛ͔ͭΊΔΑ͏ʹ ʙߴͳ1%$"Λࢦͯ͠ʙ -
େʹ͍ͯ͠Δ͜ͱ ʙҙࣝ͠ͳͯ͘ཧͰ͖ΔΈʙ w %0///" ϞσϧϩάϑΝΠϧΛɺ࣮ߦ͝ͱʹʮ࣌ʯ TV⒏YΛ͚ͭͨσΟϨΫτϦɾϑΝΠϧͰࣗಈཧ ༻͢Δಛྔ ύϥϝʔλ
$7 ֶशͨ͠Ϟσϧ ֶशϩάϑΝΠϧ GFBUVSFJNQPSUBODF -
େʹ͍ͯ͠Δ͜ͱ ʙҙࣝ͠ͳͯ͘ཧͰ͖ΔΈʙ w %0///" ϞσϧϩάϑΝΠϧΛɺ࣮ߦ͝ͱʹʮ࣌ʯ TV⒏YΛ͚ͭͨσΟϨΫτϦɾϑΝΠϧͰࣗಈཧ ༻͢Δಛྔ ύϥϝʔλ
$7 ֶशͨ͠Ϟσϧ ֶशϩάϑΝΠϧ GFBUVSFJNQPSUBODF w 1VCMJDείΞΛσΟϨΫτϦͷ1SFpYʹ͚Δ͜ͱͰɺ 1VCMJDͱMPDBMͷείΞͷରԠΛ͔Γ͘͢ʢৄࡉޙ΄Ͳʣ -
େʹ͍ͯ͠Δ͜ͱ ʙҙࣝ͠ͳͯ͘ཧͰ͖ΔΈʙ w %0///" ϞσϧϩάϑΝΠϧΛɺ࣮ߦ͝ͱʹʮ࣌ʯ TV⒏YΛ͚ͭͨσΟϨΫτϦɾϑΝΠϧͰࣗಈཧ ༻͢Δಛྔ ύϥϝʔλ
$7 ֶशͨ͠Ϟσϧ ֶशϩάϑΝΠϧ GFBUVSFJNQPSUBODF w 1VCMJDείΞΛσΟϨΫτϦͷ1SFpYʹ͚Δ͜ͱͰɺ 1VCMJDͱMPDBMͷείΞͷରԠΛ͔Γ͘͢ʢৄࡉޙ΄Ͳʣ ৄࡉΛݟ͍͖ͯ·͢ -
ྫʣIPHFQZ features = [ "age", "pclass", "family_size", "fare", "sibsp",
"parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } runner = Runner(run_name, ModelLGB, features, dataset.get('target'), params_lgb, cv, FEATURE_DIR_NAME, MODEL_DIR_NAME) runner.run_train_cv() # ֶश runner.run_predict_cv() # ਪ Submission.create_submission(run_name) # submit࡞ -
ྫʣIPHFQZ features = [ "age", "pclass", "family_size", "fare", "sibsp",
"parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } runner = Runner(run_name, ModelLGB, features, dataset.get('target'), params_lgb, cv, FEATURE_DIR_NAME, MODEL_DIR_NAME) runner.run_train_cv() # ֶश runner.run_predict_cv() # ਪ Submission.create_submission(run_name) # submit࡞ جຊతʹ1%$"Λճ͢ͱ͖ʹ͍͡Δͷ ͜ͷIPHFQZͷΈʹ͢Δ -
ྫʣIPHFQZ features = [ "age", "pclass", "family_size", "fare", "sibsp",
"parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } runner = Runner(run_name, ModelLGB, features, dataset.get('target'), params_lgb, cv, FEATURE_DIR_NAME, MODEL_DIR_NAME) runner.run_train_cv() # ֶश runner.run_predict_cv() # ਪ Submission.create_submission(run_name) # submit࡞ ֶशʹ༻͢Δಛྔ -
ྫʣIPHFQZ features = [ "age", "pclass", "family_size", "fare", "sibsp",
"parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } runner = Runner(run_name, ModelLGB, features, dataset.get('target'), params_lgb, cv, FEATURE_DIR_NAME, MODEL_DIR_NAME) runner.run_train_cv() # ֶश runner.run_predict_cv() # ਪ Submission.create_submission(run_name) # submit࡞ ϋΠύʔύϥϝʔλ -
ྫʣIPHFQZ features = [ "age", "pclass", "family_size", "fare", "sibsp",
"parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } runner = Runner(run_name, ModelLGB, features, dataset.get('target'), params_lgb, cv, FEATURE_DIR_NAME, MODEL_DIR_NAME) runner.run_train_cv() # ֶश runner.run_predict_cv() # ਪ Submission.create_submission(run_name) # submit࡞ $7ͷઃఆ -
ྫʣIPHFQZ features = [ "age", "pclass", "family_size", "fare", "sibsp",
"parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } runner = Runner(run_name, ModelLGB, features, dataset.get('target'), params_lgb, cv, FEATURE_DIR_NAME, MODEL_DIR_NAME) runner.run_train_cv() # ֶश runner.run_predict_cv() # ਪ Submission.create_submission(run_name) # submit࡞ ֶशɾਪɾTVC࡞ -
ྫʣIPHFQZ features = [ "age", "pclass", "family_size", "fare", "sibsp",
"parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } runner = Runner(run_name, ModelLGB, features, dataset.get('target'), params_lgb, cv, FEATURE_DIR_NAME, MODEL_DIR_NAME) runner.run_train_cv() # ֶश runner.run_predict_cv() # ਪ Submission.create_submission(run_name) # submit࡞ ֶशɾਪɾTVC࡞ w ࠶ݱੑΛ୲อ͢Δ w ߴͳ1$%"Λճͨ͢Ίͷ w ҙࣝ͠ͳͯ͘ॾʑ͕ཧͰ͖Δ ʹ͍͓ͭͯ͠͠·͢ -
࣮ৄࡉ IPHFQZ ʙֶशͷ࠶ݱੑʙ features = [ "age", "pclass", "family_size",
"fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } -
࣮ৄࡉ IPHFQZ ʙֶशͷ࠶ݱੑʙ features = [ "age", "pclass", "family_size",
"fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } QZUIPOIPHFQZ -
࣮ৄࡉ IPHFQZ ʙֶशͷ࠶ݱੑʙ features = [ "age", "pclass", "family_size",
"fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } { "use_features": [ "age", "pclass", "family_size", "fare", "sibsp", "parch", "cabin" ], "model_params": { "boosting_type": "gbdt", "objective": "fair", "metric": "fair", "num_boost_round": 20000, "early_stopping_rounds": 1000, "verbose": 1000, "random_state": 999 }, "cv": { "method": "KFold", "n_splits": 5, "random_state": 42, "shuffle": true }, "dataset": { "run_name": "lgb_1128_2003", "feature_directory": "../data/features/remove_outlier/", "target": "salary" } } IPHF@QBSBNKTPO QZUIPOIPHFQZ -
࣮ৄࡉ IPHFQZ ʙֶशͷ࠶ݱੑʙ features = [ "age", "pclass", "family_size",
"fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } { "use_features": [ "age", "pclass", "family_size", "fare", "sibsp", "parch", "cabin" ], "model_params": { "boosting_type": "gbdt", "objective": "fair", "metric": "fair", "num_boost_round": 20000, "early_stopping_rounds": 1000, "verbose": 1000, "random_state": 999 }, "cv": { "method": "KFold", "n_splits": 5, "random_state": 42, "shuffle": true }, "dataset": { "run_name": "lgb_1128_2003", "feature_directory": "../data/features/remove_outlier/", "target": "salary" } } IPHF@QBSBNKTPO QZUIPOIPHFQZ IPHFQZΛ࣮ߦ͢Δ͜ͱʹΑΓ ࣗಈతʹKTPOϑΝΠϧ͕ੜ͞Ε ༻ͨ͠ಛྔɾύϥϝʔλʔͳͲ͕શͯอଘ͞ΕΔ -
࣮ৄࡉ IPHFQZ ʙֶशͷ࠶ݱੑʙ features = [ "age", "pclass", "family_size",
"fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } { "use_features": [ "age", "pclass", "family_size", "fare", "sibsp", "parch", "cabin" ], "model_params": { "boosting_type": "gbdt", "objective": "fair", "metric": "fair", "num_boost_round": 20000, "early_stopping_rounds": 1000, "verbose": 1000, "random_state": 999 }, "cv": { "method": "KFold", "n_splits": 5, "random_state": 42, "shuffle": true }, "dataset": { "run_name": "lgb_1128_2003", "feature_directory": "../data/features/remove_outlier/", "target": "salary" } } IPHF@QBSBNKTPO QZUIPOIPHFQZ ͜Εͧ࠶ݱੑ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ ಛྔΛݮΒֶͯ͠श͍ͤͨ͞ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ features = [ "age", "pclass", # "family_size", "fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } IPHFQZ ରͷಛྔΛίϝϯτΞτ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ features = [ "age", "pclass", # "family_size", "fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } IPHFQZ QZUIPOIPHFQZ ରͷಛྔΛίϝϯτΞτ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ features = [ "age", "pclass", # "family_size", "fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } IPHFQZ QZUIPOIPHFQZ ରͷಛྔΛίϝϯτΞτ ؆୯ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ $7มֶ͑ͯश͍ͤͨ͞ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ features = [ "age", "pclass", "family_size", "fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'GroupKFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, 'cv_target':'user' } IPHFQZ DWͷهࡌΛมߋ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ features = [ "age", "pclass", "family_size", "fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'GroupKFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, 'cv_target':'user' } IPHFQZ DWͷهࡌΛมߋ QZUIPOIPHFQZ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ features = [ "age", "pclass", "family_size", "fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'GroupKFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, 'cv_target':'user' } IPHFQZ DWͷهࡌΛมߋ QZUIPOIPHFQZ ؆୯ -
࣮ৄࡉ IPHFQZ features = [ "age", "pclass", "family_size", "fare",
"sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'KFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, } ʙߴͳ1%$"Λࢦͯ͠ʙ features = [ "age", "pclass", "family_size", "fare", "sibsp", "parch", "cabin" ] params_lgb = { 'boosting_type': 'gbdt', 'objective': 'fair', 'metric': 'fair', 'num_boost_round': 20000, 'early_stopping_rounds': 1000, 'verbose': 1000, 'random_state': 999 } cv = { 'method': 'GroupKFold', 'n_splits': 5, 'random_state': 42, 'shuffle': True, 'cv_target':'user' } IPHFQZ DWͷهࡌΛมߋ QZUIPOIPHFQZ GFBUVSFJNQPSUBODFʹ͍ͭͯ -
࣮ৄࡉ IPHFQZΛ࣮ߦͨ͋͠ͱͷϑΥϧμɾϑΝΠϧ ʙҙࣝ͠ͳͯ͘ཧͰ͖ΔΈʙ w ಛྔ w ύϥϝʔλ w GFBUVSFJNQPSUBODF
w ֶशϩά w Ϟσϧ w ਪϑΝΠϧ ͕อଘ͞ΕΔ -
࣮ৄࡉ IPHFQZΛ࣮ߦͨ͋͠ͱͷϑΥϧμɾϑΝΠϧ ʙҙࣝ͠ͳͯ͘ཧͰ͖ΔΈʙ w ಛྔ w ύϥϝʔλ w GFBUVSFJNQPSUBODF
w ֶशϩά w Ϟσϧ w ਪϑΝΠϧ ͕อଘ͞ΕΔ ҙࣝͯ͠ͳ͍͚ͲউखʹཧͰ͖ͯΔ -
࣮ৄࡉ IPHFQZΛ࣮ߦͨ͋͠ͱͷϑΥϧμɾϑΝΠϧ ʙҙࣝ͠ͳͯ͘ཧͰ͖ΔΈʙ w ಛྔ w ύϥϝʔλ w GFBUVSFJNQPSUBODF
w ֶशϩά w Ϟσϧ w ਪϑΝΠϧ ͕อଘ͞ΕΔ -
·ͱΊ w ֶशɾਪύΠϓϥΠϯ͍͍ͧʂ ύΠϓϥΠϯΛߏங͢Δ͜ͱͰԼهͷΑ͏ͳϝϦοτ͕͋Γʢࠓͷॴʣݸਓత ʹΊͬͪΌྑ͍ɻ ‣ ࠶ݱੑ ‣
ߴͳ1%$" ‣ ॾʑͷཧ w ্هͷΑ͏ͳ͜ͱ͕୲อ͞ΕΔͷͰ৺ཧత҆શੑ