diff --git a/benchmarks/fev_bench/results/t0-beta.csv b/benchmarks/fev_bench/results/t0-beta.csv new file mode 100644 index 0000000..5a5f1ca --- /dev/null +++ b/benchmarks/fev_bench/results/t0-beta.csv @@ -0,0 +1,101 @@ +model_name,dataset_path,dataset_config,horizon,num_windows,initial_cutoff,window_step_size,min_context_length,max_context_length,seasonality,eval_metric,extra_metrics,quantile_levels,id_column,timestamp_column,target,generate_univariate_targets_from,known_dynamic_columns,past_dynamic_columns,static_columns,task_name,test_error,training_time_s,inference_time_s,num_forecasts,dataset_fingerprint,trained_on_this_dataset,fev_version,SQL,MASE,WAPE,WQL,model_class,model_kwargs +t0-beta,autogluon/fev_datasets,proenfo_gfc12,168,10,-1680,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,['airtemperature'],[],[],proenfo_gfc12,0.6366597782763881,0.0,8.833002624975052,110,70e6cce11b73757a,False,0.11.0.dev0,0.6366597782763881,0.7956725276410296,0.06525336284385934,0.05201745498944264,t0-beta,{} +t0-beta,autogluon/fev_datasets,proenfo_gfc14,168,20,-3360,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,['airtemperature'],[],[],proenfo_gfc14,0.3935554763807748,0.0,1.8241172068810556,20,87312edc76952605,False,0.11.0.dev0,0.3935554763807748,0.4837333668031502,0.023659331179427465,0.019247029404497025,t0-beta,{} +t0-beta,autogluon/fev_datasets,proenfo_gfc17,168,20,-3360,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,['airtemperature'],[],[],proenfo_gfc17,0.498504508667278,0.0,11.911711042979732,160,897258c227dde589,False,0.11.0.dev0,0.498504508667278,0.6208445543955393,0.039830467401534175,0.03209825370391747,t0-beta,{} +t0-beta,autogluon/fev_datasets,rohlik_sales_1D,14,1,2023-12-15T00:00:00,14,14,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,sales,,"['holiday', 'school_holidays', 'sell_price_main', 'shops_closed', 'total_orders', 'type_0_discount', 'type_1_discount', 'type_2_discount', 'type_3_discount', 'type_4_discount', 'type_5_discount', 'type_6_discount', 'winter_school_holidays']",['availability'],"['L1_category_name_en', 'L2_category_name_en', 'L3_category_name_en', 'L4_category_name_en', 'name', 'product_unique_id', 'warehouse']",rohlik_sales_1D,0.9342952341996246,0.0,331.35341158299707,4116,2c9ecdd477948cce,False,0.11.0.dev0,0.9342952341996246,1.161468881758514,0.2946363496183804,0.2358309775595918,t0-beta,{} +t0-beta,autogluon/fev_datasets,rohlik_orders_1D,61,5,2023-05-01T00:00:00,61,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,orders,,"['holiday', 'school_holidays', 'shops_closed', 'winter_school_holidays']","['blackout', 'frankfurt_shutdown', 'mini_shutdown', 'mov_change', 'precipitation', 'shutdown', 'snow', 'user_activity_1', 'user_activity_2']",[],rohlik_orders_1D,1.0600102326433887,0.0,2.5764855409797747,35,d4ccf61e0ac3ccdf,False,0.11.0.dev0,1.0600102326433887,1.32418839218739,0.061696975494407544,0.04923991827972661,t0-beta,{} +t0-beta,autogluon/fev_datasets,rohlik_sales_1W,8,1,2023-12-15T00:00:00,8,8,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,sales,,"['holiday', 'school_holidays', 'sell_price_main', 'shops_closed', 'total_orders', 'type_0_discount', 'type_1_discount', 'type_2_discount', 'type_3_discount', 'type_4_discount', 'type_5_discount', 'type_6_discount', 'winter_school_holidays']",['availability'],"['L1_category_name_en', 'L2_category_name_en', 'L3_category_name_en', 'L4_category_name_en', 'name', 'product_unique_id', 'warehouse']",rohlik_sales_1W,1.2522045205732975,0.0,88.17846975001157,3942,149ec34eb852d454,False,0.11.0.dev0,1.2522045205732975,1.5336419680848363,0.23652876328585093,0.18948541625820572,t0-beta,{} +t0-beta,autogluon/fev_datasets,rohlik_orders_1W,8,5,2023-05-01T00:00:00,8,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,orders,,"['holiday', 'school_holidays', 'shops_closed', 'winter_school_holidays']","['blackout', 'frankfurt_shutdown', 'mini_shutdown', 'mov_change', 'precipitation', 'shutdown', 'snow', 'user_activity_1', 'user_activity_2']",[],rohlik_orders_1W,1.2976465876345031,0.0,0.8687145420117304,35,4eea90b920dec215,False,0.11.0.dev0,1.2976465876345031,1.6212729739829872,0.052498519713824965,0.04193894025027729,t0-beta,{} +t0-beta,autogluon/fev_datasets,entsoe_15T,96,20,-1920,96,1,,96,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['radiation_diffuse_horizontal', 'radiation_direct_horizontal', 'temperature']",[],[],entsoe_15T,0.4727302393367692,0.0,20.093112039961852,120,8facdab2c9ff5bd5,False,0.11.0.dev0,0.4727302393367692,0.6034017769029937,0.03974237117236534,0.031343568108642965,t0-beta,{} +t0-beta,autogluon/fev_datasets,entsoe_30T,96,20,-1920,96,1,,48,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['radiation_diffuse_horizontal', 'radiation_direct_horizontal', 'temperature']",[],[],entsoe_30T,0.45767355765659257,0.0,20.332424126041587,120,98ab81cde986fe22,False,0.11.0.dev0,0.45767355765659257,0.5686110197432918,0.034519825866322906,0.028023053541845165,t0-beta,{} +t0-beta,autogluon/fev_datasets,entsoe_1H,168,20,-3360,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['radiation_diffuse_horizontal', 'radiation_direct_horizontal', 'temperature']",[],[],entsoe_1H,0.43132774609880087,0.0,20.211779332981678,120,c1b23331fc6093bc,False,0.11.0.dev0,0.43132774609880087,0.5252216417303138,0.034218416042266955,0.028512594658995215,t0-beta,{} +t0-beta,autogluon/fev_datasets,epf_be,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['Generation forecast', 'System load forecast']",[],[],epf_be,0.5152436924036575,0.0,2.5605855829489883,20,2b9df413e5ec356e,False,0.11.0.dev0,0.5152436924036575,0.6471340598590057,0.11497250523257785,0.09165872982594452,t0-beta,{} +t0-beta,autogluon/fev_datasets,epf_de,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['Ampirion Load Forecast', 'PV+Wind Forecast']",[],[],epf_de,0.44813265193181173,0.0,2.5188367489899974,20,bc6dede7cf6a507b,False,0.11.0.dev0,0.44813265193181173,0.5596997159858976,0.2638035927811763,0.21479020984774794,t0-beta,{} +t0-beta,autogluon/fev_datasets,epf_fr,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['Generation forecast', 'System load forecast']",[],[],epf_fr,0.3693032932173393,0.0,2.5423249159939587,20,a00b09fd79e0529e,False,0.11.0.dev0,0.3693032932173393,0.46014007846037747,0.07021172413467666,0.05639297343896018,t0-beta,{} +t0-beta,autogluon/fev_datasets,epf_np,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['Grid load forecast', 'Wind power forecast']",[],[],epf_np,0.5720786373485183,0.0,2.5201210010272916,20,9830dda233defc6c,False,0.11.0.dev0,0.5720786373485183,0.6930104813756149,0.03163166621520988,0.02609169905275464,t0-beta,{} +t0-beta,autogluon/fev_datasets,epf_pjm,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['System load forecast', 'Zonal COMED load foecast']",[],[],epf_pjm,0.3879626645332669,0.0,2.5163247069576755,20,9228310bf7192a0a,False,0.11.0.dev0,0.3879626645332669,0.48403353587986675,0.08432823281329149,0.06773144650038498,t0-beta,{} +t0-beta,autogluon/fev_datasets,rossmann_1D,48,10,-480,48,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,Sales,,"['DayOfWeek', 'Open', 'Promo', 'SchoolHoliday', 'StateHoliday']",['Customers'],"['Assortment', 'CompetitionDistance', 'CompetitionOpenSinceMonth', 'CompetitionOpenSinceYear', 'Promo2', 'Promo2SinceWeek', 'Promo2SinceYear', 'PromoInterval', 'Store', 'StoreType']",rossmann_1D,0.2863708513676749,0.0,304.41215362399817,11150,83a79b0a9a999f56,False,0.11.0.dev0,0.2863708513676749,0.3634704077783299,0.12216252809077321,0.09623069324712663,t0-beta,{} +t0-beta,autogluon/fev_datasets,rossmann_1W,13,8,-104,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,Sales,,"['Open', 'Promo', 'SchoolHoliday', 'StateHoliday']",['Customers'],"['Assortment', 'CompetitionDistance', 'CompetitionOpenSinceMonth', 'CompetitionOpenSinceYear', 'Promo2', 'Promo2SinceWeek', 'Promo2SinceYear', 'PromoInterval', 'Store', 'StoreType']",rossmann_1W,0.36729197673563074,0.0,60.050578291062266,8920,28f46b06b50c6d84,False,0.11.0.dev0,0.36729197673563074,0.4570436611336681,0.12410730258384664,0.09965306480854878,t0-beta,{} +t0-beta,autogluon/fev_datasets,restaurant,28,8,-224,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],"['air_area_name', 'air_genre_name', 'latitude', 'longitude']",restaurant,0.6763713374812443,0.0,14.204377791989828,6502,eb8da1769444e614,False,0.11.0.dev0,0.6763713374812443,0.8465584414225449,0.3557056967897792,0.28033092081053024,t0-beta,{} +t0-beta,autogluon/fev_datasets,hermes,52,1,-52,52,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,['external'],[],"['category', 'country']",hermes,0.674551707050121,0.0,35.746022125007585,10000,003815c209114c79,False,0.11.0.dev0,0.674551707050121,0.8560838229091269,0.0031366067727879175,0.002472608299298006,t0-beta,{} +t0-beta,autogluon/fev_datasets,walmart,39,1,-39,39,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['CPI', 'Fuel_Price', 'IsHoliday', 'MarkDown1', 'MarkDown2', 'MarkDown3', 'MarkDown4', 'MarkDown5', 'Temperature', 'Unemployment']",[],"['Dept', 'Size', 'Store', 'Type']",walmart,0.759162032229795,0.0,54.80967762498767,2936,5de0711be36aab0b,False,0.11.0.dev0,0.759162032229795,0.9556124742805552,0.1106502715113281,0.08823557062642602,t0-beta,{} +t0-beta,autogluon/fev_datasets,m5_1D,28,1,-28,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['event_Cultural', 'event_National', 'event_Religious', 'event_Sporting', 'sell_price', 'snap_CA', 'snap_TX', 'snap_WI']",[],"['cat_id', 'dept_id', 'item_id', 'state_id', 'store_id']",m5_1D,0.7145585269524386,0.0,2274.2238758750027,30490,91e836c1eb428f2d,False,0.11.0.dev0,0.7145585269524386,0.8747127345561839,0.7013676509010011,0.5513149902025548,t0-beta,{} +t0-beta,autogluon/fev_datasets,m5_1W,13,1,-13,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['event_Cultural', 'event_National', 'event_Religious', 'event_Sporting', 'sell_price', 'snap_CA', 'snap_TX', 'snap_WI']",[],"['cat_id', 'dept_id', 'item_id', 'state_id', 'store_id']",m5_1W,0.880462467221529,0.0,526.6490334589907,30490,bb4e7816349be551,False,0.11.0.dev0,0.880462467221529,1.1263467072550677,0.42153801677222,0.3314423138994695,t0-beta,{} +t0-beta,autogluon/fev_datasets,m5_1M,12,1,-12,12,6,,12,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['event_Cultural', 'event_National', 'event_Religious', 'event_Sporting', 'sell_price', 'snap_CA', 'snap_TX', 'snap_WI']",[],"['cat_id', 'dept_id', 'item_id', 'state_id', 'store_id']",m5_1M,0.9631682951827821,0.0,290.9466818750079,29364,4dcc997759fda726,False,0.11.0.dev0,0.9631682951827821,1.15952415128945,0.42880509715039233,0.34006958676445964,t0-beta,{} +t0-beta,autogluon/fev_datasets,hierarchical_sales_1D,28,10,-280,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],hierarchical_sales_1D,0.5497313807683833,0.0,9.486682126007508,1180,569b9f333dc43810,False,0.11.0.dev0,0.5497313807683833,0.6877640837094583,0.6950869067736081,0.5614659204053355,t0-beta,{} +t0-beta,autogluon/fev_datasets,hierarchical_sales_1W,13,10,-130,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],hierarchical_sales_1W,0.6243467524058355,0.0,2.007455375016434,1180,51efc7d99ad351b6,False,0.11.0.dev0,0.6243467524058355,0.7613567661734616,0.43782759983837255,0.37223393782858655,t0-beta,{} +t0-beta,autogluon/fev_datasets,favorita_stores_1D,28,10,-280,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,sales,,"['holiday', 'onpromotion']",['oil_price'],"['city', 'cluster', 'family', 'state', 'store_nbr', 'type']",favorita_stores_1D,0.9320230955957589,0.0,507.9860572499747,15790,2f332ce1cd71d139,False,0.11.0.dev0,0.9320230955957589,1.1562125656679818,0.14703948021761878,0.1186595393718887,t0-beta,{} +t0-beta,autogluon/fev_datasets,favorita_stores_1W,13,10,-130,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,sales,,['onpromotion'],['oil_price'],"['city', 'cluster', 'family', 'state', 'store_nbr', 'type']",favorita_stores_1W,2.0177392020639022,0.0,74.18426279199775,15790,34ca8727dcbb9c1a,False,0.11.0.dev0,2.0177392020639022,2.366424971313127,0.1267981418121593,0.10292378530307214,t0-beta,{} +t0-beta,autogluon/fev_datasets,favorita_stores_1M,12,2,-24,12,1,,12,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,sales,,['onpromotion'],['oil_price'],"['city', 'cluster', 'family', 'state', 'store_nbr', 'type']",favorita_stores_1M,1.8549120246796784,0.0,9.235417998977937,3158,838e35b99c2032ca,False,0.11.0.dev0,1.8549120246796784,2.219318447777495,0.2009299790443267,0.14243197168449293,t0-beta,{} +t0-beta,autogluon/fev_datasets,favorita_transactions_1D,28,10,-280,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,transactions,,['holiday'],['oil_price'],"['city', 'cluster', 'state', 'store_nbr', 'type']",favorita_transactions_1D,0.8465026494821026,0.0,11.965194081014488,510,48943abb349dc31b,False,0.11.0.dev0,0.8465026494821026,1.1103333203829695,0.07013885655412246,0.05625987656100132,t0-beta,{} +t0-beta,autogluon/fev_datasets,favorita_transactions_1W,13,10,-130,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,transactions,,[],['oil_price'],"['city', 'cluster', 'state', 'store_nbr', 'type']",favorita_transactions_1W,1.3755175806789683,0.0,1.8215780009923037,510,53825d45abbbebd4,False,0.11.0.dev0,1.3755175806789683,1.6051513180362118,0.061576057258205606,0.05189915296002386,t0-beta,{} +t0-beta,autogluon/fev_datasets,favorita_transactions_1M,12,2,-24,12,1,,12,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,transactions,,[],['oil_price'],"['city', 'cluster', 'state', 'store_nbr', 'type']",favorita_transactions_1M,1.1142655441985654,0.0,0.3190139159851242,102,ea60d62aee5c7662,False,0.11.0.dev0,1.1142655441985654,1.398227166805778,0.08108280713002766,0.06846856357316156,t0-beta,{} +t0-beta,autogluon/fev_datasets,solar_with_weather_15T,96,20,-1920,96,1,,96,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['day_length', 'humidity', 'pressure', 'rain_1h', 'snow_1h', 'temp', 'wind_speed']","['clouds_all', 'global_horizontal_irradiance']",[],solar_with_weather_15T,0.7774962132029672,0.0,7.6345733730413485,20,6e766232f6e4f0ac,False,0.11.0.dev0,0.7774962132029672,1.013653943921017,1.2207444505700464,0.9785568138693195,t0-beta,{} +t0-beta,autogluon/fev_datasets,solar_with_weather_1H,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,"['day_length', 'humidity', 'pressure', 'rain_1h', 'snow_1h', 'temp', 'wind_speed']","['clouds_all', 'global_horizontal_irradiance']",[],solar_with_weather_1H,0.7260272011413733,0.0,7.983821294124937,20,12eed0644b253066,False,0.11.0.dev0,0.7260272011413733,0.9194721351203677,1.190181348784384,0.9714267850546104,t0-beta,{} +t0-beta,autogluon/fev_datasets,solar_1D,28,10,-280,28,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],solar_1D,0.6076409556696953,0.0,2.443576747988118,1370,ae293fab54bb1beb,False,0.11.0.dev0,0.6076409556696953,0.7780889003826745,0.24445196476190842,0.19009105607143212,t0-beta,{} +t0-beta,autogluon/fev_datasets,solar_1W,13,1,-13,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],solar_1W,0.9188772622961482,0.0,0.19957600001362152,137,63a28ed11ce85dc9,False,0.11.0.dev0,0.9188772622961482,1.179386740683897,0.20053219957321647,0.1552248869024074,t0-beta,{} +t0-beta,autogluon/fev_datasets,kdd_cup_2022_10T,288,10,-2880,288,1,,144,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,Patv,,[],"['Etmp', 'Itmp', 'Ndir', 'Pab1', 'Pab2', 'Pab3', 'Prtv', 'Wdir', 'Wspd']",[],kdd_cup_2022_10T,0.4323948327738473,0.0,610.8264000419585,1340,16cef2ee148416f8,False,0.11.0.dev0,0.4323948327738473,0.5300067403075827,1.6599990980747035,1.4424459446075564,t0-beta,{} +t0-beta,autogluon/fev_datasets,kdd_cup_2022_30T,96,10,-960,96,1,,48,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,Patv,,[],"['Etmp', 'Itmp', 'Ndir', 'Pab1', 'Pab2', 'Pab3', 'Prtv', 'Wdir', 'Wspd']",[],kdd_cup_2022_30T,0.42320230572223183,0.0,566.6052229180932,1340,d65c83db5d8286cf,False,0.11.0.dev0,0.42320230572223183,0.5257137490560031,1.6124037617690614,1.3888671484988713,t0-beta,{} +t0-beta,autogluon/fev_datasets,kdd_cup_2022_1D,14,10,-140,14,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,Patv,,[],"['Etmp', 'Itmp', 'Ndir', 'Pab1', 'Pab2', 'Pab3', 'Prtv', 'Wdir', 'Wspd']",[],kdd_cup_2022_1D,0.6932956595935825,0.0,20.848009501991328,1340,4126635775d467d9,False,0.11.0.dev0,0.6932956595935825,0.888593048919029,0.7068364148693833,0.5511414534008503,t0-beta,{} +t0-beta,autogluon/fev_datasets,uci_air_quality_1H,168,20,-3360,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['C6H6(GT)', 'CO(GT)', 'NO2(GT)', 'NOx(GT)']",,"['AH', 'RH', 'T']",[],[],uci_air_quality_1H,0.8328963925592493,0.0,5.123714958957862,80,770ea30aa1f2a0fd,False,0.11.0.dev0,0.8328963925592493,1.0701212603379282,0.3442283395975089,0.26701061145816596,t0-beta,{} +t0-beta,autogluon/fev_datasets,uci_air_quality_1D,28,11,-308,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['C6H6(GT)', 'CO(GT)', 'NO2(GT)', 'NOx(GT)']",,"['AH', 'RH', 'T']",[],[],uci_air_quality_1D,1.090320805047934,0.0,0.5714739600371104,44,b6e9003deabe1eba,False,0.11.0.dev0,1.090320805047934,1.3810119125138807,0.27554484173617755,0.21752857413188845,t0-beta,{} +t0-beta,autogluon/fev_datasets,LOOP_SEATTLE_5T,288,10,-2880,288,1,,288,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],LOOP_SEATTLE_5T,0.5620928867796356,0.0,136.83326354101882,3230,7855faba94bb7184,False,0.11.0.dev0,0.5620928867796356,0.7184941637226713,0.0781440752836429,0.06101998085760145,t0-beta,{} +t0-beta,autogluon/fev_datasets,LOOP_SEATTLE_1D,28,10,-280,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],LOOP_SEATTLE_1D,0.7445029473517538,0.0,5.725027333945036,3230,d46589b2c9e45a4d,False,0.11.0.dev0,0.7445029473517538,0.9125536709856507,0.03456743748868979,0.02813033632875383,t0-beta,{} +t0-beta,autogluon/fev_datasets,LOOP_SEATTLE_1H,168,10,-1680,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],LOOP_SEATTLE_1H,0.652386282325275,0.0,132.78672741603805,3230,002ea9451d732fea,False,0.11.0.dev0,0.652386282325275,0.8160314679711476,0.07016956272423804,0.0559392937799811,t0-beta,{} +t0-beta,autogluon/fev_datasets,M_DENSE_1D,28,10,-280,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],M_DENSE_1D,0.6892200671256836,0.0,0.9678524979390204,300,f18fd4e8b2a34059,False,0.11.0.dev0,0.6892200671256836,0.8331794527059353,0.0926643395458682,0.07713732741841535,t0-beta,{} +t0-beta,autogluon/fev_datasets,M_DENSE_1H,168,10,-1680,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],M_DENSE_1H,0.5658932021597765,0.0,12.999546000995906,300,f40bfe9afb7b3e51,False,0.11.0.dev0,0.5658932021597765,0.682113099265286,0.14090315186831734,0.11647575632445213,t0-beta,{} +t0-beta,autogluon/fev_datasets,SZ_TAXI_15T,96,10,-960,96,1,,96,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],SZ_TAXI_15T,0.39341312490667973,0.0,20.45471120698494,1560,4bd3ed28835dc961,False,0.11.0.dev0,0.39341312490667973,0.4999589202354233,0.23920716647140866,0.1879227901445885,t0-beta,{} +t0-beta,autogluon/fev_datasets,SZ_TAXI_1H,168,2,-336,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],SZ_TAXI_1H,0.41090745804680484,0.0,1.1370906659867615,312,a22b45b2f394c6d4,False,0.11.0.dev0,0.41090745804680484,0.5006280463154759,0.1703042489823699,0.14032761608145505,t0-beta,{} +t0-beta,autogluon/fev_datasets,redset_5T,288,10,-2880,288,1,,288,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],['subset'],redset_5T,0.6365706109258842,0.0,55.53091891700751,1180,706c03b5beda8537,False,0.11.0.dev0,0.6365706109258842,0.7642857926627447,0.32340516207166065,0.26244542033108265,t0-beta,{} +t0-beta,autogluon/fev_datasets,redset_15T,96,10,-960,96,1,,96,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],['subset'],redset_15T,0.7037177943794382,0.0,58.04097383306362,1260,08bf434cbfb37dc0,False,0.11.0.dev0,0.7037177943794382,0.8412829354099541,0.27064006849904776,0.22406652667728197,t0-beta,{} +t0-beta,autogluon/fev_datasets,redset_1H,24,10,-240,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],['subset'],redset_1H,1.279064922266553,0.0,14.919905665970873,1380,e8e85a6b0b054d1f,False,0.11.0.dev0,1.279064922266553,1.3851902799778197,0.21491575874156127,0.18014555511196753,t0-beta,{} +t0-beta,autogluon/fev_datasets,hospital_admissions_1D,28,20,-560,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],hospital_admissions_1D,0.5540306078599234,0.0,1.6737817470275331,160,3f0a46c7e6210f5f,False,0.11.0.dev0,0.5540306078599234,0.7169101078510871,0.5318159867272909,0.41094962208389874,t0-beta,{} +t0-beta,autogluon/fev_datasets,hospital_admissions_1W,13,16,-208,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],hospital_admissions_1W,0.5809115410022369,0.0,0.6961007490463089,128,008ee200cfa88de1,False,0.11.0.dev0,0.5809115410022369,0.7558617689548583,0.2128278315973171,0.16360944438312006,t0-beta,{} +t0-beta,autogluon/fev_datasets,ETT_1W,13,5,-65,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['HUFL', 'HULL', 'LUFL', 'LULL', 'MUFL', 'MULL', 'OT']",,[],[],[],ETT_1W,2.411692734001314,0.0,0.2673310420359485,70,7eda996e7e174b76,False,0.11.0.dev0,2.411692734001314,2.8226951713191606,0.5390664993299865,0.4329246272892375,t0-beta,{} +t0-beta,autogluon/fev_datasets,ETT_1D,28,20,-560,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['HUFL', 'HULL', 'LUFL', 'LULL', 'MUFL', 'MULL', 'OT']",,[],[],[],ETT_1D,1.1115799406750897,0.0,1.1827425410156138,280,09165729d889736c,False,0.11.0.dev0,1.1115799406750897,1.338239169866152,0.31255531957333993,0.25012740245235976,t0-beta,{} +t0-beta,autogluon/fev_datasets,ETT_1H,168,20,-3360,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['HUFL', 'HULL', 'LUFL', 'LULL', 'MUFL', 'MULL', 'OT']",,[],[],[],ETT_1H,0.8467980861767117,0.0,10.583075334056048,280,8990bcbad8689c1f,False,0.11.0.dev0,0.8467980861767117,1.0849454660702524,0.23525671956852845,0.18384964535001608,t0-beta,{} +t0-beta,autogluon/fev_datasets,ETT_15T,96,20,-1920,96,1,,96,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['HUFL', 'HULL', 'LUFL', 'LULL', 'MUFL', 'MULL', 'OT']",,[],[],[],ETT_15T,0.5645470021332736,0.0,12.061242209980264,280,b25a46033df1ea43,False,0.11.0.dev0,0.5645470021332736,0.7227745105040957,0.11750531112724434,0.09111344792057628,t0-beta,{} +t0-beta,autogluon/fev_datasets,hospital,12,4,-48,12,1,,12,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],hospital,0.7043435711390388,0.0,3.11270125003648,3068,fa8f208d01d9a008,False,0.11.0.dev0,0.7043435711390388,0.8994025634950755,0.09521695682664263,0.07491146962958777,t0-beta,{} +t0-beta,autogluon/fev_datasets,jena_weather_10T,144,20,-2880,144,1,,144,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_3', 'target_4', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],jena_weather_10T,0.34524157338821865,0.0,18.16911433302448,420,289a13db9929af4f,False,0.11.0.dev0,0.34524157338821865,0.42430571273758133,0.24863247483719042,0.21497384015190804,t0-beta,{} +t0-beta,autogluon/fev_datasets,jena_weather_1D,28,11,-308,28,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_3', 'target_4', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],jena_weather_1D,1.069989669160831,0.0,0.7978654569596983,231,a8ad906bff503daf,False,0.11.0.dev0,1.069989669160831,1.3579071647873766,0.27356503888266226,0.2226615506247258,t0-beta,{} +t0-beta,autogluon/fev_datasets,jena_weather_1H,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_3', 'target_4', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],jena_weather_1H,0.35194370180835366,0.0,17.718673125986243,420,20626aa7356b69f9,False,0.11.0.dev0,0.35194370180835366,0.43703800621628314,0.2484481216623678,0.22398025540995342,t0-beta,{} +t0-beta,autogluon/fev_datasets,bizitobs_l2c_1H,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_2', 'target_3', 'target_4', 'target_5', 'target_6']",,[],[],[],bizitobs_l2c_1H,0.32331328819547006,0.0,2.080046414048411,140,b0e17ddbdd539d9e,False,0.11.0.dev0,0.32331328819547006,0.416315801759779,0.32254594190144903,0.2571295707105647,t0-beta,{} +t0-beta,autogluon/fev_datasets,bizitobs_l2c_5T,288,20,-5760,288,1,,288,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_2', 'target_3', 'target_4', 'target_5', 'target_6']",,[],[],[],bizitobs_l2c_5T,0.4251864343958934,0.0,5.649676418979652,140,cdf3cd0903314fa4,False,0.11.0.dev0,0.4251864343958934,0.5567843126009917,0.7863008800109673,0.5950307225352999,t0-beta,{} +t0-beta,autogluon/fev_datasets,australian_tourism,8,2,-16,8,1,,4,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],australian_tourism,0.8546859420108546,0.0,0.2563003750110511,178,7013fcb0b68de2bf,False,0.11.0.dev0,0.8546859420108546,1.0749153887098175,0.12230942717113302,0.09713989498821896,t0-beta,{} +t0-beta,autogluon/fev_datasets,world_tourism,5,2,-10,5,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],world_tourism,3.1939294123422055,0.0,0.4175635419669561,356,893ed5f68fefe071,False,0.11.0.dev0,3.1939294123422055,3.96415787741378,0.12134741842408163,0.09062306317317581,t0-beta,{} +t0-beta,autogluon/fev_datasets,world_co2_emissions,5,9,-45,5,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],world_co2_emissions,2.8999921544573954,0.0,1.7432209590042476,1719,0604d178ae264900,False,0.11.0.dev0,2.8999921544573954,3.543158946059153,0.09232044757282419,0.07435207006858234,t0-beta,{} +t0-beta,autogluon/fev_datasets,world_life_expectancy,5,10,-50,5,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],world_life_expectancy,1.1530741529433706,0.0,2.516593833017396,2370,34b64743f43fc2a0,False,0.11.0.dev0,1.1530741529433706,1.3953731063314112,0.011145702226141816,0.009294700467499814,t0-beta,{} +t0-beta,autogluon/fev_datasets,ercot_1H,168,20,-3360,168,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],ercot_1H,1.0157527257591967,0.0,6.10983875094098,160,6ed98f4d1d5e8850,False,0.11.0.dev0,1.0157527257591967,1.3297351759420293,0.06589813014946531,0.05011982688082404,t0-beta,{} +t0-beta,autogluon/fev_datasets,ercot_1D,28,20,-560,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],ercot_1D,0.8884238709637252,0.0,4.839801585942041,160,765631b415c5e31b,False,0.11.0.dev0,0.8884238709637252,1.1204538721957558,0.07559403326190083,0.058907213385346234,t0-beta,{} +t0-beta,autogluon/fev_datasets,ercot_1W,13,20,-260,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],ercot_1W,0.9861662927848723,0.0,1.095964164007455,160,5ac87e3ab310fdc5,False,0.11.0.dev0,0.9861662927848723,1.2562392606000514,0.05881872018739322,0.04582162383515782,t0-beta,{} +t0-beta,autogluon/fev_datasets,ercot_1M,12,15,-180,12,1,,12,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],ercot_1M,0.7964256271954522,0.0,0.5302412099554203,120,7f70bf68861c13d6,False,0.11.0.dev0,0.7964256271954522,1.0054918019483354,0.05129952697880761,0.04032989629935865,t0-beta,{} +t0-beta,autogluon/fev_datasets,uk_covid_utla_1D,28,10,-280,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,new_cases,,[],[],[],uk_covid_utla_1D/new,3.7434353252568684,0.0,7.006650457013166,2140,da1326db1e917a89,False,0.11.0.dev0,3.7434353252568684,4.482544518190697,0.4636393816832129,0.38785672403813026,t0-beta,{} +t0-beta,autogluon/fev_datasets,uk_covid_utla_1W,13,5,-65,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,cumulative_cases,,[],[],[],uk_covid_utla_1W/cumulative,16.2217899120627,0.0,1.2305457499751355,1070,4c07446a089a723d,False,0.11.0.dev0,16.2217899120627,18.504628595025828,0.18222279654341272,0.15851781612260757,t0-beta,{} +t0-beta,autogluon/fev_datasets,uk_covid_nation_1D,28,20,-560,28,1,,7,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['new_admissions', 'new_cases', 'new_deaths']",,[],"['hospital_cases', 'icu_ventilator_occupancy', 'new_vaccinated_1', 'new_vaccinated_2', 'new_vaccinated_3']",[],uk_covid_nation_1D/new,2.0666330322581414,0.0,1.7159867060254328,240,49ed8b456f954821,False,0.11.0.dev0,2.0666330322581414,2.517929186186314,0.3508211748138023,0.2806501921171388,t0-beta,{} +t0-beta,autogluon/fev_datasets,uk_covid_nation_1D,28,20,-560,28,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['cumulative_admissions', 'cumulative_cases', 'cumulative_deaths']",,[],"['cumulative_vaccinated_1', 'cumulative_vaccinated_2', 'cumulative_vaccinated_3', 'hospital_cases', 'icu_ventilator_occupancy']",[],uk_covid_nation_1D/cumulative,6.60453850945993,0.0,1.721824667067267,240,49ed8b456f954821,False,0.11.0.dev0,6.60453850945993,8.056115546345659,0.01714529802741517,0.013954276991937723,t0-beta,{} +t0-beta,autogluon/fev_datasets,uk_covid_nation_1W,8,4,-32,8,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['new_admissions', 'new_cases', 'new_deaths']",,[],"['hospital_cases', 'icu_ventilator_occupancy', 'new_vaccinated_1', 'new_vaccinated_2', 'new_vaccinated_3']",[],uk_covid_nation_1W/new,4.394749048771704,0.0,0.19071212597191334,48,92ea0a1b98c9e5ce,False,0.11.0.dev0,4.394749048771704,5.444677512357283,0.750217620368823,0.5762151705977185,t0-beta,{} +t0-beta,autogluon/fev_datasets,uk_covid_nation_1W,8,4,-32,8,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['cumulative_admissions', 'cumulative_cases', 'cumulative_deaths']",,[],"['cumulative_vaccinated_1', 'cumulative_vaccinated_2', 'cumulative_vaccinated_3', 'hospital_cases', 'icu_ventilator_occupancy']",[],uk_covid_nation_1W/cumulative,2.716459299740184,0.0,0.18537958298111334,48,92ea0a1b98c9e5ce,False,0.11.0.dev0,2.716459299740184,3.2165098598676067,0.023884487124138546,0.021782452879378137,t0-beta,{} +t0-beta,autogluon/fev_datasets,ecdc_ili,13,10,-130,13,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],ecdc_ili,2.398087832089685,0.0,0.4996379189833533,240,6c8b131f985e4218,False,0.11.0.dev0,2.398087832089685,2.8851206058488414,0.4237971529534913,0.36089647301983035,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1062,288,20,-5760,288,1,,288,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_3', 'target_4', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_1062,0.5637840217144812,0.0,17.243291248974856,420,eec81c201b8778e1,False,0.11.0.dev0,0.5637840217144812,0.6834670303922333,0.569372556969821,0.4563116778039239,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1209,288,20,-5760,288,1,,288,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_52', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_1209,0.6450837939080699,0.0,47.16822266599047,1060,161694bf9dc519eb,False,0.11.0.dev0,0.6450837939080699,0.7410570666525546,0.2906477521305527,0.24906987487233767,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1225,60,20,-1200,60,1,,1440,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_1225,0.1909080100917179,0.0,42.20891416500672,980,16ca972263ae30e6,False,0.11.0.dev0,0.1909080100917179,0.23757984871604157,0.12481333766926488,0.10061352344815902,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1230,288,20,-5760,288,1,,288,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_3', 'target_4', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_1230,1.1494760389933707,0.0,19.850794665981084,460,71842e1d5b103a62,False,0.11.0.dev0,1.1494760389933707,1.32214898935109,0.295891869952171,0.251257627328589,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1282,60,20,-1200,60,1,,1440,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_4', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_1282,0.4132995072906387,0.0,30.489875873958226,700,9387ab4d0c15b634,False,0.11.0.dev0,0.4132995072906387,0.5047672178192821,0.35055431828211503,0.2844290617672604,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1487,288,20,-5760,288,1,,288,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_52', 'target_53', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_1487,0.4090606607736541,0.0,48.58540091692703,1080,acf7534da4aae38f,False,0.11.0.dev0,0.4090606607736541,0.4943806294128576,0.1946173552394076,0.16086143812095027,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1631,96,20,-1920,96,1,,48,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_1631,0.5942056391526445,0.0,35.51570125101716,800,74d0d6952f666083,False,0.11.0.dev0,0.5942056391526445,0.7310489847156604,0.3173080142463384,0.25763139403359187,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1676,96,20,-1920,96,1,,48,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_52', 'target_53', 'target_54', 'target_55', 'target_56', 'target_57', 'target_58', 'target_59', 'target_6', 'target_60', 'target_61', 'target_62', 'target_63', 'target_64', 'target_65', 'target_66', 'target_67', 'target_68', 'target_69', 'target_7', 'target_70', 'target_71', 'target_72', 'target_73', 'target_74', 'target_75', 'target_76', 'target_77', 'target_78', 'target_79', 'target_8', 'target_80', 'target_81', 'target_82', 'target_83', 'target_84', 'target_85', 'target_86', 'target_87', 'target_88', 'target_89', 'target_9', 'target_90', 'target_91', 'target_92', 'target_93', 'target_94', 'target_95', 'target_96', 'target_97', 'target_98', 'target_99']",,[],[],[],boomlet_1676,0.5700507469414987,0.0,90.89915574991028,2000,abbc88acc77da441,False,0.11.0.dev0,0.5700507469414987,0.6951919884918376,0.3150263203166631,0.24877521208749637,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1855,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_1855,0.45094264523938454,0.0,27.538887331989827,1040,d0c354b056fd7763,False,0.11.0.dev0,0.45094264523938454,0.5285573659333027,0.1572322204159409,0.1349303169372233,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_1975,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_52', 'target_53', 'target_54', 'target_55', 'target_56', 'target_57', 'target_58', 'target_59', 'target_6', 'target_60', 'target_61', 'target_62', 'target_63', 'target_64', 'target_65', 'target_66', 'target_67', 'target_68', 'target_69', 'target_7', 'target_70', 'target_71', 'target_72', 'target_73', 'target_74', 'target_8', 'target_9']",,[],[],[],boomlet_1975,0.12640866230220413,0.0,40.04121974800364,1500,8d750d7893416ec7,False,0.11.0.dev0,0.12640866230220413,0.14964332647662565,0.08349696928626354,0.0705242523652416,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_2187,24,20,-480,24,1,,24,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_52', 'target_53', 'target_54', 'target_55', 'target_56', 'target_57', 'target_58', 'target_59', 'target_6', 'target_60', 'target_61', 'target_62', 'target_63', 'target_64', 'target_65', 'target_66', 'target_67', 'target_68', 'target_69', 'target_7', 'target_70', 'target_71', 'target_72', 'target_73', 'target_74', 'target_75', 'target_76', 'target_77', 'target_78', 'target_79', 'target_8', 'target_80', 'target_81', 'target_82', 'target_83', 'target_84', 'target_85', 'target_86', 'target_87', 'target_88', 'target_89', 'target_9', 'target_90', 'target_91', 'target_92', 'target_93', 'target_94', 'target_95', 'target_96', 'target_97', 'target_98', 'target_99']",,[],[],[],boomlet_2187,0.7158103050625576,0.0,52.924221667984966,2000,a13b0fc5b533c30e,False,0.11.0.dev0,0.7158103050625576,0.8075381646236044,0.2630215145566903,0.23133966883642115,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_285,60,20,-1200,60,1,,1440,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_52', 'target_53', 'target_54', 'target_55', 'target_56', 'target_57', 'target_58', 'target_59', 'target_6', 'target_60', 'target_61', 'target_62', 'target_63', 'target_64', 'target_65', 'target_66', 'target_67', 'target_68', 'target_69', 'target_7', 'target_70', 'target_71', 'target_72', 'target_73', 'target_74', 'target_8', 'target_9']",,[],[],[],boomlet_285,0.33457168386331704,0.0,70.0498185850447,1500,0ced6efaa846e1bd,False,0.11.0.dev0,0.33457168386331704,0.39878344324799475,0.18431554118130605,0.14909968318206407,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_619,60,20,-1200,60,1,,1440,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_619,0.31046699433374714,0.0,49.15043437303393,1040,83721e76c6faf4a6,False,0.11.0.dev0,0.31046699433374714,0.4072183309121013,0.28353552275215926,0.21722685650876666,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_772,60,20,-1200,60,1,,1440,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_28', 'target_29', 'target_3', 'target_30', 'target_31', 'target_32', 'target_33', 'target_34', 'target_35', 'target_36', 'target_37', 'target_38', 'target_39', 'target_4', 'target_40', 'target_41', 'target_42', 'target_43', 'target_44', 'target_45', 'target_46', 'target_47', 'target_48', 'target_49', 'target_5', 'target_50', 'target_51', 'target_52', 'target_53', 'target_54', 'target_55', 'target_56', 'target_57', 'target_58', 'target_59', 'target_6', 'target_60', 'target_61', 'target_62', 'target_63', 'target_64', 'target_65', 'target_66', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_772,0.2895226996975069,0.0,65.01472745803767,1340,f6a344659a6a948e,False,0.11.0.dev0,0.2895226996975069,0.33707581939147196,0.15903982362582073,0.13463635346665437,t0-beta,{} +t0-beta,autogluon/fev_datasets,boomlet_963,60,20,-1200,60,1,,1440,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['target_0', 'target_1', 'target_10', 'target_11', 'target_12', 'target_13', 'target_14', 'target_15', 'target_16', 'target_17', 'target_18', 'target_19', 'target_2', 'target_20', 'target_21', 'target_22', 'target_23', 'target_24', 'target_25', 'target_26', 'target_27', 'target_3', 'target_4', 'target_5', 'target_6', 'target_7', 'target_8', 'target_9']",,[],[],[],boomlet_963,0.7255537508515502,0.0,25.635096169018652,560,5bbc9fb121124724,False,0.11.0.dev0,0.7255537508515502,0.8422417530605053,0.39470335529863865,0.325614725604912,t0-beta,{} +t0-beta,autogluon/fev_datasets,fred_md_2025,12,20,-240,12,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['CPIAUCSL', 'FEDFUNDS', 'PAYEMS']",,[],"['M2REAL', 'NONBORRES', 'PPICMM', 'TOTRESNS']",[],fred_md_2025/cee,3.3651385659145254,0.0,0.9499526259314734,60,74fe4767146a4d96,False,0.11.0.dev0,3.3651385659145254,4.026454708722861,0.07528815293086091,0.06223419502134747,t0-beta,{} +t0-beta,autogluon/fev_datasets,fred_qd_2025,8,20,-160,8,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['CPIAUCSL', 'FEDFUNDS', 'PAYEMS']",,[],"['M2REAL', 'NONBORRES', 'PPICMM', 'TOTRESNS']",[],fred_qd_2025/cee,1.9688849870599827,0.0,0.6540877490187995,60,d4ca154cc310644e,False,0.11.0.dev0,1.9688849870599827,2.447018326243229,0.15183568368376327,0.12251327082032662,t0-beta,{} +t0-beta,autogluon/fev_datasets,fred_md_2025,12,20,-240,12,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['AAA', 'AWHMAN', 'BAA', 'BUSINVx', 'CPIAUCSL', 'CPIMEDSL', 'CUMFNS', 'DMANEMP', 'EXUSUKx', 'FEDFUNDS', 'GS1', 'GS10', 'GS5', 'HOUST', 'INDPRO', 'INVEST', 'IPBUSEQ', 'IPCONGD', 'IPDCONGD', 'IPDMAT', 'IPFINAL', 'IPMANSICS', 'IPMAT', 'IPNCONGD', 'IPNMAT', 'M2REAL', 'MANEMP', 'NDMANEMP', 'NONBORRES', 'PAYEMS', 'PERMIT', 'PPICMM', 'S&P 500', 'SRVPRD', 'TB3MS', 'TB6MS', 'TOTRESNS', 'UEMP15T26', 'UEMP27OV', 'UEMP5TO14', 'UEMPLT5', 'UEMPMEAN', 'UMCSENTx', 'UNRATE', 'USCONS', 'USFIRE', 'USGOOD', 'USGOVT', 'USTPU', 'USTRADE', 'VIXCLSx']",,[],[],[],fred_md_2025/macro,5.2826322287455465,0.0,3.6688975430151913,1020,74fe4767146a4d96,False,0.11.0.dev0,5.2826322287455465,6.18324680359369,0.07805368265312632,0.06405630435939061,t0-beta,{} +t0-beta,autogluon/fev_datasets,fred_qd_2025,8,20,-160,8,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['AAA', 'AWHMAN', 'BAA', 'BUSINVx', 'CPIAUCSL', 'CPIMEDSL', 'CUMFNS', 'DMANEMP', 'EXUSUKx', 'FEDFUNDS', 'GS1', 'GS10', 'GS5', 'HOUST', 'INDPRO', 'INVEST', 'IPBUSEQ', 'IPCONGD', 'IPDCONGD', 'IPDMAT', 'IPFINAL', 'IPMANSICS', 'IPMAT', 'IPNCONGD', 'IPNMAT', 'M2REAL', 'MANEMP', 'NDMANEMP', 'NONBORRES', 'PAYEMS', 'PERMIT', 'PPICMM', 'S&P 500', 'SRVPRD', 'TB3MS', 'TB6MS', 'TOTRESNS', 'UEMP15T26', 'UEMP27OV', 'UEMP5TO14', 'UEMPLT5', 'UEMPMEAN', 'UMCSENTx', 'UNRATE', 'USCONS', 'USFIRE', 'USGOOD', 'USGOVT', 'USTPU', 'USTRADE', 'VIXCLSx']",,[],[],[],fred_qd_2025/macro,3.3963243499412363,0.0,1.907093083980726,1020,d4ca154cc310644e,False,0.11.0.dev0,3.3963243499412363,4.126863634821758,0.10792573691773041,0.08590239616737218,t0-beta,{} +t0-beta,autogluon/fev_datasets,us_consumption_1M,12,10,-120,12,1,,12,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],us_consumption_1M,1.5587330340791776,0.0,1.3012679569947068,310,dee11a908eddc948,False,0.11.0.dev0,1.5587330340791776,1.8571741691969401,0.023273664347491372,0.020254795539488654,t0-beta,{} +t0-beta,autogluon/fev_datasets,us_consumption_1Q,8,10,-80,8,1,,4,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],us_consumption_1Q,1.8687457582071452,0.0,0.655338040960487,310,05aa7957a46336f0,False,0.11.0.dev0,1.8687457582071452,2.338509924007952,0.036142958662097976,0.029643102019801672,t0-beta,{} +t0-beta,autogluon/fev_datasets,us_consumption_1Y,5,10,-50,5,1,,1,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,target,,[],[],[],us_consumption_1Y,4.186575269980564,0.0,0.33802174901938997,310,34316f35c64781aa,False,0.11.0.dev0,4.186575269980564,5.144902886890928,0.0683048133091853,0.055287442945497055,t0-beta,{} +t0-beta,autogluon/fev_datasets,gvar,8,10,-80,8,1,,4,SQL,"['MASE', {'name': 'WAPE', 'epsilon': 1.0}, {'name': 'WQL', 'epsilon': 1.0}]","[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]",id,timestamp,"['Dp', 'ep', 'eq', 'lr', 'r', 'y']",,[],"['pmat', 'pmetal', 'poil']",[],gvar,0.5808244644939119,0.0,4.456063874036772,1980,d404e6be8d6f82fd,False,0.11.0.dev0,0.5808244644939119,0.7174401581625833,0.018563041183193494,0.015088065611144457,t0-beta,{} diff --git a/models/t0-beta/README.md b/models/t0-beta/README.md new file mode 100644 index 0000000..127b411 --- /dev/null +++ b/models/t0-beta/README.md @@ -0,0 +1,31 @@ +# t0-beta + +`t0-beta` is a 256M-parameter open-weights time series foundation model from +[The Forecasting Company](https://theforecastingcompany.com/). It is a +decoder-style patch transformer that emits 21 quantile levels natively and +conditions on past and known-future covariates. + +- Weights: https://huggingface.co/theforecastingcompany/t0-beta +- Code: https://github.com/theforecastingcompany/tfc-t0 +- Runtime: [`tfc-t0`](https://pypi.org/project/tfc-t0/) `>=0.5.0` + +```bash +python models/evaluate.py -m t0-beta +``` + +## How the wrapper maps a task + +A task's columns become variates of one t0 sample, which attend to one another +through the model's group attention: + +| Task column | t0 variate | +| --- | --- | +| every target column | `TARGET`, all forecast in one pass | +| `known_dynamic_columns` | `FUTURE`, spanning context and horizon | +| `past_dynamic_columns` | `HISTORICAL`, stopping at the forecast start | + +Non-numeric covariates are label-encoded per request, as the model has no +categorical embedding. Context is truncated to the most recent 8192 steps. + +Ablations: `-k '{"use_covariates": false}'` drops the covariate rows, and +`-k '{"as_univariate": true}'` forecasts each target column on its own. diff --git a/models/t0-beta/model.py b/models/t0-beta/model.py new file mode 100644 index 0000000..0262853 --- /dev/null +++ b/models/t0-beta/model.py @@ -0,0 +1,231 @@ +"""t0-beta model wrapper for fev evaluation. + +t0 is a decoder-style patch transformer from The Forecasting Company that emits +quantile forecasts and natively conditions on covariates. `t0-beta` is its +256M-parameter second public checkpoint. + +- Weights: https://huggingface.co/theforecastingcompany/t0-beta +- Runtime: https://pypi.org/project/tfc-t0/ + +A task's columns map onto t0 variates of one sample, which attend to one another +through the model's group attention: every target column is a TARGET row and all +are forecast in one pass, ``known_dynamic_columns`` span context and horizon as +FUTURE rows, and ``past_dynamic_columns`` stop at the forecast start as +HISTORICAL rows. ``use_covariates=False`` drops the covariate rows and +``as_univariate=True`` forecasts each target column on its own, for ablations. + +Usage: + python models/evaluate.py -m t0-beta +""" + +import datasets +import numpy as np +import torch +from t0 import MaskType, T0Forecaster, TimeSeries, VariateType, batch_series + +import fev + +# Longest history fed to the model; longer series keep their most recent steps. +DEFAULT_MAX_CONTEXT = 8192 +DEFAULT_BATCH_SIZE = 32 + + +def cast_as_numeric(*sequences) -> list[np.ndarray]: + """Coerce covariate values to float32, label-encoding non-numeric ones.""" + try: + return [np.asarray(s, dtype=np.float32) for s in sequences] + except (ValueError, TypeError): + seqs = [[str(v) for v in s] for s in sequences] + codes = {value: float(idx) for idx, value in enumerate(sorted({v for s in seqs for v in s}))} + return [np.asarray([codes[v] for v in s], dtype=np.float32) for s in seqs] + + +class T0BetaModel(fev.ForecastingModel): + """t0-beta forecasting foundation model from The Forecasting Company.""" + + model_name = "t0-beta" + # t0-beta was trained with no data leakage. + trained_on_datasets = [] + + def __init__( + self, + model_path: str = "theforecastingcompany/t0-beta", + device: str = "auto", + max_context_length: int = DEFAULT_MAX_CONTEXT, + batch_size: int = DEFAULT_BATCH_SIZE, + use_covariates: bool = True, + as_univariate: bool = False, + ): + super().__init__() + self.model_path = model_path + self.device = device + self.max_context_length = max_context_length + self.batch_size = batch_size + self.use_covariates = use_covariates + self.as_univariate = as_univariate + self._model = None + + def _load_model(self) -> T0Forecaster: + """Load the forecaster once and reuse it across tasks.""" + if self._model is None: + model_path = fev.utils.maybe_cache_from_s3(self.model_path) + self._model = T0Forecaster.from_pretrained(model_path).to(self._resolve_device()).eval() + return self._model + + def _resolve_device(self) -> str: + """Honour an explicit device, else take the best of cuda, mps, cpu.""" + if self.device != "auto": + return self.device + if torch.cuda.is_available(): + return "cuda" + return "mps" if torch.backends.mps.is_available() else "cpu" + + def convert_series( + self, + past_cols, + future_cols, + index: int, + targets: list[str], + known: list[str], + past_only: list[str], + horizon: int, + ) -> TimeSeries: + """One series as a t0 ``TimeSeries``: target, known-future and historical rows. + + Every target column becomes a TARGET row of the same sample, so the columns + attend to one another and are forecast in a single pass. + """ + context = np.stack([cast_as_numeric(past_cols[c][index])[0][-self.max_context_length :] for c in targets]) + context_len = context.shape[1] + + future_covariates = None + if known: + # A known column spans context and horizon: its past half lives in the + # window's past_data, its future half in future_data. + rows = [] + for c in known: + seen, ahead = cast_as_numeric(past_cols[c][index], future_cols[c][index]) + rows.append(np.concatenate([seen[-context_len:], ahead[:horizon]])) + future_covariates = torch.from_numpy(np.stack(rows))[None] + + series = TimeSeries.from_array( + torch.from_numpy(context)[None], # (1, targets, context) + future_covariates=future_covariates, + horizon=horizon, + ) + if not past_only: + return series + + # Historical rows are known over the context and unknown across the + # horizon, so their forecast region is MISSING rather than WITHHELD. + width = series.seq_len + values = np.stack([cast_as_numeric(past_cols[c][index])[0][-context_len:] for c in past_only]) + hist = torch.zeros((len(past_only), width), dtype=torch.float32) + hist[:, :context_len] = torch.from_numpy(np.nan_to_num(values, nan=0.0)) + mask = torch.full((len(past_only), width), MaskType.MISSING, dtype=torch.int8) + mask[:, :context_len] = MaskType.VALID + # A gap in a covariate is an absent observation, not a zero: left VALID it + # would feed NaN to the scaler and turn the whole forecast non-finite. + gaps = torch.from_numpy(np.isnan(values)) + mask[:, :context_len] = torch.where( + gaps, torch.tensor(MaskType.MISSING, dtype=torch.int8), mask[:, :context_len] + ) + return TimeSeries( + variates=torch.cat([series.variates, hist]), + mask=torch.cat([series.mask, mask]), + group_ids=torch.cat([series.group_ids, torch.zeros_like(mask, dtype=torch.long)]), + variate_type=torch.cat( + [series.variate_type, torch.full_like(mask, VariateType.HISTORICAL, dtype=torch.long)] + ), + ) + + def _fit_predict(self, task: fev.Task) -> list[datasets.DatasetDict]: + device = self._resolve_device() + model = self._load_model() + + known = list(task.known_dynamic_columns or []) if self.use_covariates else [] + past_only = list(task.past_dynamic_columns or []) if self.use_covariates else [] + # as_univariate splits a multivariate task into one instance per target column, + # so the model then sees a single target row and the columns cannot attend. + targets = ["target"] if self.as_univariate else list(task.target_columns) + + # fev scores the point forecast with the task's own metric; a squared-error + # metric wants the mean, everything else the median. The model emits + # quantiles only, so 0.5 stands in for the mean. + levels = list(task.quantile_levels) + point_levels = sorted(levels if 0.5 in levels else [*levels, 0.5]) + + predictions_per_window = [] + for window in task.iter_windows(): + past_data, future_data = fev.convert_input_data( + window, adapter="datasets", as_univariate=self.as_univariate + ) + n_series = len(past_data) + # Read each column once. Indexing an Arrow dataset row-by-row inside the + # batch loop re-materialises every column of that row each time. + past_cols = {c: past_data[c] for c in [*targets, *known, *past_only]} + future_cols = {c: future_data[c] for c in known} if future_data is not None else {} + + quantiles_all = [] + with self._record_inference_time(): + rows_per_series = len(targets) + len(known) + len(past_only) + series_per_batch = max(1, self.batch_size // rows_per_series) + for start in range(0, n_series, series_per_batch): + stop = min(start + series_per_batch, n_series) + if known or past_only: + chunk = [ + self.convert_series(past_cols, future_cols, i, targets, known, past_only, task.horizon) + for i in range(start, stop) + ] + model_input = TimeSeries.batch(chunk).to(device) + out = self._predict(model, model_input, task.horizon, point_levels, batched=True) + else: + # One (targets, context) entry per series: batch_series gives its rows a + # shared group id, which is what makes the columns attend to one another. + contexts = [ + np.stack( + [cast_as_numeric(past_cols[c][i])[0][-self.max_context_length :] for c in targets] + ) + for i in range(start, stop) + ] + context, mask, group_ids = batch_series(contexts) + out = self._predict( + model, + context.to(device), + task.horizon, + point_levels, + mask=mask.to(device), + group_ids=group_ids.to(device), + ) + quantiles_all.append(out) + + # Target rows come back in input order, so the series axis splits off cleanly. + quantiles_np = np.concatenate(quantiles_all, axis=0).reshape( + n_series, len(targets), task.horizon, len(point_levels) + ) + + per_target = {} + for position, column in enumerate(targets): + predictions_dict = {"predictions": quantiles_np[:, position, :, point_levels.index(0.5)]} + for level in task.quantile_levels: + predictions_dict[str(level)] = quantiles_np[:, position, :, point_levels.index(level)] + per_target[column] = datasets.Dataset.from_dict(predictions_dict) + + if self.as_univariate: + # Each target column was its own instance; fev knows how to fold them back. + predictions_per_window.append( + fev.combine_univariate_predictions_to_multivariate( + per_target["target"], target_columns=task.target_columns + ) + ) + else: + predictions_per_window.append(datasets.DatasetDict(per_target)) + + return predictions_per_window + + def _predict(self, model, model_input, horizon, levels, *, batched=False, **kwargs) -> np.ndarray: + with torch.inference_mode(): + if batched: + kwargs["context_length"] = model_input.seq_len - horizon + out = model.predict(model_input, horizon=horizon, quantile_levels=levels, **kwargs) + return out.quantiles.float().cpu().numpy() diff --git a/models/t0-beta/requirements.txt b/models/t0-beta/requirements.txt new file mode 100644 index 0000000..be92395 --- /dev/null +++ b/models/t0-beta/requirements.txt @@ -0,0 +1 @@ +tfc-t0==0.5.0