Hugging Face's logo

Datasets:
Diaugeia
/
TSEval-Static

Modalities:
Text
Tags:
time-series
forecasting
benchmark
License:
Dataset card Data Studio Files
xet
 Community
Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 321 new columns ({'42', '12', '224', '298', '321', '318', '43', '89', '107', '167', '314', '57', '188', '265', '15', '156', '61', '26', '147', '249', '178', '154', '115', '93', '187', '137', '85', '133', '157', '144', '35', '9', '166', '113', '54', '295', '84', '8', '19', '141', '112', '260', '301', '159', '50', '67', '177', '47', '244', '5', '56', '235', '78', '273', '251', '92', '310', '109', '80', '223', '36', '259', '271', '7', '316', '130', '13', '300', '105', '100', '281', '185', '202', '38', '31', '64', '172', '319', '55', '193', '58', '132', '277', '103', '219', '239', '211', '320', '24', '87', '189', '155', '204', '91', '214', '275', '163', '212', '6', '256', '95', '108', '262', '86', '79', '118', '39', '272', '127', '236', '30', '213', '313', '90', '299', '306', '139', '169', '152', '125', '258', '282', '59', '34', '182', '270', '153', '20', '2', '195', '97', '257', '146', '46', '81', '206', '194', '17', '96', '74', '229', '274', '77', '252', '11', '237', '246', '288', '216', '176', '110', '44', '183', '263', '225', '222', '66', '291', '73', '269', '205', '196', '33', '199', '317', '253', '234', '145', '102', '63', '279', '241', '201', '4', '29', '70', '99', '60', '168', '192', '117', '124', '303', '268', '200', '131', '243', '27', '247', '69', '3', '71', '304', '120', '186', '287', '173', '197', '289', '309', '215', '129', '217', '162', '114', '180', '175', '161', '218', '245', '293', '22', '82', '45', '290', '37', '18', '142', '48', '106', '128', '233', '149', '181', '28', '135', '261', '138', '111', '284', '210', '65', '174', '164', '311', '101', '226', '150', '76', '136', '104', '238', '228', '278', '165', '191', '312', '254', '209', '75', '143', '220', '25', '53', '292', '134', '179', '98', '140', '285', '308', '14', '171', '16', '315', '32', '276', '242', '121', '203', '170', '207', '51', '1', '62', '158', '190', '23', '52', '88', '227', '283', '41', '94', '148', '255', '280', '302', '49', '83', '267', '208', '151', '72', '126', '264', '221', '294', '307', '250', '10', '123', '122', '21', '266', '119', '231', '248', '116', '232', '68', '296', '297', '40', '184', '240', '286', '198', '160', '230', '305'}) and 7 missing columns ({'OT', 'MUFL', 'LULL', 'HULL', 'HUFL', 'MULL', 'LUFL'}).

This happened while the csv dataset builder was generating data using

hf://datasets/Diaugeia/TSEval-Static/electricity.csv (at revision 1e76820425cfc08d4eb8886cbe75affd82edca3c), ['hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/ETTh1.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/ETTh2.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/ETTm1.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/ETTm2.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/electricity.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/traffic.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/weather.csv']

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1837, in _prepare_split_single
                  writer.write_table(table)
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 765, in write_table
                  self._write_table(pa_table, writer_batch_size=writer_batch_size)
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 773, in _write_table
                  pa_table = table_cast(pa_table, self._schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              date: string
              1: double
              2: double
              3: double
              4: double
              5: double
              6: double
              7: double
              8: double
              9: double
              10: double
              11: double
              12: double
              13: double
              14: double
              15: double
              16: double
              17: double
              18: double
              19: double
              20: double
              21: double
              22: double
              23: double
              24: double
              25: double
              26: double
              27: double
              28: double
              29: double
              30: double
              31: double
              32: double
              33: double
              34: double
              35: double
              36: double
              37: double
              38: double
              39: double
              40: double
              41: double
              42: double
              43: double
              44: double
              45: double
              46: double
              47: double
              48: double
              49: double
              50: double
              51: double
              52: double
              53: double
              54: double
              55: double
              56: double
              57: double
              58: double
              59: double
              60: double
              61: double
              62: double
              63: double
              64: double
              65: double
              66: double
              67: double
              68: double
              69: double
              70: double
              71: double
              72: double
              73: double
              74: double
              75: double
              76: double
              77: double
              78: double
              79: double
              80: double
              81: double
              82: double
              83: double
              84: double
              85: double
              86: double
              87: double
              88: double
              89: double
              90: double
              91: double
              92: double
              93: double
              94: double
              95: double
              96: double
              97: double
              98: double
              99: double
              100: double
              101: double
              102: double
              103: double
              104: double
              105: double
              106: double
              107: double
              108: double
              109: double
              110: double
              111: double
              112: double
              113: double
              114: double
              115: double
              116: double
              117: double
              118: double
              119: double
              120: double
              121: double
              122: double
              123: double
              124: double
              125: double
              126: double
              127: double
              128: double
              129: double
              130: double
              131: double
              132: double
              133: double
              ...
              ble
              206: double
              207: double
              208: double
              209: double
              210: double
              211: double
              212: double
              213: double
              214: double
              215: double
              216: double
              217: double
              218: double
              219: double
              220: double
              221: double
              222: double
              223: double
              224: double
              225: double
              226: double
              227: double
              228: double
              229: double
              230: double
              231: double
              232: double
              233: double
              234: double
              235: double
              236: double
              237: double
              238: double
              239: double
              240: double
              241: double
              242: double
              243: double
              244: double
              245: double
              246: double
              247: double
              248: double
              249: double
              250: double
              251: double
              252: double
              253: double
              254: double
              255: double
              256: double
              257: double
              258: double
              259: double
              260: double
              261: double
              262: double
              263: double
              264: double
              265: double
              266: double
              267: double
              268: double
              269: double
              270: double
              271: double
              272: double
              273: double
              274: double
              275: double
              276: double
              277: double
              278: double
              279: double
              280: double
              281: double
              282: double
              283: double
              284: double
              285: double
              286: double
              287: double
              288: double
              289: double
              290: double
              291: double
              292: double
              293: double
              294: double
              295: double
              296: double
              297: double
              298: double
              299: double
              300: double
              301: double
              302: double
              303: double
              304: double
              305: double
              306: double
              307: double
              308: double
              309: double
              310: double
              311: double
              312: double
              313: double
              314: double
              315: double
              316: double
              317: double
              318: double
              319: double
              320: double
              321: double
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 34502
              to
              {'date': Value('string'), 'HUFL': Value('float64'), 'HULL': Value('float64'), 'LUFL': Value('float64'), 'LULL': Value('float64'), 'MUFL': Value('float64'), 'MULL': Value('float64'), 'OT': Value('float64')}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1361, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 940, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1683, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1839, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 321 new columns ({'42', '12', '224', '298', '321', '318', '43', '89', '107', '167', '314', '57', '188', '265', '15', '156', '61', '26', '147', '249', '178', '154', '115', '93', '187', '137', '85', '133', '157', '144', '35', '9', '166', '113', '54', '295', '84', '8', '19', '141', '112', '260', '301', '159', '50', '67', '177', '47', '244', '5', '56', '235', '78', '273', '251', '92', '310', '109', '80', '223', '36', '259', '271', '7', '316', '130', '13', '300', '105', '100', '281', '185', '202', '38', '31', '64', '172', '319', '55', '193', '58', '132', '277', '103', '219', '239', '211', '320', '24', '87', '189', '155', '204', '91', '214', '275', '163', '212', '6', '256', '95', '108', '262', '86', '79', '118', '39', '272', '127', '236', '30', '213', '313', '90', '299', '306', '139', '169', '152', '125', '258', '282', '59', '34', '182', '270', '153', '20', '2', '195', '97', '257', '146', '46', '81', '206', '194', '17', '96', '74', '229', '274', '77', '252', '11', '237', '246', '288', '216', '176', '110', '44', '183', '263', '225', '222', '66', '291', '73', '269', '205', '196', '33', '199', '317', '253', '234', '145', '102', '63', '279', '241', '201', '4', '29', '70', '99', '60', '168', '192', '117', '124', '303', '268', '200', '131', '243', '27', '247', '69', '3', '71', '304', '120', '186', '287', '173', '197', '289', '309', '215', '129', '217', '162', '114', '180', '175', '161', '218', '245', '293', '22', '82', '45', '290', '37', '18', '142', '48', '106', '128', '233', '149', '181', '28', '135', '261', '138', '111', '284', '210', '65', '174', '164', '311', '101', '226', '150', '76', '136', '104', '238', '228', '278', '165', '191', '312', '254', '209', '75', '143', '220', '25', '53', '292', '134', '179', '98', '140', '285', '308', '14', '171', '16', '315', '32', '276', '242', '121', '203', '170', '207', '51', '1', '62', '158', '190', '23', '52', '88', '227', '283', '41', '94', '148', '255', '280', '302', '49', '83', '267', '208', '151', '72', '126', '264', '221', '294', '307', '250', '10', '123', '122', '21', '266', '119', '231', '248', '116', '232', '68', '296', '297', '40', '184', '240', '286', '198', '160', '230', '305'}) and 7 missing columns ({'OT', 'MUFL', 'LULL', 'HULL', 'HUFL', 'MULL', 'LUFL'}).
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/Diaugeia/TSEval-Static/electricity.csv (at revision 1e76820425cfc08d4eb8886cbe75affd82edca3c), ['hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/ETTh1.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/ETTh2.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/ETTm1.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/ETTm2.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/electricity.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/traffic.csv', 'hf://datasets/Diaugeia/TSEval-Static@1e76820425cfc08d4eb8886cbe75affd82edca3c/weather.csv']
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

date
string
HUFL
float64
HULL
float64
LUFL
float64
LULL
float64
MUFL
float64
MULL
float64
OT
float64
2016-07-01 00:00:00
5.827
2.009
4.203
1.34
1.599
0.462
30.531
2016-07-01 01:00:00
5.693
2.076
4.142
1.371
1.492
0.426
27.787001
2016-07-01 02:00:00
5.157
1.741
3.777
1.218
1.279
0.355
27.787001
2016-07-01 03:00:00
5.09
1.942
3.807
1.279
1.279
0.391
25.044001
2016-07-01 04:00:00
5.358
1.942
3.868
1.279
1.492
0.462
21.948
2016-07-01 05:00:00
5.626
2.143
4.051
1.371
1.528
0.533
21.174
2016-07-01 06:00:00
7.167
2.947
5.026
1.858
2.132
0.782
22.792
2016-07-01 07:00:00
7.435
3.282
5.087
2.224
2.31
1.031
23.143999
2016-07-01 08:00:00
5.559
3.014
2.955
1.432
2.452
1.173
21.667
2016-07-01 09:00:00
4.555
2.545
2.68
1.371
1.919
0.817
17.445999
2016-07-01 10:00:00
4.957
2.545
2.955
1.492
1.99
0.853
19.979
2016-07-01 11:00:00
5.76
2.545
3.442
1.492
2.203
0.853
20.118999
2016-07-01 12:00:00
4.689
2.545
2.833
1.523
1.812
0.853
19.205
2016-07-01 13:00:00
4.689
2.679
3.107
1.614
1.777
1.244
18.572001
2016-07-01 14:00:00
5.09
2.947
2.559
1.432
2.452
1.35
19.556
2016-07-01 15:00:00
5.09
3.148
2.589
1.523
2.487
1.35
17.305
2016-07-01 16:00:00
4.22
2.411
2.619
1.492
1.706
0.782
19.486
2016-07-01 17:00:00
4.756
2.344
3.076
1.492
1.635
0.711
19.134001
2016-07-01 18:00:00
5.626
2.88
3.076
1.492
2.523
1.208
20.681999
2016-07-01 19:00:00
5.492
3.014
3.015
1.553
2.452
1.208
18.712
2016-07-01 20:00:00
5.358
3.014
2.863
1.523
2.452
1.208
17.868
2016-07-01 21:00:00
5.09
2.947
2.68
1.523
2.381
1.208
18.009001
2016-07-01 22:00:00
4.823
2.947
2.619
1.523
2.203
1.173
18.009001
2016-07-01 23:00:00
4.622
2.88
2.467
1.492
2.132
1.137
19.768
2016-07-02 00:00:00
5.224
3.081
2.437
1.523
2.701
1.315
21.104
2016-07-02 01:00:00
5.157
3.014
2.345
1.432
2.878
1.35
19.697001
2016-07-02 02:00:00
5.157
3.148
2.284
1.432
2.878
1.492
20.049
2016-07-02 03:00:00
5.157
3.081
2.193
1.401
2.914
1.492
20.752001
2016-07-02 04:00:00
4.555
3.081
2.193
1.401
2.452
1.492
21.385
2016-07-02 05:00:00
5.425
3.282
2.437
1.462
3.092
1.706
22.23
2016-07-02 06:00:00
5.492
3.282
2.985
1.462
2.523
1.492
20.26
2016-07-02 07:00:00
5.626
3.215
3.076
1.523
2.487
1.492
21.104
2016-07-02 08:00:00
5.559
3.282
2.924
1.523
2.594
1.67
20.612
2016-07-02 09:00:00
5.224
3.215
2.68
1.462
2.559
1.564
18.361
2016-07-02 10:00:00
9.913
4.957
3.046
1.553
6.645
3.305
20.962999
2016-07-02 11:00:00
11.788
5.425
3.686
1.675
8.173
2.523
19.416
2016-07-02 12:00:00
9.645
4.957
3.107
1.828
6.752
2.132
20.823
2016-07-02 13:00:00
10.382
5.76
2.985
1.767
7.462
2.559
20.190001
2016-07-02 14:00:00
8.774
4.689
2.894
1.919
6.112
2.025
21.315001
2016-07-02 15:00:00
10.449
5.157
2.772
1.736
6.965
2.452
22.018999
2016-07-02 16:00:00
9.846
4.823
2.894
1.767
7.036
2.665
20.681999
2016-07-02 17:00:00
9.913
4.823
3.229
1.736
6.894
2.416
25.466
2016-07-02 18:00:00
10.65
4.689
3.381
1.797
6.929
2.452
25.888
2016-07-02 19:00:00
10.114
4.354
3.107
1.736
6.645
1.812
27.857
2016-07-02 20:00:00
9.98
4.153
3.411
1.767
6.574
1.954
27.295
2016-07-02 21:00:00
9.31
4.22
3.229
1.858
6.005
2.132
22.23
2016-07-02 22:00:00
9.444
4.622
2.955
1.858
6.965
2.168
21.948
2016-07-02 23:00:00
9.444
4.287
2.589
1.736
6.823
2.559
27.295
2016-07-03 00:00:00
10.382
5.425
2.955
1.675
7.604
2.31
29.334999
2016-07-03 01:00:00
9.779
5.224
2.65
1.675
6.716
2.843
26.028
2016-07-03 02:00:00
10.382
4.689
2.985
1.858
7.32
2.203
24.34
2016-07-03 03:00:00
9.779
4.153
2.528
1.675
6.823
1.99
26.450001
2016-07-03 04:00:00
10.717
4.756
2.65
1.797
7.356
2.807
25.958
2016-07-03 05:00:00
10.315
4.689
2.924
1.858
7.391
2.452
24.059
2016-07-03 06:00:00
12.592
5.224
3.716
1.949
8.671
2.203
25.325001
2016-07-03 07:00:00
11.119
4.622
3.625
1.919
7.889
2.843
23.636999
2016-07-03 08:00:00
10.65
4.421
3.594
1.919
7.036
2.025
26.379999
2016-07-03 09:00:00
10.047
4.22
3.686
1.949
6.432
1.67
27.365
2016-07-03 10:00:00
11.721
5.09
3.564
1.858
7.889
2.559
28.068001
2016-07-03 11:00:00
12.123
5.358
4.082
1.919
8.066
2.487
29.475
2016-07-03 12:00:00
9.98
5.023
3.29
1.858
6.858
2.559
26.802
2016-07-03 13:00:00
9.243
4.957
3.137
1.888
6.29
2.63
29.968
2016-07-03 14:00:00
10.181
5.425
3.076
1.888
7.178
3.02
30.389999
2016-07-03 15:00:00
9.645
5.425
3.015
1.828
7.107
2.665
31.164
2016-07-03 16:00:00
9.779
4.89
3.076
2.01
6.503
2.985
29.757
2016-07-03 17:00:00
11.119
5.157
3.807
1.98
7.32
2.914
32.289001
2016-07-03 18:00:00
11.052
4.957
3.686
1.98
7.391
2.523
31.938
2016-07-03 19:00:00
10.784
4.89
3.594
1.888
7.214
2.487
28.561001
2016-07-03 20:00:00
11.186
4.89
3.96
1.919
7.178
2.345
21.525999
2016-07-03 21:00:00
10.449
4.89
3.807
2.041
6.61
2.31
22.23
2016-07-03 22:00:00
9.578
5.76
3.259
1.888
6.787
3.127
19.416
2016-07-03 23:00:00
9.31
5.76
3.168
1.888
6.61
3.056
18.572001
2016-07-04 00:00:00
9.913
5.894
3.015
1.858
6.254
2.63
21.667
2016-07-04 01:00:00
8.975
4.957
2.863
1.828
6.29
2.665
25.535999
2016-07-04 02:00:00
8.64
4.823
2.924
1.828
6.148
2.594
27.857
2016-07-04 03:00:00
9.176
5.492
2.863
1.858
5.579
2.381
27.927999
2016-07-04 04:00:00
9.109
4.823
2.772
1.797
5.65
2.523
24.621
2016-07-04 05:00:00
9.846
5.559
3.107
1.888
5.97
2.949
23.848
2016-07-04 06:00:00
11.588
5.425
3.807
1.98
7.391
2.807
23.073999
2016-07-04 07:00:00
11.788
6.095
3.899
2.041
7.214
2.985
22.511
2016-07-04 08:00:00
10.583
5.961
3.655
2.071
7.143
2.914
21.667
2016-07-04 09:00:00
11.588
6.296
3.472
2.01
7.569
3.056
25.395
2016-07-04 10:00:00
11.922
6.229
3.746
1.949
7.711
3.056
25.184
2016-07-04 11:00:00
12.324
5.559
4.203
1.98
8.422
3.234
29.546
2016-07-04 12:00:00
10.382
5.894
3.564
1.949
6.858
2.63
29.475
2016-07-04 13:00:00
10.047
5.425
3.32
1.949
6.752
3.02
29.264
2016-07-04 14:00:00
10.516
6.028
3.137
1.919
7.107
3.376
30.952999
2016-07-04 15:00:00
10.717
6.095
3.168
2.01
6.787
3.02
31.726
2016-07-04 16:00:00
9.98
5.023
3.442
2.041
6.503
2.559
33.132999
2016-07-04 17:00:00
11.32
5.09
3.868
2.041
7.356
2.452
28.983
2016-07-04 18:00:00
11.387
4.957
4.295
2.193
7.356
2.452
28.983
2016-07-04 19:00:00
9.377
3.885
2.467
1.188
6.894
2.239
31.726
2016-07-04 20:00:00
10.114
4.086
2.955
1.462
7.143
2.239
25.184
2016-07-04 21:00:00
10.382
4.823
3.503
2.01
6.894
2.31
30.531
2016-07-04 22:00:00
9.645
4.89
3.259
1.919
6.61
1.919
27.646
2016-07-04 23:00:00
12.726
6.497
3.168
1.98
9.346
3.482
25.466
2016-07-05 00:00:00
11.989
5.626
3.198
1.98
8.777
2.949
25.958
2016-07-05 01:00:00
12.525
6.296
3.137
2.01
8.955
3.163
25.958
2016-07-05 02:00:00
12.324
6.296
2.985
1.919
8.813
3.376
26.028
2016-07-05 03:00:00
10.717
5.425
2.833
1.858
8.066
2.878
28.913
End of preview.

TSEval-Static

This repository holds the static forecasting benchmarks for TSEval — an open, reproducible leaderboard for time-series forecasting. It is the data side of the static track: the fixed, public datasets that every static submission is evaluated on.

TSEval ranks community submissions transparently across tracks, datasets, and horizons. Each entry is one agent trajectory plus one verified result. This repo provides the ground-truth inputs for the static track, so every number on the board traces back to the same data.

Live leaderboard: diaugeia.ai/tseval

Role in TSEval

TSEval has two tiers of tracks:

  • Static — fixed public benchmark data. The subject of this repo.
  • RealTime — periodically-refreshed live datasets. The first is the CSI-300 (沪深300) stock index, hosted in TSEval-RealTime.

The static track is split by task mode into three groups:

  • time_series — plain univariate / multivariate forecasting on the standard long-sequence benchmarks. Present today.
  • spatiotemporal — forecasting with explicit spatial / graph structure. Planned.
  • covariate — forecasting with exogenous covariates. Planned.

Right now this repo contains the time_series group only. The other two groups are coming.

Files present today

The time_series group ships the standard public long-term time-series forecasting (LTSF) benchmarks, as CSVs:

Dataset Description
ETTh1 Electricity Transformer Temperature, hourly, station 1
ETTh2 Electricity Transformer Temperature, hourly, station 2
ETTm1 Electricity Transformer Temperature, 15-minute, station 1
ETTm2 Electricity Transformer Temperature, 15-minute, station 2
electricity Hourly electricity consumption across clients
traffic Road occupancy rates from highway sensors
weather Local meteorological indicators
solar Solar power production records

These are the same public benchmarks used across the LTSF literature. We host them here unchanged so that static-track submissions read identical inputs, and so the data splits and evaluation are fixed by the framework rather than redefined per run.

Layout

The repo is laid out by task mode, one directory per static group:

time_series/      # present: ETTh1, ETTh2, ETTm1, ETTm2, electricity, traffic, weather, solar
spatiotemporal/   # planned
covariate/        # planned

New datasets and task modes are added by appending directories. Existing files do not move.

How it fits together

The static data here is the input. The producer framework, the submission contract, and the leaderboard sit around it:

  • ModernTSF — the producer framework. Clone it, run experiments through the tsf CLI, and capture the agent's trajectory. Static datasets in this repo are wired in directly. github.com/Diaugeia/ModernTSF
  • Submission contract — each submission is a trajectory.jsonl, one schema-valid RunRecord, and a short human-readable report. The schema is defined by TSF-Core (a pydantic-only layer inside ModernTSF) and exported as JSON Schema; the leaderboard reads only that schema.
  • Leaderboard build — deterministic CI (tsf leaderboard-build, no torch). It reads every submission, checks that the result and trajectory are present and schema-valid, then collates and ranks per (track, dataset, horizon) by MSE.
  • Reproducible by construction — trained checkpoints are archived separately and referenced by sha256, so a number is always traceable to the exact weights that produced it.

To participate: clone ModernTSF, run experiments via the tsf CLI, capture the trajectory with tsf trace, then tsf submit --push to open a community PR on the Submissions dataset.

The TSEval repos

License

MIT. The benchmark CSVs are standard public LTSF datasets, redistributed here for reproducible evaluation.
Downloads last month
-
Total file size:
214 MB

Space using Diaugeia/TSEval-Static 1