Hugging Face's logo

Datasets:
Wendelia
/
ohlcv-br-ativos

Tasks:
Time Series Forecasting
Languages:
Portuguese
Size:
1K<n<10K
Tags:
finance
brazilian-stocks
timesfm
ohlcv
technical-indicators
lora-finetuning
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 16 new columns ({'rsi14', 'macd', 'retorno', 'bb_inferior', 'log_ret', 'volume_rel', 'macd_hist', 'sma50', 'sma200', 'volume_medio_20d', 'sma20', 'macd_sinal', 'bb_superior', 'bb_medio', 'volatilidade_20d', 'range_pct'})

This happened while the csv dataset builder was generating data using

hf://datasets/Wendelia/ohlcv-br-ativos/dados_processados/ITUB4_tecnicos.csv (at revision 8bf0336774ff64679c72172c75e0874e8982c11c), [/tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/ITUB4_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/ITUB4_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/ITUB4_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/ITUB4_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/MGLU3_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/MGLU3_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/MGLU3_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/MGLU3_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/PETR4_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/PETR4_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/PETR4_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/PETR4_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/VALE3_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/VALE3_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/VALE3_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/VALE3_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/WEGE3_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/WEGE3_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/WEGE3_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/WEGE3_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/resultados_backtest.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/resultados_backtest.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/timesfm_lora_output/resultados_lora_teste.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/timesfm_lora_output/resultados_lora_teste.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 1800, 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 2321, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2249, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              date: string
              open: double
              high: double
              low: double
              close: double
              volume: int64
              split_ratio: double
              dividend: double
              sma20: double
              sma50: double
              sma200: double
              rsi14: double
              macd: double
              macd_sinal: double
              macd_hist: double
              bb_medio: double
              bb_superior: double
              bb_inferior: double
              retorno: double
              volatilidade_20d: double
              volume_medio_20d: double
              volume_rel: double
              log_ret: double
              range_pct: double
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 3048
              to
              {'date': Value('string'), 'open': Value('float64'), 'high': Value('float64'), 'low': Value('float64'), 'close': Value('float64'), 'volume': Value('int64'), 'split_ratio': Value('float64'), 'dividend': 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 1347, in compute_config_parquet_and_info_response
                  parquet_operations = convert_to_parquet(builder)
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
                  builder.download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 882, in download_and_prepare
                  self._download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 943, in _download_and_prepare
                  self._prepare_split(split_generator, **prepare_split_kwargs)
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1646, 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 1802, 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 16 new columns ({'rsi14', 'macd', 'retorno', 'bb_inferior', 'log_ret', 'volume_rel', 'macd_hist', 'sma50', 'sma200', 'volume_medio_20d', 'sma20', 'macd_sinal', 'bb_superior', 'bb_medio', 'volatilidade_20d', 'range_pct'})
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/Wendelia/ohlcv-br-ativos/dados_processados/ITUB4_tecnicos.csv (at revision 8bf0336774ff64679c72172c75e0874e8982c11c), [/tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/ITUB4_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/ITUB4_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/ITUB4_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/ITUB4_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/MGLU3_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/MGLU3_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/MGLU3_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/MGLU3_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/PETR4_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/PETR4_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/PETR4_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/PETR4_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/VALE3_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/VALE3_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/VALE3_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/VALE3_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/WEGE3_daily.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/WEGE3_daily.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/WEGE3_tecnicos.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/dados_processados/WEGE3_tecnicos.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/resultados_backtest.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/resultados_backtest.csv), /tmp/hf-datasets-cache/medium/datasets/13363388404475-config-parquet-and-info-Wendelia-ohlcv-br-ativos-99e5d250/hub/datasets--Wendelia--ohlcv-br-ativos/snapshots/8bf0336774ff64679c72172c75e0874e8982c11c/timesfm_lora_output/resultados_lora_teste.csv (origin=hf://datasets/Wendelia/ohlcv-br-ativos@8bf0336774ff64679c72172c75e0874e8982c11c/timesfm_lora_output/resultados_lora_teste.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
open
float64
high
float64
low
float64
close
float64
volume
int64
split_ratio
float64
dividend
float64
2022-01-03
18.684908
19.249779
18.614298
18.984995
37,222,335
0
0.015578
2022-01-04
19.082083
19.576345
18.914387
19.52339
49,793,763
0
0
2022-01-05
19.49691
19.61165
19.002647
19.152691
35,851,405
0
0
2022-01-06
19.099735
19.629301
18.99382
19.54104
40,624,734
0
0
2022-01-07
19.49691
19.991173
19.311562
19.97352
48,590,177
0
0
2022-01-10
19.920565
20.194174
19.691086
20.158871
47,125,888
0
0
2022-01-11
20.114738
20.609001
20.035303
20.609001
63,361,665
0
0
2022-01-12
20.529568
20.741394
20.238304
20.520741
47,231,371
0
0
2022-01-13
20.379522
21.02383
20.344217
20.900265
49,188,741
0
0
2022-01-14
20.803177
21.050308
20.661959
20.741394
33,902,872
0
0
2022-01-17
20.706091
20.962048
20.43248
20.741394
19,948,957
0
0
2022-01-18
20.644306
21.147396
20.538393
20.86496
49,389,056
0
0
2022-01-19
21.02383
21.253309
20.653133
20.723742
35,038,364
0
0
2022-01-20
20.776699
20.891438
20.458958
20.556046
29,391,946
0
0
2022-01-21
20.458958
20.873787
20.406
20.564873
42,676,371
0
0
2022-01-24
20.609001
20.785524
20.185349
20.538393
44,921,297
0
0
2022-01-25
20.353045
21.315092
20.300089
21.279787
44,585,362
0
0
2022-01-26
21.473963
21.624006
21.015003
21.509268
63,314,645
0
0
2022-01-27
21.615179
21.994703
21.359222
21.615179
47,070,145
0
0
2022-01-28
21.323917
21.897615
21.29744
21.879965
33,360,845
0
0
2022-01-31
21.774052
22.515446
21.641659
22.356575
50,129,358
0
0
2022-02-01
22.32127
22.674314
22.224182
22.497793
32,598,902
0
0.015578
2022-02-02
22.480141
22.515446
21.994703
22.144749
39,095,411
0
0
2022-02-03
22.241835
22.533096
21.809355
22.400705
47,261,735
0
0
2022-02-04
22.27714
22.736097
22.241835
22.471315
39,194,888
0
0
2022-02-07
22.462488
22.550751
22.268314
22.34775
26,166,974
0
0
2022-02-08
22.312445
22.700794
22.091791
22.594879
42,299,308
0
0
2022-02-09
22.021181
22.427183
21.685789
21.694616
100,922,428
0
0
2022-02-10
21.668137
22.338923
21.588701
22.109444
68,333,609
0
0
2022-02-11
23.25684
23.830538
23.097969
23.41571
182,038,996
0
0
2022-02-14
23.477493
23.556929
23.186232
23.362753
43,043,123
0
0
2022-02-15
23.654015
23.742277
23.389233
23.583406
33,336,032
0
0
2022-02-16
23.636362
23.715797
23.283318
23.336275
44,408,048
0
0
2022-02-17
23.30097
23.371578
22.912621
23.009708
38,194,109
0
0
2022-02-18
23.142099
23.433361
23.000881
23.097969
29,924,682
0
0
2022-02-21
23.150927
23.195057
22.268314
22.533096
25,401,293
0
0
2022-02-22
22.727272
23.071491
22.497793
22.727272
34,318,570
0
0.012056
2022-02-23
22.744925
23.018534
22.356575
22.409533
25,506,096
0
0
2022-02-24
21.84466
22.180052
21.482788
22.030008
49,889,615
0
0
2022-02-25
21.950573
22.506618
21.897615
22.480141
44,070,414
0
0
2022-03-02
22.215357
22.427183
21.985878
22.091791
36,789,303
0
0.015578
2022-03-03
22.100616
22.815533
21.800529
22.109444
50,467,105
0
0
2022-03-04
22.127096
22.180052
21.429832
21.774052
34,369,894
0
0
2022-03-07
21.421005
21.579876
21.200354
21.306267
59,892,079
0
0
2022-03-08
21.3857
21.588701
20.997353
21.103266
49,265,672
0
0
2022-03-09
21.36805
22.338923
21.323917
22.153574
58,459,174
0
0
2022-03-10
21.835835
22.171227
21.606354
22.171227
33,553,115
0
0
2022-03-11
22.268314
22.586054
21.871138
21.924095
48,649,207
0
0
2022-03-14
22.082966
22.541924
22.074139
22.233009
36,267,443
0
0
2022-03-15
22.162399
22.453663
21.800529
22.153574
41,720,798
0
0
2022-03-16
22.374228
22.647837
22.188879
22.586054
45,741,022
0
0
2022-03-17
22.630184
23.115622
22.471315
23.080317
49,108,185
0
0
2022-03-18
23.027361
23.477493
23.009708
23.248014
73,408,542
0
0
2022-03-21
23.336275
23.909973
23.265665
23.821712
40,745,625
0
0
2022-03-22
24.007059
24.298323
23.918798
24.148277
41,741,192
0
0
2022-03-23
24.174757
24.395411
23.980581
24.042364
30,594,059
0
0
2022-03-24
24.121799
24.474846
24.033539
24.201235
53,835,967
0
0
2022-03-25
24.289495
24.289495
23.945278
24.245365
36,342,901
0
0
2022-03-28
24.30715
24.457193
23.936451
24.227715
24,411,278
0
0
2022-03-29
24.457193
24.783758
24.174757
24.510151
33,838,857
0
0
2022-03-30
24.571932
24.677847
24.289495
24.563107
37,371,325
0
0
2022-03-31
24.607237
24.71315
24.086496
24.254192
46,340,719
0
0
2022-04-01
24.448366
24.607237
24.121799
24.30715
45,495,048
0
0.015578
2022-04-04
24.227715
24.28067
23.84819
24.086496
28,469,684
0
0
2022-04-05
24.033539
24.068844
23.495146
23.601059
41,644,661
0
0
2022-04-06
23.451015
23.751102
23.353928
23.592234
38,196,035
0
0
2022-04-07
23.398058
23.618711
23.309795
23.574579
26,732,002
0
0
2022-04-08
23.442188
23.742277
23.309795
23.530451
29,650,496
0
0
2022-04-11
23.477493
23.592234
23.336275
23.442188
30,882,294
0
0
2022-04-12
23.654015
23.742277
22.956751
23.027361
33,798,183
0
0
2022-04-13
23.195057
23.424536
22.921448
23.089144
38,370,517
0
0
2022-04-14
23.009708
23.362753
22.956751
23.239187
26,768,711
0
0
2022-04-18
23.159752
23.442188
23.133274
23.371578
23,802,743
0
0
2022-04-19
23.406883
23.486319
22.727272
22.956751
39,035,022
0
0
2022-04-20
22.965578
23.168577
22.82436
23.106796
28,485,092
0
0
2022-04-22
22.859665
22.965578
22.471315
22.656664
34,773,809
0
0
2022-04-25
22.506618
22.639011
22.144749
22.568401
42,282,427
0
0
2022-04-26
22.294792
22.506618
21.756397
21.800529
47,838,319
0
0
2022-04-27
22.056486
22.153574
21.535746
21.624006
27,688,480
0
0
2022-04-28
21.632833
21.703442
21.235657
21.447483
41,170,160
0
0
2022-04-29
21.588701
22.021181
21.067961
21.067961
50,954,295
0
0
2022-05-02
21.067961
21.253309
20.688437
20.962048
44,535,284
0
0.015578
2022-05-03
21.006178
21.323917
20.900265
21.067961
29,699,668
0
0
2022-05-04
20.970873
21.041483
20.653133
21.041483
53,745,441
0
0
2022-05-05
20.820829
20.820829
20.211826
20.556046
45,795,293
0
0
2022-05-06
20.547218
21.376875
20.538393
20.997353
68,490,643
0
0
2022-05-09
20.95322
21.094439
20.308914
20.697264
63,819,963
0
0
2022-05-10
20.820829
20.95322
20.450132
20.591351
40,331,174
0
0
2022-05-11
20.635481
21.376875
20.626654
21.279787
64,249,824
0
0
2022-05-12
21.120916
21.774052
21.120916
21.615179
61,754,278
0
0
2022-05-13
21.791702
21.994703
21.624006
21.871138
29,721,422
0
0
2022-05-16
21.93292
22.338923
21.729919
22.100616
30,939,624
0
0
2022-05-17
22.383053
22.612534
22.250662
22.294792
30,953,220
0
0
2022-05-18
22.224182
22.391878
21.924095
21.977053
43,092,975
0
0
2022-05-19
21.994703
22.180052
21.791702
21.906443
25,636,391
0
0
2022-05-20
22.144749
22.268314
21.93292
22.180052
43,072,014
0
0
2022-05-23
22.32127
22.947926
22.312445
22.753752
43,627,637
0
0
2022-05-24
22.665489
23.212709
22.550751
23.097969
44,474,328
0
0
2022-05-25
22.992056
23.062666
22.647837
22.683142
32,876,147
0
0
2022-05-26
22.70962
23.097969
22.700794
22.965578
39,868,230
0
0
End of preview.

OHLCV Brazilian Stocks with Technical Indicators

Dataset processado para análise preditiva de ações brasileiras usando TimesFM 2.5.

Ativos

Ticker Nome Período
PETR4 Petrobras PN 2022-05 a 2026-05
VALE3 Vale ON 2022-05 a 2026-05
ITUB4 Itaú Unibanco PN 2022-05 a 2026-05
WEGE3 WEG ON 2022-05 a 2026-05
MGLU3 Magazine Luiza ON 2022-05 a 2026-05

Estrutura

dados_processados/

  • {TICKER}_daily.csv — OHLCV bruto (open, high, low, close, volume, split_ratio, dividend)
  • {TICKER}_tecnicos.csv — OHLCV + indicadores técnicos (RSI14, MACD, SMA20/50, vol20, volume_rel)

dataset_lora_timesfm.csv / .parquet

Dataset consolidado para fine-tuning LoRA do TimesFM 2.5. Colunas:

  • ativo, date, open, high, low, close, volume
  • rsi14, macd, macd_sinal, sma20, sma50, vol20, volume_rel
  • close_norm, rsi14_norm, macd_norm, volume_rel_norm
  • 4405 registros, 19 colunas

resultados_backtest.csv

Backtesting baseline (modelo sem fine-tuning, 128 dias horizonte):

  • PETR4: MAPE 10.27%, VALE3: MAPE 7.62%, ITUB4: MAPE 14.13%
  • WEGE3: MAPE 11.32%, MGLU3: MAPE 24.17%

timesfm_lora_output/resultados_lora_teste.csv

Resultados pós fine-tuning LoRA (rank=8, alpha=16, 3 epochs, 0.70% params treináveis):

  • Melhorias: ITUB4 -3.8pp, MGLU3 -3.5pp, PETR4 -1.1pp, WEGE3 -0.8pp
  • Média geral: 11.9% vs 13.5% baseline (↓1.6pp)

Configuração LoRA

  • rank=8, alpha=16, dropout=0.05
  • target_modules: q_proj, k_proj, v_proj, o_proj (20 layers × 4 = 80 módulos)
  • Trainable: 1,638,400 / 232,927,680 (0.70%)
  • Contexto: 128 steps, Horizonte: 128 steps
  • Precision: float32 (RTX 4060 8GB)
  • LR: 5e-4, batch=1, grad_accum=8

Fonte dos dados

OpenBB (provider: yfinance) via Hermes Agent. Ações brasileiras identificadas pelo sufixo .SA.

Uso 

from datasets import load_dataset

ds = load_dataset("wendelia/ohlcv-br-ativos", data_files="dataset_lora_timesfm.parquet")

Licença

MIT
Downloads last month
45
Total file size:
2.84 MB