Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
num_rows
int64
498,535
500,368
499,097
500,278
499,500
498,159
499,659
500,237
499,420
498,986
499,224
499,386
498,921
499,912
499,490
500,539
498,326
500,739
500,350
499,600
500,772
500,654
499,562
498,368
499,565
498,890
499,538
498,882
500,522
499,299
499,276
499,363
499,566
499,205
500,256
498,725
499,716
500,189
499,560
499,591
500,982
499,416
500,337
497,796
498,284
501,117
499,724
499,758
500,523
499,577
499,771
499,338
497,945
499,181
499,621
500,470
499,592
498,749
500,198
501,108
499,903
500,341
499,848
499,552
499,543
499,912
500,140
498,888
500,068
499,323
499,988
499,403
499,904
499,919
498,857
498,116
499,492
498,035
499,459
499,220
499,926
499,358
499,298
498,051
499,678
499,631
500,106
500,157
499,293
500,539
498,764
500,005
499,212
498,650
498,576
498,857
498,427
499,437
498,916
499,421
End of preview. Expand in Data Studio

KataGo shuffled data

Generated from KataGo archive training data plus kata1/extradata/zhizi/ using KataGo's official shuffle script.

  • Input range: 2024-05-27 through 2026-06-07
  • Extra data: kata1/extradata/zhizi/
  • KataGo commit: 2a2fb4a44203f969e4ebe3dc46d96855d4bcd4fd
  • Total input rows: 2038453604
  • Validation split: md5-path [0.9999, 1.0000) (0.01%)
  • Training split: md5-path [0.0000, 0.9999) (99.99%)
  • Train shuffle: -num-waves 16 -num-processes 64 -approx-rows-per-out-file 500000 -approx-rows-per-bucket 500000 -worker-group-size 500000 -exclude-qvalues
  • Validation shuffle: -num-waves 1 -num-processes 64 -approx-rows-per-out-file 50000 -approx-rows-per-bucket 500000 -worker-group-size 500000 -exclude-qvalues
  • Training kept rows: unknown
  • Validation kept rows: unknown
  • Training output files: 8160
  • Validation output files: 20

Output is under shuffleddata/katago_20240527_20260607_zhizi_20260610-150619/.

Downloads last month
44