LeelaChessZero/lczero-training

Issues

• Functions get_weights() and fill_net() in tf/net.py not up-to-date

  #226 opened 4 months ago by zz4032
  1

• which python version is required

  #195 opened 3 years ago by arch-user-france1
  2

• ZeroDivisionError in squeeze_excitation function due to SE_ratio being set to zero

  #224 opened a year ago by ChinChangYang
  1

• Question about the meaning of input shape and output shape

  #223 opened a year ago by tianyu-z
  2

• How to load the weight pb file to tensorflow model?

  #221 opened a year ago by tianyu-z
  1

• 跑谱

  #219 opened a year ago by Biao-K
  0

• VarianceScaling Initializer Is Unseeded

  #218 opened 2 years ago by cn4750
  8

• Model cannot be saved in a saved model format

  #207 opened 2 years ago by Kovax007
  1

• Training without GPU

  #206 opened 2 years ago by alexmolas
  0

• Training cfg error

  #205 opened 2 years ago by omarzeb
  0

• Training script crashes after hitting a checkpoint

  #159 opened 3 years ago by magicianfromriga
  1

• assertion error

  #197 opened 3 years ago by arch-user-france1
  2

• No proto files after git clone execution

  #199 opened 3 years ago by teck45
  1

• question about training

  #198 opened 3 years ago by Flawy12
  0

• requirements file is broken

  #196 opened 3 years ago by arch-user-france1
  1

• Material Evaluation by LC0

  #157 opened 4 years ago by musketeerchess
  0

• leelalogs writing in train_input folder bug on colab

  #154 opened 4 years ago by teck45
  3

• Raise error on training data from different format versions

  #152 opened 4 years ago by YodaEmbedding
  1

• UnboundLocalError: 'mse_loss' is used before assignment when using `value: classical` in yaml file

  #136 opened 4 years ago by kennyfrc
  0

• http://lczero.org/training_data leads to 404

  #132 opened 4 years ago by junfengy
  0

• Softmax Policy Target

  #128 opened 5 years ago by oscardssmith
  0

• Training MLH off of result MLHZ and prediction MLHR as MLHRZ_RATIO

  #123 opened 5 years ago by TesseractA
  4

• decode_training.py does not support V5 training format data

  #126 opened 5 years ago by cn4750
  0

• 404 at training data link

  #122 opened 5 years ago by mhorton18
  6

• update training data section in readme

  #114 opened 5 years ago by skondrashov
  0

• ChunkParser fails with more than 1 worker in Python 3.7

  #75 opened 6 years ago by Lutzy
  0

• (Small) Possibility of Duplicated Training Games

  #72 opened 6 years ago by xghostxx
  1

• Training data is missing

  #49 opened 6 years ago by vishalkumarchaudhary
  1

• net_to_model.py - ValueError: Number of filters in YAML doesn't match the network

  #65 opened 6 years ago by namebrandon
  2

• Make batchnorm gammas trainable for reset

  #41 opened 6 years ago by gonzalezjo
  5

• Add ratio of weights:updates to TensorBoard

  #42 opened 6 years ago by cn4750
  0

• Input pipeline

  #44 opened 6 years ago by Pytq
  1

• Split script trim assumes that test and train are correct ratios before starting.

  #21 opened 6 years ago by Tilps
  2

• Discrepancy of batch norm to A0 paper

  #28 opened 6 years ago by Cyanogenoid
  4

• Report a 'train' number before step 1 like for 'test'.

  #27 opened 6 years ago by Tilps
  0

• Analyze blunders

  #5 opened 7 years ago by dubslow
  36

• (issue mover duplicate)

  #6 opened 7 years ago by dubslow
  35