Commit History

Author SHA1 Message Date
  Matthias Reso cad284c66f Replace new model url 10 months ago
  Matthias Reso 8b0a233c1a Use new chat format in custom dataset 10 months ago
  Matthias Reso fac41298b0 Adapt test_custom_dataset to new model 10 months ago
  Matthias Reso 960014a3bb Fix test_custom_dataset by introducing a stable sort algorithm 10 months ago
  Matthias Reso 147aaa29bc Remove deprecated pytest_cmdline_preparse 1 year ago
  Matthias Reso 0022d97163 remove decapoda-research/llama-7b-hf tokenizer and skip tests if meta-llama/Llama-2-7b is not available 1 year ago
  Matthias Reso 53fd82355f Add missing changes 1 year ago
  Matthias Reso 8620ab8ac2 Fix invalid labels for context in custom dataset/oasst1 1 year ago
  Matthias Reso 52c417b7d5 Merge branch 'fix/invalidate_label_for_chat' into feature/length_based_batch_sampling 1 year ago
  Matthias Reso eafea7b366 Invalidate labels in dialog dataset to disable loss 1 year ago
  Matthias Reso 10f9367e56 fix missing labels in datasets 1 year ago
  Matthias Reso ca41c1c697 Adjust tests to len based batch sampling 1 year ago
  Matthias Reso eccf7cb3dc Check additional example in resulting dataset 1 year ago
  Matthias Reso ec00a2f722 Fix batching error 1 year ago
  Matthias Reso dc507b4e55 Finish implementation oasst preprocessing for of custom dataset 1 year ago
  Matthias Reso 26b9b7dbb2 Give an explicit error message if custom datset function is not found 1 year ago
  Matthias Reso 9cb3fddfcf Move datasets tests in their own subfolder 1 year ago