ThilinaRajapakse/simpletransformers

Issues

• Target is multiclass but average='binary'. Please choose another average setting, one of [None, 'micro', 'macro', 'weighted'].

  #1566 opened 2 months ago by MuvvaThriveni
  4

• Loading a fine tuned Seq2Seq MarianMT model gives wrong predictions

  #1571 opened 17 days ago by ziweizh24
  3

• Default learning rate on pretrained layers?

  #1570 opened a month ago by mehak25
  0

• Problem with the absence of attention_mask when using sliding_window

  #1565 opened 3 months ago by DarijaNS
  0

• Binary classification .predict raises a ValueError: could not broadcast input array from shape (2,2) into shape (1,2)

  #1563 opened 4 months ago by gaborfodor720818
  3

• change threshold value in predict for multi-label classification?

  #1562 opened 4 months ago by lececefifi
  1

• Unable to do hyperparamters search on sentence-pair classification(regression) task

  #1561 opened 4 months ago by Gregory5949
  0

• saving pytorch_model.bin instead of safetensors

  #1560 opened 5 months ago by oldaandozerskaya
  1

• RoBERTa model Prediction taking Infinite time with no error but Multiprocessing disabled warning

  #1558 opened 6 months ago by etqadkhan
  0

• Error with prediction using multilabel classification

  #1556 opened 6 months ago by willkattrup
  0

• argmax sliding_window tiebreaker bug

  #1554 opened 7 months ago by fenri-sulfr
  0

• RuntimeError: "nll_loss_forward_reduce_cuda_kernel_2d_index" not implemented for 'Int'

  #1553 opened 7 months ago by celsowm
  0

• Convert generate model into .tflite for token classification

  #1552 opened 7 months ago by Hir98
  0

• Update of macos to Sonoma 14.0 breaks train_model when using 'mps' device

  #1549 opened 7 months ago by drmatthewclark
  1

• `ValueError` occurred with `representation_model`

  #1548 opened 8 months ago by JINHXu
  0

• Error with pretrain T5 model for paraphrase

  #1521 opened a year ago by LeMoussel
  2

• `calculate_results` appears to only use the first correct answer, not all correct answers

  #1545 opened 9 months ago by kpounder
  0

• The training process cannot continue

  #1536 opened 10 months ago by xgySTATISICT
  7

• Memory Leak Issue

  #1544 opened 9 months ago by sanwark11
  1

• MMBT no longer supported by huggingface transformers.

  #1543 opened 10 months ago by nicholas-botzer
  1

• Version 0.63.11 not compatible with Transformers 4.31.0

  #1538 opened 9 months ago by owendw1
  3

• ModuleNotFoundError: No module named 'transformers.models.mmbt'

  #1539 opened 10 months ago by DeepakGaurAerospace
  12

• ValueError: multiclass-multioutput is not supported

  #1541 opened 10 months ago by zelekemarew
  6

• save_steps not working , checkpoint getting generated on every epoch morethan once

  #1542 opened 10 months ago by guruprasaad123
  2

• AttributeError: module 'fsspec.callbacks' has no attribute 'TqdmCallback'

  #1540 opened 10 months ago by ColinFerguson
  3

• How can we run NER task on t5 model is there any example for that ?

  #1537 opened 10 months ago by TekeshwarHirwani
  0

• can I get tokenization results during prediction

  #1535 opened a year ago by lececefifi
  0

• Ways to speed up training using CPU?

  #1534 opened a year ago by lececefifi
  0

• Change in predictions of BERT-QA when data is shuffled

  #1530 opened a year ago by soumyasj
  2

• Seq2seq: failure when evaluating while training

  #1522 opened a year ago by Futyn-Maker
  8

• for a while im trying to run this model and i'm getting this error

  #1519 opened a year ago by cehguru
  2

• `MultiModalClassificationModel` parameters descriptions are not informative

  #1502 opened a year ago by adhadseKavida
  1

• CUDA error: CUBLAS_STATUS_EXECUTION_FAILED during training a Multiclass Model with weights

  #1503 opened a year ago by simonkleinfeld
  2

• _sentencepiece.SentencePieceProcessor_LoadFromFile can't load existing file.

  #1504 opened a year ago by csanadpoda
  1

• TypeError: an integer is required (got type NoneType)

  #1506 opened a year ago by ijazul-haq
  1

• How to print model summary?

  #1511 opened a year ago by samfch
  1

• Why is cls_token_segment set to 1 and should it be set to 0?

  #1514 opened a year ago by peilongchencc
  1

• Can't train T5 for classification

  #1513 opened a year ago by RanaBan
  2

• TypeError: cached_path() got an unexpected keyword argument 'proxies'

  #1532 opened a year ago by MSaqibShah
  0

• TypeError when training T5 with evaluate_during_training=True

  #1524 opened a year ago by timmartin
  1

• model.train_model(train_df, eval_data=eval_df, **eval_metrics) is giving error "All arrays must be of the same length"

  #1531 opened a year ago by Moustafa-Banbouk
  2

• ELECTRA continual pre-training doesn't accept xlm-roberta type tokenizer.

  #1529 opened a year ago by aloka-fernando
  0

• Why is it so much faster to use simpletransformers than huggingface transformer when training

  #1520 opened a year ago by 645187919
  1

• Cannot import required library (Repost after multiple attempts at workarounds)

  #1526 opened a year ago by Fshrink
  2

• Can't load previously trained GPT-2 Language generation model

  #1527 opened a year ago by timmartin
  0

• interact method not loading trained model:

  #1525 opened a year ago by Fshrink
  0

• Cannot import ConvAIModel

  #1523 opened a year ago by Fshrink
  1

• "ModuleNotFound" error for simpletransformers.classification

  #1518 opened a year ago by schonlau
  2

• How do I not load the weights of a model?

  #1515 opened a year ago by chence08
  1

• error when loading a saved fine tunned model

  #1508 opened a year ago by RafieeAshkan
  2