dmlc/gluon-nlp

Issues

• ImportError [...]/lib/python3.10/site-packages/gluonnlp/data/fast_bert_tokenizer.cpython-310-x86_64-linux-gnu.so: undefined symbol: _PyGen_Send

  #1595 opened 2 years ago by MBrede
  5

• nlp version error

  #1597 opened 2 years ago by IluvRiver
  1

• Installation Error in google Collab env with Python3.7

  #1598 opened 2 years ago by AbdirayimovS
  1

• [Document] Improve documentation about the conversion toolkit

  #1511 opened 4 years ago by sxjscience
  1

• Problems not being imported from colab

  #1596 opened 2 years ago by stratosphere1492
  1

• (not a bug) question about bert `create_pretraining_data.tokenize_lines()`

  #1592 opened 2 years ago by kiukchung
  1

• Wheel fails to build setup.py for gluonnlp

  #1589 opened 2 years ago by hb0313
  2

• [Error Message] Improve error message in SentencepieceTokenizer when arguments are not expected.

  #1449 opened 4 years ago by preeyank5
  7

• Upgrade the version of WikiExtractor used in nlp_data

  #1533 opened 4 years ago by sxjscience
  2

• Wrong ETA for max_seq_length != 512

  #1586 opened 3 years ago by bartekkuncer
  0

• difference between gluonnlp 0.10.0 and gluonnlp 1.0.0 RoBERTaModel?

  #1563 opened 3 years ago by makua-bernal
  0

• Implementation of Incremental Decoding

  #1582 opened 3 years ago by barry-jin
  0

• I want BigRNN back...

  #1577 opened 3 years ago by yifeim
  5

• Difference in naming convention between op.py and test_op.py

  #1580 opened 3 years ago by AetherPrior
  0

• clean_tok_mono_corpus

  #1520 opened 3 years ago by araitats
  1

• Merge clean_tok_para_corpus and clean_tok_mono_corpus

  #1490 opened 3 years ago by sxjscience
  0

• Operator npx.broadcast_like

  #1552 opened 3 years ago by barry-jin
  3

• GPT2 tests mysteriously killed

  #1567 opened 4 years ago by barry-jin
  3

• XLMR Conversion fails on parameter dictionary KeyError

  #1570 opened 4 years ago by davisliang
  3

• gluon-nlp/scripts/question_answering/train.py

  #1569 opened 4 years ago by rrjia
  2

• Badge link was broken in README

  #1564 opened 4 years ago by zheyuye
  0

• Dead link to build_steps.groovy in contribute.rst#make-changes

  #1562 opened 4 years ago by sfraczek
  0

• [v0.x] Some CPU tests fail after switching to use MXNet 1.8.0

  #1559 opened 4 years ago by barry-jin
  1

• Loading 'distilbert_6_768_12' is broken

  #1549 opened 4 years ago by craffel
  2

• Upgrade sacremoses to 0.0.44

  #1548 opened 4 years ago by barry-jin
  0

• nlp.models.bert.get_pretrained_bert provides slow tokenizer

  #1538 opened 4 years ago by leezu
  3

• Pre-training scripts should allow resuming from checkpoints

  #1526 opened 4 years ago by szha
  3

• [Website] Improve website of the master version to prepare for the 1.0 release

  #1472 opened 4 years ago by sxjscience
  8

• v0.x CI is broken

  #1529 opened 4 years ago by leezu
  2

• strip_accents should be None by default in WordPiece

  #1528 opened 4 years ago by sxjscience
  3

• Issues in ELECTRA-base pre-training and fine-tuning

  #1505 opened 4 years ago by szha
  3

• Downloading problem of model weights

  #1522 opened 4 years ago by sxjscience
  1

• gluon-nlp/scripts/processing/learn_subword.py

  #1519 opened 4 years ago by araitats
  0

• Have problom in BERT pre-training: how to training on multiple GPUs

  #1508 opened 4 years ago by yangshuo0323
  13

• [TVM Integration] Support TVM conversion with FP16 data type

  #1510 opened 4 years ago by sxjscience
  0

• [Backbone] Add DeBERTa

  #1507 opened 4 years ago by sxjscience
  0

• [API][Rename] Rename T5NMTInference to T5Inference

  #1501 opened 4 years ago by sxjscience
  5

• BERT pretraining loss and accuracy is not stabilized

  #1497 opened 4 years ago by rondogency
  4

• Incorrect link to the mentioned book in the README.rst file

  #1465 opened 4 years ago by realBjornRoden
  2

• [Bug][Conversion Tools] mxnet block.py Overriding Parameter attribute %s is not allowed

  #1485 opened 4 years ago by makua-bernal
  4

• Add version check to mxnet and move all files related to mxnet to a separate namespace

  #1486 opened 4 years ago by sxjscience
  0

• [Example] Add summarization example with finetuning BART model

  #1484 opened 4 years ago by sxjscience
  2

• [Bug] Bug of top-p sampling

  #1481 opened 4 years ago by sxjscience
  0

• [Benchmark] Improve NLP Backbone Benchmark

  #1473 opened 4 years ago by sxjscience
  0

• [Enhancement] Add sequence-level distillation to NMT training

  #1463 opened 4 years ago by sxjscience
  0

• BERT/Eelectra vocab file and tokenizer model.

  #1451 opened 4 years ago by araitats
  1

• [Conversion Tools] Refactor conversion tools to support converting to/from huggingface

  #1458 opened 4 years ago by sxjscience
  0

• [CI][CD] Scheduled release to docker hub

  #1453 opened 4 years ago by sxjscience
  0

• [BUG] CPU Unittest (macos-python3.7) Failed for ArrowTypeError

  #1455 opened 4 years ago by barry-jin
  6

• [Tests] It takes long time to test full suite of backbone models

  #1445 opened 4 years ago by barry-jin
  5