|
Issues with translating inputs containing repeated phrases
|
|
1
|
1525
|
September 9, 2020
|
|
T5 weights auto conversion from TF to PyTorch
|
|
1
|
676
|
September 10, 2020
|
|
Merging bert-base-uncased models after trainer but before predict
|
|
6
|
2404
|
September 10, 2020
|
|
Inference API detailed request
|
|
5
|
2216
|
September 11, 2020
|
|
T5 Generates very short summaries
|
|
22
|
5524
|
September 11, 2020
|
|
What I know and don't know about sequence to sequence batching
|
|
3
|
2028
|
September 11, 2020
|
|
How to get weights indicating the importance of each words in a sentence corresponding to the label
|
|
2
|
1273
|
September 12, 2020
|
|
Current best practice for final linear classifier layer(s)?
|
|
3
|
2392
|
September 12, 2020
|
|
Generating NER data sample for RoBERTa model
|
|
3
|
469
|
September 14, 2020
|
|
Finetuning a specific task when pretrained model isn't trained on that specific task? Using the task model vs using the base model
|
|
4
|
1015
|
September 14, 2020
|
|
Unstable Reformer training on toy task
|
|
8
|
1117
|
September 14, 2020
|
|
Bart input confusion
|
|
2
|
3847
|
September 14, 2020
|
|
Saving underlying language model after trained on downstream task
|
|
0
|
420
|
September 14, 2020
|
|
Sharing BERT formatted corpus
|
|
7
|
1738
|
September 15, 2020
|
|
T5forConditionalGeneration
|
|
2
|
2212
|
September 15, 2020
|
|
[Proposal] Copy Pasting modeling_bart.py
|
|
4
|
600
|
September 15, 2020
|
|
Marian: Language Discovery questions
|
|
6
|
1570
|
September 15, 2020
|
|
Not all BLEU scores were created equal
|
|
0
|
313
|
September 15, 2020
|
|
Model() output issue during migration from pytorch_pretrained_bert to transformers
|
|
0
|
545
|
September 15, 2020
|
|
Fine-tune, or train from scratch?
|
|
6
|
3425
|
September 16, 2020
|
|
CircleCI+Github Actions: Which tests run where and when
|
|
2
|
401
|
September 16, 2020
|
|
New seq2seq tool: search hparam space with run_eval.py
|
|
5
|
347
|
September 17, 2020
|
|
[tool] easy branch rebase
|
|
0
|
311
|
September 17, 2020
|
|
How to return word replacements when returning masked word predictions?
|
|
0
|
608
|
September 17, 2020
|
|
How much memory is needed for training ByteLevelBPETokenizer?
|
|
3
|
1486
|
September 18, 2020
|
|
Pipeline with custom dataset tokenizer: when to save/load manually
|
|
18
|
5575
|
September 18, 2020
|
|
Using Pegasus Model for Transfer Learning is generating garbage summaries
|
|
2
|
704
|
September 18, 2020
|
|
Does the default weight_decay of 0.0 in transformers.AdamW make sense?
|
|
2
|
11407
|
September 18, 2020
|
|
Is_pretokenized argument for tokenizer doesn't work?
|
|
1
|
1780
|
September 18, 2020
|
|
How to train a language model from scratch when my dataset is bigger than RAM?
|
|
19
|
9731
|
September 18, 2020
|