|
Is it Possible to modify the zero-shot classier?
|
|
0
|
16
|
May 24, 2022
|
|
How to represent paginated documents as a single instance of training data for whole document classification?
|
|
5
|
145
|
May 24, 2022
|
|
Zero-Shot Classification
|
|
0
|
16
|
May 24, 2022
|
|
How to order sentences based on pairwise probabilities?
|
|
0
|
20
|
May 24, 2022
|
|
Inference result is SequenceClassifierOutput instance?
|
|
0
|
17
|
May 24, 2022
|
|
The best way to load pertained weights then continue training BERT?
|
|
0
|
33
|
May 23, 2022
|
|
Training and evaluation loss goes down however, WER score stays the same
|
|
0
|
30
|
May 23, 2022
|
|
Using Trainer at inference time
|
|
7
|
1321
|
May 23, 2022
|
|
Text classifier is trained incorrectly using BERT transformers (f1 = 0) for a certain amount of dataset
|
|
0
|
36
|
May 22, 2022
|
|
Vision Transformer reconstruct image
|
|
0
|
39
|
May 22, 2022
|
|
EncoderDecoderModel for Machine Translation
|
|
0
|
58
|
May 21, 2022
|
|
Cannot use the new model built
|
|
1
|
70
|
May 21, 2022
|
|
The best way to install and edit the transformers package locally?
|
|
2
|
74
|
May 21, 2022
|
|
Multi-label token classification
|
|
23
|
510
|
May 20, 2022
|
|
How to add additional module to BERT architecture, then load the original weight and use it
|
|
0
|
59
|
May 20, 2022
|
|
Error in fine tuning T5 model for Seq2Seq translation task
|
|
0
|
71
|
May 20, 2022
|
|
Logging which decoder selected in generation
|
|
0
|
79
|
May 19, 2022
|
|
Logging & Experiment tracking with W&B
|
|
68
|
8377
|
May 18, 2022
|
|
Inference API offline model limit
|
|
0
|
70
|
May 18, 2022
|
|
Swin transformer hidden states( feature map) different
|
|
0
|
84
|
May 18, 2022
|
|
Lower Memory Usage for TF GPT-J
|
|
0
|
91
|
May 17, 2022
|
|
New pipeline for zero-shot text classification
|
|
98
|
38788
|
May 17, 2022
|
|
Is the reported loss averaged over logging steps
|
|
2
|
91
|
May 17, 2022
|
|
How to input word2vec embeddings to gpt2 model?
|
|
0
|
89
|
May 17, 2022
|
|
ValueError: Mixed precision training with AMP or APEX (`--fp16` or `--bf16`) and half precision evaluation (`--fp16_full_eval` or `--bf16_full_eval`) can only be used on CUDA devices
|
|
0
|
87
|
May 17, 2022
|
|
Problem with Adding LayerNorm after BART's Encoder for Summarization
|
|
0
|
97
|
May 16, 2022
|
|
Export M2M100 model to ONNX
|
|
1
|
138
|
May 16, 2022
|
|
How to log the eval metrics every `eval_steps` to a file?
|
|
1
|
153
|
May 16, 2022
|
|
How to use 1 model for 2 downstream tasks?
|
|
0
|
101
|
May 16, 2022
|
|
How to represent paginated documents as a single training data instance
|
|
2
|
167
|
May 16, 2022
|