pet

pet copied to clipboard

Missing optimizer step

2

If max_steps or data length is not divisible by gradient_accumulation_steps some gradients are lost. Since updating only takes place at `if (step + 1) % gradient_accumulation_steps == 0:`

FlorisFok

I trained a model using pet and I want to evaluate it over some validation set. While I remove `--do_train` and only leave `--do_eval` option, it still loads training data...

puraminy

ZeroDivisionError when reduction is set to 'wmean' while training iPET

1

I ran the code with: ``` python cli.py \ --method ipet \ --data_dir ../dataset/data \ --model_type bert \ --model_name_or_path bert-base-cased \ --task_name my-task \ --output_dir ./bert_ipet_10_test \ --pattern_ids 0 1...

xruifan

I find there is no softmax function when I should get the distribution of prediction. ----------------- wrapper.py def mlm_train_step(self, labeled_batch: Dict[str, torch.Tensor], unlabeled_batch: Optional[Dict[str, torch.Tensor]] = None, lm_training: bool =...

bookpen

The program is stuck at "There are 0 examples for label [my_label]" while iPETs are choosing examples per label for the next generation. This is because there is no condition...

xruifan

Bumps [transformers](https://github.com/huggingface/transformers) from 3.0.2 to 4.30.0. Release notes Sourced from transformers's releases. v4.30.0: 100k, Agents improvements, Safetensors core dependency, Swiftformer, Autoformer, MobileViTv2, timm-as-a-backbone 100k Transformers has just reached 100k stars...

dependabot[bot]

In the paper “Automatically Identifying Words That Can Serve as Labels for Few-Shot Text Classification”, I do not quite understand the following. Could you please give me an explanation?

Ywandung-Lyou

Clarification on how to interpret PET's results

2

When running PET on my task I obtain two "results_text" files with different metric values. One is in the main directory, while the other is in the "final" directory. I...

aliromagnoli

How to reproduce results of the paper？

1

I read the paper and downloaded the AG news dataset，and tested PET model on it，but there is a great margin between my results and the author's results. I set parameters...

Devil-Ideal

nieallen