imbalanced-dataset-sampler

imbalanced-dataset-sampler copied to clipboard

ValueError: Cannot set a frame with no defined index and a value that cannot be converted to a Series

1

Hi, I am using BERT for multi label classification. The dataset is imbalance and I use ImbalancedDatasetSampler as the sampler. The train data has been tokenized, has id, mask and...

zhangjizby

updates: - [github.com/pre-commit/pre-commit-hooks: v4.1.0 → v5.0.0](https://github.com/pre-commit/pre-commit-hooks/compare/v4.1.0...v5.0.0) - [github.com/PyCQA/isort: 5.10.1 → 5.13.2](https://github.com/PyCQA/isort/compare/5.10.1...5.13.2) - https://github.com/pre-commit/mirrors-yapf → https://github.com/google/yapf - [github.com/google/yapf: v0.32.0 → v0.40.2](https://github.com/google/yapf/compare/v0.32.0...v0.40.2) - [github.com/PyCQA/flake8: 4.0.1 → 7.1.1](https://github.com/PyCQA/flake8/compare/4.0.1...7.1.1)

pre-commit-ci[bot]

Suggestion: Change y limits of learning curves in README

1

If I can make one small quick suggestion, perhaps the two following plots in your README should have the same y axis limits. 

RylanSchaeffer

A new sampler for your reference

1

https://github.com/zzw-zwzhang/Yoneed/blob/main/sampler.py#L15

zzw-zwzhang

Subset sampling entire dataset

1

Hi everyone, I have a question concering using subsets with this sampler. According to the code it chooses samples from all entries in the parent dataset: https://github.com/ufoym/imbalanced-dataset-sampler/blob/e9dd2deca6e058771533678b29b38a60843b0a85/torchsampler/imbalanced.py#L49-L50 Shouldn't it only...

dajopr

I have added a few lines that allow to work with pytorch-geometric dataset. Since Pytorch-geometric data is saved as a list before being loaded by a Pytorch-geometric Dataloader, the modification...

avarbella

callback_get_label no longer works as described

3

`callback_get_label: a callback-like function which takes two arguments - dataset and index` This [no longer seems to be the case](https://github.com/ufoym/imbalanced-dataset-sampler/commit/ad50e22240c5316f8855db972542e1e2c388f839#diff-2df1f4597623f25e583732eb253eba0b08914c4a36d219d19dda17b9c7b3932dR42)? Please update how the new use-case looks like, because above...

zimonitrome

Does val_dataset need balanced sampler?

1

I don't know if the val_dataset needle a balance. Thanks.

Joevaen

ConcatDataset support

1

Thanks for the great work! I try to combine two datasets by using "dataset = dataset1+dataset2", and it gives me such error: AttributeError: 'ConcatDataset' object has no attribute 'get_labels' Is...

jasonbian97

ERROR label_to_count not callable

1

Hi ! I noticed that they are some bugs introduce with the last commit [ad50e22](https://github.com/ufoym/imbalanced-dataset-sampler/commit/ad50e22240c5316f8855db972542e1e2c388f839) Step to reproduce ` import torch from torchsampler import ImbalancedDatasetSampler mnist = torchvision.datasets.MNIST('.', train=True, download=True,...

samsja