pyg-team/pytorch-frame

Issues

• Potential Issue with the ExcelFormer Example

  #460 opened 3 months ago by mtreca
  1

• Drop Python 3.8 support

  #450 opened 3 months ago by akihironitta
  0

• BUG: Target_col count for each category was wrong

  #459 opened 3 months ago by hzm-January
  3

• Address `torch.compile` graph breaks in models

  #452 opened 4 months ago by akihironitta
  0

• Remove unnecessary CUDA synchronisations

  #453 opened 4 months ago by akihironitta
  0

• `AttributeError: module 'torch_frame' has no attribute 'multicategorical'`

  #440 opened 5 months ago by ValentinLo62
  4

• Use `torch.load(weights_only=True)` in the codebase

  #422 opened 5 months ago by akihironitta
  0

• EmbeddingBag.cu:143: EmbeddingBag_updateOutputKernel_sum_mean: block: [61,0,0], thread: [0,2,0] Assertion `input[emb] < numRows` failed.

  #439 opened 5 months ago by davidfstein
  3

• Wish to provide more methods and a roadmap

  #426 opened 5 months ago by qychen2001
  1

• TypeError in `Dataset.materialize` when ser.dtype is bool

  #430 opened 5 months ago by Kh4L
  1

• XGBoost did not use the best one iteration to predict

  #419 opened 5 months ago by qychen2001
  1

• Support for additional hyperparameters

  #416 opened 6 months ago by qychen2001
  0

• Support for pyspark dataframe

  #348 opened 6 months ago by saankhya-mondal
  2

• Support Multi-label Classification problem for `Dataset`

  #350 opened 6 months ago by XinweiHe
  0

• Fix broken Cohere embedding interface

  #406 opened 7 months ago by yiweny
  0

• Incorrect link in README Implemented Deep Tabular Models section

  #405 opened 7 months ago by Kksk43
  1

• Supports hyperparameter tuning for nn methods

  #404 opened 7 months ago by qychen2001
  2

• Problem in MutualInformationSort

  #388 opened 8 months ago by bernardocecchetto
  10

• Feature Importance

  #282 opened a year ago by xnuohz
  5

• Problem to reproduce similar results when saving the model

  #393 opened 8 months ago by bernardocecchetto
  0

• About Excelformer

  #387 opened 9 months ago by WhatAShot
  2

• Improve current the split and support more split methods

  #352 opened a year ago by XinweiHe
  0

• Error in Handling Heterogeneous Semantic Types (in documentation)

  #338 opened a year ago by preritt
  1

• Select rows in `TensorFrame` with a bool `Tensor` mask

  #331 opened a year ago by toenshoff
  4

• Split dataset by `timestamp`

  #335 opened a year ago by xnuohz
  1

• Support more `stype` in `LinearModelEncoder`

  #305 opened a year ago by weihua916
  0

• Try it with custom dataset

  #322 opened a year ago by sevimcengiz
  1

• Integration with TorchData

  #329 opened a year ago by nimaous
  1

• dataset download errors

  #314 opened a year ago by huyuaaaray
  3

• Can it be used for relational datasets?

  #312 opened a year ago by echatzikyriakidis
  6

• Make text finetuning version of Trompt, just like "transformers_text.py" for FTtransformer?

  #310 opened a year ago by hohoCode
  3

• feature importance frameworks integration (SHAP, Captum and etc)

  #313 opened a year ago by bernardocecchetto
  2

• Remove implicit clone in stype encoders

  #254 opened a year ago by yiweny
  0

• NaN masking not working in multicategorical stype

  #301 opened a year ago by yiweny
  0

• Fix `examples/mercari.py`

  #293 opened a year ago by akihironitta
  0

• Fix `examples/transformers_text.py`

  #294 opened a year ago by akihironitta
  2

• Fix the image of the architecture overview in our documentation

  #288 opened a year ago by akihironitta
  2

• Support saving/loading of GBDT models

  #242 opened a year ago by weihua916
  0

• Refactor advanced indexing in `MultiEmbeddingTensor` and `MultiNestedTensor`

  #235 opened a year ago by akihironitta
  0

• Support `LightGBM`

  #245 opened a year ago by xnuohz
  2

• Allow different `text_embedder_cfg` for each column

  #206 opened a year ago by toenshoff
  7

• Handle differnt `dtype` in `MultiNestedTensor`/`MultiEmbeddingTensor`

  #184 opened a year ago by akihironitta
  3

• Add ColumnWiseEncoder

  #232 opened a year ago by yiweny
  1

• Auto inference of `stype`

  #215 opened a year ago by weihua916
  0

• Clean up `TextTokenizer` logic

  #216 opened a year ago by weihua916
  4

• sequence numerical stats computation

  #195 opened a year ago by weihua916
  0

• Handle the case when input series is all nans's in MutualInformationSort

  #173 opened a year ago by yiweny
  0

• Add default NAStrategy to models

  #178 opened a year ago by yiweny
  0

• Handle the case when target_mean is nan in CatToNumTransform

  #172 opened a year ago by yiweny
  0

• Handle the case when a numerical sequence is all nans

  #169 opened a year ago by yiweny
  1