search

huggingface / transformers

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
https://huggingface.co/transformers
Apache License 2.0
133.97k stars 26.79k forks source link

add doctests for data2VecText #17338

Closed artemisep closed 2 years ago

artemisep commented 2 years ago

Feature request

Enable doctests for data2VecText model, as part of https://github.com/huggingface/transformers/issues/16292

Motivation

please see https://github.com/huggingface/transformers/issues/16292

Your contribution

implement this feature

patrickvonplaten commented 2 years ago

Hey @Ruihua-Fang,

Would you like to give it a try? :-)

artemisep commented 2 years ago

Hey @patrickvonplaten , yep, Thanks :)

artemisep commented 2 years ago

Following the instruction in https://github.com/huggingface/transformers/issues/16292 as listed below: Make sure to run the doc example doc test locally as described in https://github.com/huggingface/transformers/tree/master/docs#for-python-files 5 failed, 2 passed see attached file for detailed doctest_data2vec_text_errormsg.txt error messages

p.s, for sanity check, I also run the doctest sample for the following:

bigbird_pegasus: all 5 tests passed data2vec_audio in the same folder: 1 failed, 4 passed

error message for data2vec_audio: [doctest] transformers.models.data2vec.modeling_data2vec_audio.Data2VecAudioForAudioFrameClassification.forward _____ 1420 heads. 1421 1422 Example: 1423 1424 ```python 1425 >>> from transformers import Wav2Vec2FeatureExtractor, Data2VecAudioForAudioFrameClassification 1426 >>> from datasets import load_dataset 1427 >>> import torch 1428 1429 >>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation") Expected nothing Got: Downloading and preparing dataset librispeech_asr/clean to /home/ruihua/.cache/huggingface/datasets/hf-internal-testing___librispeech_asr/clean/2.1.0/d3bc4c2bc2078fcde3ad0f0f635862e4c0fef78ba94c4a34c4c250a097af240b... Dataset librispeechasr downloaded and prepared to /home/ruihua/.cache/huggingface/datasets/hf-internal-testinglibrispeech_asr/clean/2.1.0/d3bc4c2bc2078fcde3ad0f0f635862e4c0fef78ba94c4a34c4c250a097af240b. Subsequent calls will reuse this data.

/home/ruihua/project/huggingface/tf/transformers/src/transformers/models/data2vec/modeling_data2vec_audio.py:1429: DocTestFailure

github-actions[bot] commented 2 years ago

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.