Unable to load ted_hrlr_translate/pt_to_en dataset

user_10933724 0
Viewed 355

I was running transformer[link] code in my jupyter notebook by looking at the google collab code. The tfds.load was able to load ted_hrlr_translate/pt_to_en dataset in collab but was unable in jupyter notebook. The error that hit the screen was :

Failed to construct dataset ted_hrlr_translate: Message type "tensorflow_datasets.DatasetInfo" has no field named "releaseNotes". Available Fields(except extensions): ['name', 'description', 'version', 'configName', 'configDescription', 'citation', 'sizeInBytes', 'downloadSize', 'location', 'downloadChecksums', 'schema', 'splits', 'supervisedKeys', 'redistributionInfo', 'moduleName', 'disableShuffling', 'fileFormat']

Is this a bug or I am too dumb to understand the error? Do I need to download the dataset? If so then i was able to load mnist data in jupyter notebook without downloading it.

transformer-model tensorflow-datasets jupyter-notebook tensorflow python
1 Answers

This did not show any error to me when I tried to import this dataset using same line of code in jupyter notebook with TensorFlow 2.7. However please let us know which TensorFlow version are you using to replicate the issue into the same.

Meanwhile, you can try with installing below packages before of importing datasets in jupyter notebook.

#For the stable version, released every few months.
!pip install tensorflow-datasets  

#Released every day, contains the last versions of the datasets.
!pip install tfds-nightly
user_142906810
Related