How severe does this issue affect your experience of using Ray?
- High: It blocks me to complete my task.
I would reply on this API to generated tfrecord files with metadata. My application direclty processed it in tensorflow.
How severe does this issue affect your experience of using Ray?
I would reply on this API to generated tfrecord files with metadata. My application direclty processed it in tensorflow.
I believe Dataset.write_tfrecords
should already be writing .tfrecords
files. Are you using any custom logic, like FilenameProvider
or a custom datasource? ray.data.Dataset.write_tfrecords — Ray 2.34.0