Susan Xueqing Liu
Susan Xueqing Liu
If use ray: override _save_checkpoint, but add with tune.checkpoint_dir(...) to make ray keep track of the checkpoint. else: manually delete the checkpoints other than the best one in TransformersEstimator.fit(), after...
Hi @jmrichardson, also it would be great if you could add a test for your pull request. That is, adding the two following two lines at the end of test_object...
The current meta_feature function does not support different meta feature methods, need to add another parameter such as "algorithm" to support inputting the meta feature method: meta_feature(task, X_train, y_train) ->...
Add auto detect multimodal: we should detect multi modal, if the columns contain text field. To do that, please add an @property function text_columns in DataTransformer, set it to a...
@mossadhelali thanks for your question. Indeed, you should be using the NLP module with the task = 'seq-classification' option. Here is the sample code for using FLAML to classify the...
change _is_nlp_Task to is nlp estimator (distiller or transformers)
Add one concrete example of NLP of selecting parallel vs sequential.
Hi, can you share the code so we could reproduce the issue?
I tried with pydoc-markdown==4.5.0 and saw the same error message. @sonichi
> Was looking at the failed build and noticed that there were some import issues. > > * Python 3.6 doesn't include data classes, but there is a backport that...