search

DeepRec-AI / HybridBackend

A high-performance framework for training wide-and-deep recommender systems on heterogeneous cluster
Apache License 2.0
156 stars 30 forks source link

hb.keras.model evaluate error #118

Closed karterotte closed 1 year ago

karterotte commented 1 year ago

Current behavior

hb.keras.model evaluate report error like:

Traceback (most recent call last): File "models/recommend/MultiTowerDnn.py", line 100, in base.run(MultiTowerDnn) File "/var/workspace/models/recommend/base.py", line 331, in run scores = m.evaluate(valid_ds, verbose=0, steps=None) File "/usr/local/lib/python3.6/dist-packages/hybridbackend/tensorflow/keras/model.py", line 609, in evaluate super().evaluate(*args, **kwargs) File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/keras/engine/training.py", line 832, in evaluate use_multiprocessing=use_multiprocessing) File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/keras/engine/training_arrays.py", line 704, in evaluate callbacks=callbacks) File "/usr/local/lib/python3.6/dist-packages/hybridbackend/tensorflow/keras/model.py", line 345, in wrapped_model_iteration self._save_best_mode): File "/usr/local/lib/python3.6/dist-packages/hybridbackend/tensorflow/keras/model.py", line 147, in init if 'acc' in self._monitor or self._monitor.startswith('fmeasure'): TypeError: argument of type 'NoneType' is not iterable

Willing to contribute

Yes