-
Notifications
You must be signed in to change notification settings - Fork 292
Open
Description
Hi.
When runnin notebook in colab on the line
dcn_result = run_models(use_cross_layer=True,
deep_layer_sizes=[192, 192])
the ValueError
got raised
ValueError Traceback (most recent call last)
[/tmp/ipython-input-21-1445969553.py](https://localhost:8080/#) in <cell line: 0>()
----> 1 dcn_result = run_models(use_cross_layer=True,
2 deep_layer_sizes=[192, 192])
3 frames
[/tmp/ipython-input-19-2513347440.py](https://localhost:8080/#) in run_models(use_cross_layer, deep_layer_sizes, projection_dim, num_runs)
4
5 for i in range(num_runs):
----> 6 model = DCN(use_cross_layer=use_cross_layer,
7 deep_layer_sizes=deep_layer_sizes,
8 projection_dim=projection_dim)
[/tmp/ipython-input-17-2634010612.py](https://localhost:8080/#) in __init__(self, use_cross_layer, deep_layer_sizes, projection_dim)
27 vocabulary = vocabularies[feature_name]
28 self._embeddings[feature_name] = tf.keras.Sequential(
---> 29 [tf.keras.layers.IntegerLookup(
30 vocabulary=vocabulary, mask_value=None),
31 tf.keras.layers.Embedding(len(vocabulary) + 1,
[/usr/local/lib/python3.11/dist-packages/keras/src/layers/preprocessing/integer_lookup.py](https://localhost:8080/#) in __init__(self, max_tokens, num_oov_indices, mask_token, oov_token, vocabulary, vocabulary_dtype, idf_weights, invert, output_mode, sparse, pad_to_max_tokens, name, **kwargs)
337 f"vocabulary_dtype={vocabulary_dtype}"
338 )
--> 339 super().__init__(
340 max_tokens=max_tokens,
341 num_oov_indices=num_oov_indices,
[/usr/local/lib/python3.11/dist-packages/keras/src/layers/preprocessing/index_lookup.py](https://localhost:8080/#) in __init__(self, max_tokens, num_oov_indices, mask_token, oov_token, vocabulary_dtype, vocabulary, idf_weights, invert, output_mode, sparse, pad_to_max_tokens, name, **kwargs)
191 kwargs.pop("dtype", None)
192 if kwargs:
--> 193 raise ValueError(f"Unrecognized keyword argument(s): {kwargs}")
194
195 if invert:
ValueError: Unrecognized keyword argument(s): {'mask_value': None}
Metadata
Metadata
Assignees
Labels
No labels