Bug: KeyError: 'text_block_1/max_tokens'
gitzev opened this issue · comments
Bug Description
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/base_tuner.py", line 270, in _try_run_and_update_trial
self._run_and_update_trial(trial, *fit_args, **fit_kwargs)
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/base_tuner.py", line 235, in _run_and_update_trial
results = self.run_trial(trial, *fit_args, **fit_kwargs)
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/tuner.py", line 287, in run_trial
obj_value = self._build_and_fit_model(trial, *args, **copied_kwargs)
File "/usr/local/lib/python3.10/dist-packages/autokeras/engine/tuner.py", line 91, in _build_and_fit_model
model = self._try_build(trial.hyperparameters)
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/tuner.py", line 155, in _try_build
model = self._build_hypermodel(hp)
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/tuner.py", line 146, in _build_hypermodel
model = self.hypermodel.build(hp)
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/hypermodel.py", line 115, in _build_wrapper
return self._build(hp, *args, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/autokeras/graph.py", line 250, in build
outputs = block.build(hp, inputs=temp_inputs)
File "/usr/local/lib/python3.10/dist-packages/autokeras/engine/block.py", line 38, in _build_wrapper
return super()._build_wrapper(hp, *args, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/hypermodel.py", line 115, in _build_wrapper
return self._build(hp, *args, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/autokeras/blocks/wrapper.py", line 159, in build
output_node = self._build_block(hp, output_node, block_type)
File "/usr/local/lib/python3.10/dist-packages/autokeras/blocks/wrapper.py", line 165, in _build_block
max_tokens = self.max_tokens or hp.Choice(
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/hyperparameters/hyperparameters.py", line 295, in Choice
return self._retrieve(hp)
File "/usr/local/lib/python3.10/dist-packages/keras_tuner/engine/hyperparameters/hyperparameters.py", line 203, in _retrieve
return self.values[hp.name]
KeyError: 'text_block_1/max_tokens'
Bug Reproduction
When fit the model on random trial i have error. It can be trial=7 or trial =63
model2 = ak.AutoModel(
inputs=[ak.TextInput(), ak.StructuredDataInput()],
outputs=[
ak.RegressionHead(metrics=["mae"])
],
overwrite=False,
max_trials=100,
objective ='val_mae'
)
model2.fit(
[texts_all_np_train, xTrain_train],
y_train,
epochs=8,
validation_split=0.2
)
Data used by the code:
texts_all_np_train= df['Stage_Description'].values #str
Expected Behavior
Setup Details
Include the details about the versions of:
- OS type and version: google colab
- Python: 3.10
- autokeras: <1.1.0>
- keras-tuner: 1.3.5
- scikit-learn: 1.2.2
- numpy: 1.22.4
- pandas: 1.5.3
- tensorflow: 2.12.0
Additional context
I'm getting this issue too, were you able to fix it?
Me, too 😕
I have encountered a similar error.
Search: Running Trial #5
Value |Best Value So Far |Hyperparameter
vanilla |vanilla |text_block_1/block_type
none |none |text_block_1/embedding_1/pretraining
64 |64 |text_block_1/embedding_1/embedding_dim
0.25 |0.25 |text_block_1/embedding_1/dropout
5 |5 |text_block_1/conv_block_1/kernel_size
False |False |text_block_1/conv_block_1/separable
False |False |text_block_1/conv_block_1/max_pooling
1 |1 |text_block_1/conv_block_1/num_blocks
1 |1 |text_block_1/conv_block_1/num_layers
256 |256 |text_block_1/conv_block_1/filters_0_0
128 |128 |text_block_1/conv_block_1/filters_0_1
0 |0 |text_block_1/conv_block_1/dropout
64 |64 |text_block_1/conv_block_1/filters_1_0
32 |32 |text_block_1/conv_block_1/filters_1_1
0 |0 |classification_head_1/dropout
sgd |adam |optimizer
0.001 |0.001 |learning_rate
Traceback (most recent call last):
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\base_tuner.py", line 273, in _try_run_and_update_trial
self._run_and_update_trial(trial, *fit_args, **fit_kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\base_tuner.py", line 238, in _run_and_update_trial
results = self.run_trial(trial, *fit_args, **fit_kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\tuner.py", line 314, in run_trial
obj_value = self._build_and_fit_model(trial, *args, **copied_kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\engine\tuner.py", line 91, in _build_and_fit_model
model = self._try_build(trial.hyperparameters)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\tuner.py", line 164, in _try_build
model = self._build_hypermodel(hp)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\tuner.py", line 155, in _build_hypermodel
model = self.hypermodel.build(hp)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\hypermodel.py", line 120, in _build_wrapper
return self._build(hp, *args, **kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\graph.py", line 250, in build
outputs = block.build(hp, inputs=temp_inputs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\engine\block.py", line 38, in _build_wrapper
return super()._build_wrapper(hp, *args, **kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\hypermodel.py", line 120, in _build_wrapper
return self._build(hp, *args, **kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\blocks\wrapper.py", line 159, in build
output_node = self._build_block(hp, output_node, block_type)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\blocks\wrapper.py", line 165, in _build_block
max_tokens = self.max_tokens or hp.Choice(
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\hyperparameters\hyperparameters.py", line 300, in Choice
return self._retrieve(hp)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\hyperparameters\hyperparameters.py", line 208, in _retrieve
return self.values[hp.name]
KeyError: 'text_block_1/max_tokens'
Traceback (most recent call last):
File "", line 1, in
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\tasks\text.py", line 160, in fit
history = super().fit(
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\auto_model.py", line 292, in fit
history = self.tuner.search(
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\engine\tuner.py", line 193, in search
super().search(
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\base_tuner.py", line 234, in search
self.on_trial_end(trial)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\base_tuner.py", line 338, in on_trial_end
self.oracle.end_trial(trial)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\oracle.py", line 108, in wrapped_func
ret_val = func(*args, **kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\oracle.py", line 586, in end_trial
self._check_consecutive_failures()
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\oracle.py", line 543, in _check_consecutive_failures
raise RuntimeError(
RuntimeError: Number of consecutive failures exceeded the limit of 3.
Traceback (most recent call last):
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\base_tuner.py", line 273, in _try_run_and_update_trial
self._run_and_update_trial(trial, *fit_args, **fit_kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\base_tuner.py", line 238, in _run_and_update_trial
results = self.run_trial(trial, *fit_args, **fit_kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\tuner.py", line 314, in run_trial
obj_value = self._build_and_fit_model(trial, *args, **copied_kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\engine\tuner.py", line 91, in _build_and_fit_model
model = self._try_build(trial.hyperparameters)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\tuner.py", line 164, in _try_build
model = self._build_hypermodel(hp)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\tuner.py", line 155, in _build_hypermodel
model = self.hypermodel.build(hp)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\hypermodel.py", line 120, in _build_wrapper
return self._build(hp, *args, **kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\graph.py", line 250, in build
outputs = block.build(hp, inputs=temp_inputs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\engine\block.py", line 38, in _build_wrapper
return super()._build_wrapper(hp, *args, **kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\hypermodel.py", line 120, in _build_wrapper
return self._build(hp, *args, **kwargs)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\blocks\wrapper.py", line 159, in build
output_node = self._build_block(hp, output_node, block_type)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\autokeras\blocks\wrapper.py", line 165, in _build_block
max_tokens = self.max_tokens or hp.Choice(
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\hyperparameters\hyperparameters.py", line 300, in Choice
return self._retrieve(hp)
File "C:\Users\yvorlv\miniconda3\envs\tensorflow\lib\site-packages\keras_tuner\engine\hyperparameters\hyperparameters.py", line 208, in _retrieve
return self.values[hp.name]
KeyError: 'text_block_1/max_tokens'
I am facing this issue too on auto-keras == 1.1.0.
This seems to happen when the tuner is set to either of 'bayesian', 'hyperband' or 'random'.
Contribution is welcome on this issue!