Cast string to float is not supported / Failed to convert a NumPy array to a Tensor (Unsupported object type float)

[y-sheng]

Hi,

I am trying to train an autokeras model using a mixed-type of data including float, int and object (category). After running for several successful trails, it crashed with an error message saying either 'Cast string to float is not supported' or 'Failed to convert a NumPy array to a Tensor (Unsupported object type float)'.

I have tried using StructuredDataRegressor and AutoModel, both gave me this error.

I am using autokeras version 1.0.20 on windows in jupyter notebook, using the following lines:
au_try = ak.AutoModel(inputs = ak.StructuredDataInput(),
outputs=ak.RegressionHead(loss='mse',
metrics=[tf.keras.metrics.MeanAbsolutePercentageError()]),
tuner='bayesian',
overwrite=True,
max_trials=100)

The data I am using for training is a csv dataframe containing the following columns and their dtypes:
A category
B float64
C int64
D category
E category
F float64

Could you advise what I should do with this issue, please? Many thanks

[Eric Hulburd]

I received a similar error when I had tensorflow 2.11.0 installed. The offending line was:

tf.cast(self.weight_decay, variable.dtype)

where self.weight_decay = 'AdamWeightDecay' and variable.dtype = tf.float32. I don't have a re-producing script to run but, I hit this when running StructuredDataClassifier.fit.

Rolling back to tensorflow 2.10.1 resolved this issue for me.

Full stacktrace

``` Out[2]: 'AdamWeightDecay' Out[3]: 2022-12-06 11:32:24.139522: W tensorflow/core/framework/op_kernel.cc:1807] OP_REQUIRES failed at cast_op.cc:121 : UNIMPLEMENTED: Cast string to float is not supported 2022-12-06 11:32:24.141576: W tensorflow/core/framework/op_kernel.cc:1807] OP_REQUIRES failed at cast_op.cc:121 : UNIMPLEMENTED: Cast string to float is not supported Traceback (most recent call last): File "~/repo/.venv/lib/python3.8/site-packages/autokeras/utils/utils.py", line 101, in run_with_adaptive_batch_size history = func(x=x, validation_data=validation_data, **fit_kwargs) File "~/repo/.venv/lib/python3.8/site-packages/autokeras/utils/utils.py", line 89, in batch_size, lambda **kwargs: model.fit(**kwargs), **fit_kwargs File "~/repo/.venv/lib/python3.8/site-packages/keras/utils/traceback_utils.py", line 70, in error_handler raise e.with_traceback(filtered_tb) from None File "~/repo/.venv/lib/python3.8/site-packages/tensorflow/python/eager/execute.py", line 52, in quick_execute tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name, tensorflow.python.framework.errors_impl.UnimplementedError: Graph execution error:

Detected at node 'Cast_5' defined at (most recent call last):
File "/Applications/PyCharm CE.app/Contents/plugins/python-ce/helpers/pydev/pydevd.py", line 2141, in
main()
File "/Applications/PyCharm CE.app/Contents/plugins/python-ce/helpers/pydev/pydevd.py", line 2132, in main
globals = debugger.run(setup['file'], None, None, is_module)
File "/Applications/PyCharm CE.app/Contents/plugins/python-ce/helpers/pydev/pydevd.py", line 1441, in run
return self._exec(is_module, entry_point_fn, module_name, file, globals, locals)
File "/Applications/PyCharm CE.app/Contents/plugins/python-ce/helpers/pydev/pydevd.py", line 1448, in _exec
pydev_imports.execfile(file, globals, locals) # execute the script
File "/Applications/PyCharm CE.app/Contents/plugins/python-ce/helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "/repo/qcaan/classifier.py", line 54, in
train(Classifier.AUTO_KERAS)
File "/repo/qcaan/classifier.py", line 41, in train
train_results = classifier.train()
File "/repo/qcaan/classifier.py", line 21, in train
return autokeras.train()
File "/repo/qcaan/classifiers/autokeras.py", line 108, in train
ak_classifier.fit(
File "/repo/.venv/lib/python3.8/site-packages/autokeras/tasks/structured_data.py", line 326, in fit
history = super().fit(
File "/repo/.venv/lib/python3.8/site-packages/autokeras/tasks/structured_data.py", line 139, in fit
history = super().fit(
File "/repo/.venv/lib/python3.8/site-packages/autokeras/auto_model.py", line 292, in fit
history = self.tuner.search(
File "/repo/.venv/lib/python3.8/site-packages/autokeras/engine/tuner.py", line 193, in search
super().search(
File "/repo/.venv/lib/python3.8/site-packages/keras_tuner/engine/base_tuner.py", line 183, in search
results = self.run_trial(trial, *fit_args, **fit_kwargs)
File "/repo/.venv/lib/python3.8/site-packages/keras_tuner/engine/tuner.py", line 295, in run_trial
obj_value = self._build_and_fit_model(trial, *args, **copied_kwargs)
File "/repo/.venv/lib/python3.8/site-packages/autokeras/engine/tuner.py", line 101, in _build_and_fit_model
_, history = utils.fit_with_adaptive_batch_size(
File "/repo/.venv/lib/python3.8/site-packages/autokeras/utils/utils.py", line 88, in fit_with_adaptive_batch_size
history = run_with_adaptive_batch_size(
File "/repo/.venv/lib/python3.8/site-packages/autokeras/utils/utils.py", line 101, in run_with_adaptive_batch_size
history = func(x=x, validation_data=validation_data, **fit_kwargs)
File "/repo/.venv/lib/python3.8/site-packages/autokeras/utils/utils.py", line 89, in
batch_size, lambda **kwargs: model.fit(**kwargs), **fit_kwargs
File "/repo/.venv/lib/python3.8/site-packages/keras/utils/traceback_utils.py", line 65, in error_handler
return fn(*args, **kwargs)
File "/repo/.venv/lib/python3.8/site-packages/keras/engine/training.py", line 1650, in fit
tmp_logs = self.train_function(iterator)
File "/repo/.venv/lib/python3.8/site-packages/keras/engine/training.py", line 1249, in train_function
return step_function(self, iterator)
File "/repo/.venv/lib/python3.8/site-packages/keras/engine/training.py", line 1233, in step_function
outputs = model.distribute_strategy.run(run_step, args=(data,))
File "/repo/.venv/lib/python3.8/site-packages/keras/engine/training.py", line 1222, in run_step
outputs = model.train_step(data)
File "/repo/.venv/lib/python3.8/site-packages/keras/engine/training.py", line 1027, in train_step
self.optimizer.minimize(loss, self.trainable_variables, tape=tape)
File "/repo/.venv/lib/python3.8/site-packages/keras/optimizers/optimizer_experimental/optimizer.py", line 527, in minimize
self.apply_gradients(grads_and_vars)
File "/repo/.venv/lib/python3.8/site-packages/autokeras/keras_layers.py", line 360, in apply_gradients
return super(AdamWeightDecay, self).apply_gradients(
File "/repo/.venv/lib/python3.8/site-packages/keras/optimizers/optimizer_experimental/optimizer.py", line 1140, in apply_gradients
return super().apply_gradients(grads_and_vars, name=name)
File "/repo/.venv/lib/python3.8/site-packages/keras/optimizers/optimizer_experimental/optimizer.py", line 632, in apply_gradients
self._apply_weight_decay(trainable_variables)
File "/repo/.venv/lib/python3.8/site-packages/keras/optimizers/optimizer_experimental/optimizer.py", line 1159, in _apply_weight_decay
File "/repo/.venv/lib/python3.8/site-packages/keras/optimizers/optimizer_experimental/optimizer.py", line 1155, in distributed_apply_weight_decay
for variable in variables:
File "~/repo/.venv/lib/python3.8/site-packages/keras/optimizers/optimizer_experimental/optimizer.py", line 1151, in weight_decay_fn
Node: 'Cast_5'
Cast string to float is not supported
[[{{node Cast_5}}]] [Op:__inference_train_function_333096]
python-BaseException

Process finished with exit code 130 (interrupted by signal 2: SIGINT)

</details>

[y-sheng]

I received a similar error when I had tensorflow 2.11.0 installed. The offending line was:

tf.cast(self.weight_decay, variable.dtype)

where self.weight_decay = 'AdamWeightDecay' and variable.dtype = tf.float32. I don't have a re-producing script to run but, I hit this when running StructuredDataClassifier.fit.

Rolling back to tensorflow 2.10.1 resolved this issue for me.

Full stacktrace

Many thanks @erichulburd, using TensorFlow 2.10.1 solved the issue.