import scipy

[brccabral]

I had to explicitly import scipy to run model.fit()

[jasmcaus]

What do you mean?

[jasmcaus]

Without importing scipy, could you please paste the full traceback you get?

[brccabral]

I've just run it again but didn't get any error. But in the previous run I didn't have scipy installed, I guess I had to import it to make it available, but the second time it was already installed and there was no error. Thanks

[jasmcaus]

We don't use Scipy in the tutorial. Could you do this though?

1. Uninstall scipy from your machine
2. Run the script where you got an error before
3. If you still have the error, paste the traceback here

[brccabral]

I created a clean virtual environment, installed canaro (which installs caer, opencv, tensorflow, numpy, and others), installed matplotlib

python -m venv .venvclean
source .venvclean/bin/activate
pip install canaro
pip install matplotlib

Executed the code. When "model.fit()", this is the trace

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
/tmp/ipykernel_63315/3468149526.py in <module>
----> 1 training = model.fit(train_gen,
      2                     steps_per_epoch=len(x_train)//BATCH_SIZE,
      3                     epochs=EPOCHS,
      4                     validation_data=(x_val, y_val),
      5                     validation_steps=len(y_val)//BATCH_SIZE,

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras/engine/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)
   1132          training_utils.RespectCompiledTrainableState(self):
   1133       # Creates a `tf.data.Dataset` and handles batch and epoch iteration.
-> 1134       data_handler = data_adapter.get_data_handler(
   1135           x=x,
   1136           y=y,

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras/engine/data_adapter.py in get_data_handler(*args, **kwargs)
   1381   if getattr(kwargs["model"], "_cluster_coordinator", None):
   1382     return _ClusterCoordinatorDataHandler(*args, **kwargs)
-> 1383   return DataHandler(*args, **kwargs)
   1384 
   1385 

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras/engine/data_adapter.py in __init__(self, x, y, sample_weight, batch_size, steps_per_epoch, initial_epoch, epochs, shuffle, class_weight, max_queue_size, workers, use_multiprocessing, model, steps_per_execution, distribute)
   1136 
   1137     adapter_cls = select_data_adapter(x, y)
-> 1138     self._adapter = adapter_cls(
   1139         x,
   1140         y,

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras/engine/data_adapter.py in __init__(self, x, y, sample_weights, shuffle, workers, use_multiprocessing, max_queue_size, model, **kwargs)
    915     self._keras_sequence = x
    916     self._enqueuer = None
--> 917     super(KerasSequenceAdapter, self).__init__(
    918         x,
    919         shuffle=False,  # Shuffle is handed in the _make_callable override.

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras/engine/data_adapter.py in __init__(self, x, y, sample_weights, workers, use_multiprocessing, max_queue_size, model, **kwargs)
    792     # Since we have to know the dtype of the python generator when we build the
    793     # dataset, we have to look at a batch to infer the structure.
--> 794     peek, x = self._peek_and_restore(x)
    795     peek = self._standardize_batch(peek)
    796     peek = _process_tensorlike(peek)

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras/engine/data_adapter.py in _peek_and_restore(x)
    926   @staticmethod
    927   def _peek_and_restore(x):
--> 928     return x[0], x
    929 
    930   def _handle_multiprocessing(self, x, workers, use_multiprocessing,

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras_preprocessing/image/iterator.py in __getitem__(self, idx)
     63         index_array = self.index_array[self.batch_size * idx:
     64                                        self.batch_size * (idx + 1)]
---> 65         return self._get_batches_of_transformed_samples(index_array)
     66 
     67     def __len__(self):

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras_preprocessing/image/numpy_array_iterator.py in _get_batches_of_transformed_samples(self, index_array)
    159             x = self.x[j]
    160             params = self.image_data_generator.get_random_transform(x.shape)
--> 161             x = self.image_data_generator.apply_transform(
    162                 x.astype(self.dtype), params)
    163             x = self.image_data_generator.standardize(x)

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras_preprocessing/image/image_data_generator.py in apply_transform(self, x, transform_parameters)
    861         img_channel_axis = self.channel_axis - 1
    862 
--> 863         x = apply_affine_transform(x, transform_parameters.get('theta', 0),
    864                                    transform_parameters.get('tx', 0),
    865                                    transform_parameters.get('ty', 0),

~/PythonProjects/ImageAndVideo/.venvclean/lib/python3.8/site-packages/keras_preprocessing/image/affine_transformations.py in apply_affine_transform(x, theta, tx, ty, shear, zx, zy, row_axis, col_axis, channel_axis, fill_mode, cval, order)
    279     """
    280     if scipy is None:
--> 281         raise ImportError('Image transformations require SciPy. '
    282                           'Install SciPy.')
    283     transform_matrix = None

ImportError: Image transformations require SciPy. Install SciPy.

[jasmcaus]

Seems to be an issue here: apple/tensorflow_macos#87

If you install (but not import) Scipy in your Virtual Environment, does this error go away?

[brccabral]

Yes, it does go away.

I've found this, I commented but it is closed, tensorflow/tensorflow#51736

[dineshpiyasamara]

I have reinstall scipy and reopen jupyter notebook. It is worked. Thank you