experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Add missing exception to Model.fit() docstring

Browse Source 
The shuffle argument to the fit() method in the Model class gets ignored
if the input x to the method is a tf.data.Dataset object. But this isn't
documented in the docstring of the method.

Signed-off-by: Suraj Upadhyay <usuraj35@gmail.com>
This commit is contained in:
Suraj Upadhyay 2021-02-04 00:42:59 +05:30
parent 8ddc7459a8
commit c8d44014fb
No known key found for this signature in database
GPG Key ID: 9E6AA57805DA49F4
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
tensorflow/python/keras/engine/training.py
View File

@ -949,7 +949,8 @@ class Model(base_layer.Layer, version_utils.ModelVersionSelector):
are supported in `x`, eg, dict, generator or `keras.utils.Sequence`.
shuffle: Boolean (whether to shuffle the training data
before each epoch) or str (for 'batch'). This argument is ignored
when `x` is a generator. 'batch' is a special option for dealing
when `x` is a generator or an object of tf.data.Dataset.
'batch' is a special option for dealing
with the limitations of HDF5 data; it shuffles in batch-sized
chunks. Has no effect when `steps_per_epoch` is not `None`.
class_weight: Optional dictionary mapping class indices (integers)