Release 3.3.0

New release

Finally!

By Narcon Nicolas on 2022-07-20T11:20:46 (imported from GitLab)

Je pige pas, ça coince au niveau du crop (il râle parce qu'on crop de 96 de chaque coté, alors que le patch fait 128 de côté).

Epoch 1/100
/opt/otbtf/lib/python3/site-packages/keras/engine/functional.py:559: UserWarning: Input dict contained keys ['labels'] which did not match any model input. They will be ignored by the model.
  inputs = self._flatten_to_reference_inputs(inputs)
Traceback (most recent call last):
  File "train_from_patches-images.py", line 86, in <module>
    fcnn_model.train(params, ds_train, ds_valid, ds_test, ds.output_shapes)
  File "/src/otbtf/otbtf/examples/tensorflow_v2x/fcnn/fcnn_model.py", line 97, in train
    model.fit(ds_train, epochs=params.nb_epochs, validation_data=ds_valid)
  File "/opt/otbtf/lib/python3/site-packages/keras/utils/traceback_utils.py", line 67, in error_handler
    raise e.with_traceback(filtered_tb) from None
  File "/opt/otbtf/lib/python3/site-packages/tensorflow/python/framework/func_graph.py", line 1147, in autograph_handler
    raise e.ag_error_metadata.to_exception(e)
ValueError: in user code:

    File "/opt/otbtf/lib/python3/site-packages/keras/engine/training.py", line 1021, in train_function  *
        return step_function(self, iterator)
    File "/opt/otbtf/lib/python3/site-packages/keras/engine/training.py", line 1010, in step_function  **
        outputs = model.distribute_strategy.run(run_step, args=(data,))
    File "/usr/local/lib/python3.8/dist-packages/six.py", line 719, in reraise
        raise value
    File "/opt/otbtf/lib/python3/site-packages/keras/engine/training.py", line 1000, in run_step  **
        outputs = model.train_step(data)
    File "/opt/otbtf/lib/python3/site-packages/keras/engine/training.py", line 859, in train_step
        y_pred = self(x, training=True)
    File "/opt/otbtf/lib/python3/site-packages/keras/utils/traceback_utils.py", line 67, in error_handler
        raise e.with_traceback(filtered_tb) from None
    File "/opt/otbtf/lib/python3/site-packages/keras/layers/convolutional.py", line 3556, in call
        raise ValueError('Argument `cropping` must be '

    ValueError: Exception encountered when calling layer "tconv4_crop96" (type Cropping2D).
    
    Argument `cropping` must be greater than the input shape. Received: inputs.shape=(8, 128, 128, 6), and cropping=((96, 96), (96, 96))
    
    Call arguments received:
      • inputs=tf.Tensor(shape=(8, 128, 128, 6), dtype=float32)

Pourtant j'ai regardé dans decloud, on faisait exactement la même chose...

By Cresson Remi on 2022-07-20T11:27:25 (imported from GitLab)

Dans decloud les patches faisaient 256 de côté

By Narcon Nicolas on 2022-07-20T11:29:36 (imported from GitLab)

oui et on croppait jusque 256 (https://github.com/CNES/decloud/blob/master/decloud/preprocessing/constants.py#L42)

By Cresson Remi on 2022-07-20T11:32:21 (imported from GitLab)

Ah oui tiens, peut être le changement de version TF ?

By Narcon Nicolas on 2022-07-20T12:13:52 (imported from GitLab)

je ne pense pas

By Cresson Remi on 2022-07-20T12:56:22 (imported from GitLab)

enabled an automatic merge when the pipeline for 01d13b0b succeeds

By Cresson Remi on 2022-07-28T07:37:34 (imported from GitLab)

merged

mentioned in commit b124a937

By Cresson Remi on 2022-07-28T07:51:10 (imported from GitLab)