index [32768] = [1, 16, 1, 0] не индексируется в форме [10,16,16,512] - PullRequest
Новая
       7

index [32768] = [1, 16, 1, 0] не индексируется в форме [10,16,16,512]

0 голосов
/ 11 декабря 2018

Я делаю сегментацию в керасе.После того, как я ввел свой набор данных, я получаю эту ошибку ниже, я думаю, что это должно быть из model.fit-generator в строке 88 в train.py, но я не знаю, почему я получаю эту ошибку и как я должен ее исправить.

model.fit_generator(train_gen, steps_per_epoch=args.epoch_steps,
        epochs=args.n_epochs, validation_data=val_gen,
        validation_steps=args.val_steps)

ошибка: 

Traceback (most recent call last):
  File "train.py", line 96, in <module>
    main(args)
  File "train.py", line 88, in main
    validation_steps=args.val_steps)
  File "/root/anaconda3/envs/my_env/lib/python3.6/site-packages/keras/legacy/interfaces.py", line 91, in wrapper
    return func(*args, **kwargs)
  File "/root/anaconda3/envs/my_env/lib/python3.6/site-packages/keras/engine/training.py", line 1418, in fit_generator
    initial_epoch=initial_epoch)
  File "/root/anaconda3/envs/my_env/lib/python3.6/site-packages/keras/engine/training_generator.py", line 217, in fit_generator
    class_weight=class_weight)
  File "/root/anaconda3/envs/my_env/lib/python3.6/site-packages/keras/engine/training.py", line 1217, in train_on_batch
    outputs = self.train_function(ins)
  File "/root/anaconda3/envs/my_env/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py", line 2715, in __call__
    return self._call(inputs)
  File "/root/anaconda3/envs/my_env/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py", line 2675, in _call
    fetched = self._callable_fn(*array_vals)
  File "/root/anaconda3/envs/my_env/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1439, in __call__
    run_metadata_ptr)
  File "/root/anaconda3/envs/my_env/lib/python3.6/site-packages/tensorflow/python/framework/errors_impl.py", line 528, in __exit__
    c_api.TF_GetCode(self.status.status))
tensorflow.python.framework.errors_impl.InvalidArgumentError: indices[32768] = [1, 16, 1, 0] does not index into shape [10,16,16,512]
     [[{{node max_unpooling2d_1/max_unpooling2d_1/ScatterNd}} = ScatterNd[T=DT_FLOAT, Tindices=DT_INT32, _class=["loc:@train...d/GatherNd"], _device="/job:localhost/replica:0/task:0/device:CPU:0"](max_unpooling2d_1/max_unpooling2d_1/transpose, max_unpooling2d_1/max_unpooling2d_1/Reshape_2, max_unpooling2d_1/max_unpooling2d_1/ScatterNd/shape)]]
...