Question

我正在尝试训练一个模型来识别图像类别。图片为50x50像素宽。尝试运行代码时，出现错误

2019-04-15 08:55:48.737185: W tensorflow/core/framework/op_kernel.cc:1378] OP_REQUIRES failed at
strided_slice_op.cc:88 : Invalid argument: Attr shrink_axis_mask has value 4294967295 out of rang
e for an int32
Traceback (most recent call last):
  File "Final.py", line 108, in <module>
    model.fit(as_np_arr, all_image_labels, epochs=5, shuffle=False)
  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/keras/engine/training.py", line
880, in fit
    validation_steps=validation_steps)
  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/keras/engine/training_arrays.py"
, line 310, in model_iteration
    ins_batch = slice_arrays(ins[:-1], batch_ids) + [ins[-1]]
  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/keras/utils/generic_utils.py", l
ine 526, in slice_arrays
    return [None if x is None else x[start] for x in arrays]
  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/keras/utils/generic_utils.py", l
ine 526, in <listcomp>
    return [None if x is None else x[start] for x in arrays]
  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/ops/array_ops.py", line 654, in
_slice_helper
    name=name)
  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/ops/array_ops.py", line 820, in
strided_slice
    shrink_axis_mask=shrink_axis_mask)
  File "/usr/local/lib/python3.7/site-packages/tensorflow/python/ops/gen_array_ops.py", line 9334
, in strided_slice
    _six.raise_from(_core._status_to_exception(e.code, message), None)
  File "<string>", line 3, in raise_from
tensorflow.python.framework.errors_impl.InvalidArgumentError: Attr shrink_axis_mask has value 4294967295 out of range for an int32 [Op:StridedSlice] name: strided_slice/

这是正在使用的代码。 tf_raw是表示为numpy数组的图像列表，而all_image_labels是每个图像的类的python列表。

print("Creating model")
model = keras.Sequential({
    # keras.layers.Flatten(input_shape=(3670, 50, 50, 3)),
    keras.layers.Dense(2500, activation = tf.nn.relu),
    keras.layers.Dense(20, activation = tf.nn.softmax)
})
print("Compiled")
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics = ['accuracy'])
print("Turning into array")
as_np_arr = tf.convert_to_tensor(tf_raws)
print("Converted images")
all_image_labels = tf.convert_to_tensor(all_image_labels)
print(str(len(as_np_arr)) + " -> " + str(len(all_image_labels)))
print("TRAINING THE MODEL")
model.fit(as_np_arr, all_image_labels, epochs=5, shuffle=False)
print("Model is trained!")

Answer 1

我遇到了同样的问题，我尝试将numpy数组（而不是张量）馈送到fit()函数。基本上将代码的第二行替换为：

model.fit(as_np_arr.numpy(), all_image_labels.numpy(), epochs=5, shuffle=False)

我希望这对您来说是一个快速的解决方法。但是，我希望有一个经验丰富的人提供适当的解释。

在图像上训练Keras模型时，出现错误“无效参数：Attrrinke_axis_mask的值4294967295超出int32的范围”

1 个答案: