Question

我正在使用以下模型进行回归分析；输入大小为2，输出大小为28。

from keras.models import Sequential
from keras.layers import Dense
from keras.optimizers import SGD

model = Sequential()
model.add(Dense(16, input_dim=2, activation='relu'))
model.add(Dense(16, activation='relu'))
model.add(Dense(28, activation='linear'))
sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='mean_squared_error',optimizer=sgd)

在训练中一切进展顺利，但是当我保存并重新加载模型时；作为南，我正在减肥。

from keras.models import model_from_json

model_json = model.to_json()
with open('/models/model_ar.json', "w") as json_file:
     json_file.write(model_json)
model.save_weights('/models/model_wt.h5')

json_file = open('/models/model_ar.json', 'r')
loaded_model_json = json_file.read()
json_file.close()
new_model = model_from_json(loaded_model_json)
# load weights into new model
new_model.load_weights('/models/model_wt.h5')

将权重设为“ nan”。将所有权重设为nan的原因是什么

    new_model.get_weights()
[array([[nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan],
        [nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan]], dtype=float32),
 array([nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
        nan, nan, nan], dtype=float32),
 array([[nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan],
        [nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan],
        [nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan],
        [nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan],
        [nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan],
        [nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan]], dtype=float32),
 array([nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
        nan, nan, nan], dtype=float32),
 array([[nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan],
        [nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan],
        [nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
         nan, nan]], dtype=float32),
 array([nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
        nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan, nan,
        nan, nan], dtype=float32)]

Answer 1

尝试

sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True, clipvalue=0.5)

来自https://www.dlology.com/blog/how-to-deal-with-vanishingexploding-gradients-in-keras/
您也可以尝试clipnorm=1.或尝试使用值更小的参数之一。

这限制了权重在梯度下降的每个步骤中可以改变多少。当我遇到同样的问题时，它对我有用，希望对您有所帮助！

Answer 2

就我而言，数据集中有NaN行我没有注意到。因此，请检查您的数据集和值。检查答案here

keras模型的权重是nan

2 个答案: