Keras model.predict giving same values

问题

I am using Keras model.predict to get label for new sentences not present in the dataset. But the prediction is always giving the same value irrespective of the sentence.

Here is my prediction code

from sklearn.preprocessing import LabelEncoder

maxlen = 300
### PREDICT NEW UNSEEN DATA ###
tokenizer = Tokenizer()
label_enc = LabelEncoder()
label_enc.fit(tar_list)
X_test = ['asdsadav dawd','this is boring', 'wow i like this you did a great job', 'ima cry tht was mean','1 nov 1968 george harrison became the first beatle to release a solo album in the u k with the soundtrack to ']

X_test = tokenizer.texts_to_sequences(X_test)
X_test = sequence.pad_sequences(X_test, maxlen=maxlen)

print(X_test)

a = (model.predict(X_test)>0.5).astype(int).ravel()
print(a)

reverse_pred = label_enc.inverse_transform(a.ravel())
print(reverse_pred)

print(model.predict(X_test))

And here is the output

 [[0 0 0 ... 0 0 0]
 [0 0 0 ... 0 0 0]
 [0 0 0 ... 0 0 0]
 [0 0 0 ... 0 0 0]
 [0 0 0 ... 0 0 0]]
[1 0 1 0 1 0 1 0 1 0]
[1 0 1 0 1 0 1 0 1 0]
[[0.988675   0.01132498]
 [0.988675   0.01132498]
 [0.988675   0.01132498]
 [0.988675   0.01132498]
 [0.988675   0.01132498]]

As we can see, the probability output is same for every sentence.

My code for training model is

model = Sequential()
model.add(Embedding(max_words, 300, input_length=max_len))
model.add(BatchNormalization())
model.add(Activation('tanh'))
model.add(SpatialDropout1D(0.5))
model.add(Conv1D(16, kernel_size=3, activation='relu'))
model.add(Bidirectional(LSTM(16)))
model.add(BatchNormalization())
model.add(Activation('tanh'))
model.add(Dropout(0.5))
model.add(Dense(2, activation='softmax'))
model.summary()
model.compile(loss='sparse_categorical_crossentropy', metrics=['accuracy'], optimizer = 'adam')

This is the tokenizer fitting for X_train

max_words = 3000
max_len = 300
tok = Tokenizer(num_words = max_words)
tok.fit_on_texts(X_train)
sequences = tok.texts_to_sequences(X_train)
sequences_matrix = sequence.pad_sequences(sequences, maxlen = max_len)
print(sequences_matrix)
Y_train = np.array(Y_train)
Y_test = np.array(Y_test)

Output is

[[  0   0   0 ...  11  28  33]
 [  0   0   0 ...   2 125  63]
 [  0   0   0 ...   9 184  91]
 ...
 [  0   0   0 ... 105  22  85]
 [  0   0   0 ...  22  42 512]
 [  0   0   0 ...   9  28 406]]

来源：https://stackoverflow.com/questions/63227885/keras-model-predict-giving-same-values