LSTM 模型在 model.predict() 中得到错误。这些是数据的形状-shapes。这是一个有 7 个不同类别的分类模型。处理音频数据。当我使用自己的数据测试模型时出现错误。
model1 = Sequential()
model1.add(LSTM(units=128, dropout=0.05, recurrent_dropout=0.35, return_sequences=True, input_shape=(180,12004,)))
model1.add(LSTM(units=32, dropout=0.05, recurrent_dropout=0.35, return_sequences=False))
model1.add(Dense(units=7, activation="softmax"))
print("Compiling ...")
opt = Adam()
model1.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"])
model1.summary()
print("Training ...")
batch_size = 35 # num of training examples per minibatch
num_epochs = 100
model.fit(
X_train,
YY_train,
batch_size=batch_size,
epochs=num_epochs,
)
res1=model1.predict(testaudio.reshape(1, -1))
