machine-learning - ValueError : Error when checking input: expected lstm_1_input to have shape (None, 296, 2048) 但得到形状为 (296, 2048, 1) 的数组

Question

我遇到了标题中的错误。我有数千个视频，每个视频有 37 帧。我用 CNN 模型提取了每一帧的特征并保存了它们。 我有一个堆叠 LSTM 模型:

batch_size = 8
features_length = 2048
seq_length = 37*batch_size
in_shape = (seq_length, features_length)
lstm_model = Sequential()
lstm_model.add(LSTM(2048, return_sequences=True, input_shape = in_shape, dropout=0.5))
lstm_model.add(Flatten())
lstm_model.add(Dense(512, activation='relu'))
lstm_model.add(Dropout(0.5))
lstm_model.add(Dense(number_of_classes, activation='softmax'))
optimizer = Adam(lr=1e-6)
lstm_model.compile(loss='categorical_crossentropy', optimizer=optimizer, metrics = metrics)
lstm_model.fit_generator(generator = generator, steps_per_epoch = train_steps_per_epoch, epochs = nb_epoch, verbose = 1, callbacks=[checkpointer, tb, early_stopper, csv_logger], validation_data=val_generator, validation_steps = val_steps_per_epoch)

我有一台发电机；数据包括所有训练视频。

def generator(data):

    while 1:
        X, y = [], []
        for _ in range(batch_size):
            sequence = None
            sample = random.choice(data)
            folder_content, folder_name, class_name, video_features_loc = get_video_features(sample)
            for f in folder_content:
                image_feature_location = video_features_loc + f
                feat = get_extracted_feature(image_feature_location)

                X.append(feat)
                y.append(get_one_class_rep(class_name))         
        yield np.array(X), np.array(y)

生成器数据中 X 的形状为 = (296, 2048, 1)

生成器数据中 y 的形状为 = (296, 27)

此代码引发错误。我知道有几个类似的问题。我尝试了那里的建议，但没有运气。例如，建议之一是 reshape 数组；

X = np.reshape(X, (X.shape[2], X.shape[0], X.shape[1]))

我如何将输入输入到 LSTM？

提前致谢

Answer 1

错误消息告诉您所需的一切。

X 的形状应为 (样本数，296, 2048) - 根据 X 的形状，您似乎只有一个样本。

<小时/>

但是如果你有 37 帧，你绝对应该改变你的模型以接受:(Batch size, 37, 2048) - 这里，batch size 似乎是 8。

seq_length=37