相关疑难解决方法(0)

我正在使用Keras预测图像类.它适用于Google Cloud ML(GCML),但为了提高效率,需要将其更改为传递base64字符串而不是json数组. 相关文档

我可以轻松运行python代码将base64字符串解码为json数组,但是当使用GCML时,我没有机会运行预处理步骤(除非在Keras中使用Lambda层,但我认为不是正确的方法).

另一个答案建议添加tf.placeholder类型tf.string,这是有道理的,但如何将其纳入Keras模型？

以下是培训模型和保存GCML导出模型的完整代码...

import os
import numpy as np
import tensorflow as tf
import keras
from keras import backend as K
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten
from keras.layers import Conv2D, MaxPooling2D
from keras.preprocessing import image
from tensorflow.python.platform import gfile

IMAGE_HEIGHT = 138
IMAGE_WIDTH = 106
NUM_CLASSES = 329

def preprocess(filename):
    # decode the image file starting from the filename
    # end up with pixel values that …

我正在尝试将keras模型转换为用于Google Cloud的ml-engine预测的模型。我有一个预先训练的分类器，将numpy数组作为输入。我发送给model.predict的正常工作数据称为input_data。

我将其转换为base 64，并json使用以下几行将其转储到文件中：

data = {}
data['image_bytes'] = [{'b64':base64.b64encode(input_data.tostring())}]

with open('weights/keras/example.json', 'w') as outfile:
    json.dump(data, outfile)

现在，我尝试从现有模型创建TF模型：

from keras.models import model_from_json
import tensorflow as tf
from keras import backend as K
from tensorflow.python.saved_model import builder as saved_model_builder
from tensorflow.python.saved_model import utils
from tensorflow.python.saved_model import tag_constants, signature_constants
from tensorflow.python.saved_model.signature_def_utils_impl import build_signature_def, predict_signature_def
init = tf.global_variables_initializer()
with tf.Session() as sess:
    K.set_session(sess)
    sess.run(init)
    print("Keras model & weights loading...")
    K.set_learning_phase(0)
    with open(json_file_path, 'r') as file_handle:
        model = …