Tensorflow Serving 在预训练的 Keras ResNet50 模型上返回始终相同的预测

Tensorflow Serving on pretrained Keras ResNet50 model returning always same predictions

我正在使用以下代码将预训练的 ResNet50 keras 模型导出到 tensorflow,用于 tensorflow-serving:

import tensorflow as tf
sess = tf.Session()
from keras import backend as K
K.set_session(sess)
K.set_learning_phase(0)

# Modelo resnet con pesos entrenados en imagenet
from keras.applications.resnet50 import ResNet50
model = ResNet50(weights='imagenet')

# exportar en tensorflow
import os
version_number = max([ int(x) for x in os.listdir('./resnet-classifier') ]) + 1
export_path = './resnet-classifier/{}'.format(version_number)
with tf.keras.backend.get_session() as sess:
    init_op = tf.global_variables_initializer()
    sess.run(init_op)
    tf.saved_model.simple_save(sess, export_path,
            inputs=dict(input_image=model.input),
            outputs={t.name:t for t in model.outputs}
    )

我尝试了上述的一些变体,所有这些都具有相同的结果(由 tensorflow 服务提供相同的预测)。

然后我 运行 tensorflow-serving 就像:

docker run -p 8501:8501 \
  -v ./resnet-classifier:/models/resnet-classifier \
  -e MODEL_NAME=resnet-classifier -e MODEL_BASE_PATH=/models \
  -t tensorflow/serving

最后,我使用以下函数对 tensorflow 服务进行预测:

def imagepath_to_tfserving_payload(img_path):
    import numpy as np
    from keras.preprocessing import image
    from keras.applications.resnet50 import preprocess_input
    img = image.img_to_array(image.load_img(img_path, target_size=(224, 224)))
    X = np.expand_dims(img, axis=0).astype('float32')
    X = preprocess_input(X)
    payload = dict(instances=X.tolist())
    payload = json.dumps(payload)
    return payload

def tfserving_predict(image_payload, url=None):
    import requests
    if url is None:
        url = 'http://localhost:8501/v1/models/resnet-classifier:predict'
    r = requests.post(url, data=image_payload)
    pred_json = json.loads(r.content.decode('utf-8'))
    from keras.applications.resnet50 import decode_predictions
    predictions = decode_predictions(np.asarray(pred_json['predictions']), top=3)[0]
    return predictions

然后我使用上面的两个函数,从 ipython shell 到 select 来自 imagenet 的 val 集的随机图像,我已经在本地存储了。问题是 tensorflow 服务总是对我发送的所有图像返回相同的预测。

每次我使用上面的第一个脚本导出模型时,我都会变得略有不同 classes,第一个 class 的置信度为“1”,其他的置信度为“0”,例如:

# Serialization 1, in ./resnet-classifier/1 always returning:
[
  [
    "n07745940",
    "strawberry",
    1.0
  ],
  [
    "n02104029",
    "kuvasz",
    1.4013e-36
  ],
  [
    "n15075141",
    "toilet_tissue",
    0.0
  ]
]

# Serialization 2, in ./resnet-classifier/2 always returning:
[
  [
    "n01530575",
    "brambling",
    1.0
  ],
  [
    "n15075141",
    "toilet_tissue",
    0.0
  ],
  [
    "n02319095",
    "sea_urchin",
    0.0
  ]
]

这可能与 Tensorflow : serving model return always the same prediction 有关,但我不知道那里的答案(没有被接受的答案)有何帮助。

有人知道上面出了什么问题吗?如何解决?

我有时会在忘记规范化图像时遇到这种问题。我认为 resnet 接受 0. 和 1.(或者 -1. 到 1.)之间的浮点数格式的图像。我不知道 preprocess_input 函数的作用,但您可以检查它是否 returns 数组以预期的格式。

我发现调用 sess.run(tf.global_variables_initializer()) 会覆盖预训练的权重,线索在 http://zachmoshe.com/2017/11/11/use-keras-models-with-tf.html

我的解决方案非常简单,只需将原始问题中的第一段代码更改为以下代码,即调用 tf.global_variables_initializer() before 模型实例化/ 负重:

import tensorflow as tf
sess = tf.Session()
sess.run(tf.global_variables_initializer())

from keras import backend as K
K.set_session(sess)
K.set_learning_phase(0)

# Modelo resnet con pesos entrenados en imagenet
from keras.applications.resnet50 import ResNet50
model = ResNet50(weights='imagenet')

# exportar en tensorflow
import os
versions = [ int(x) for x in os.listdir('./resnet-classifier') ]
version_number = max(versions) + 1 if versions else 1
export_path = './resnet-classifier/{}'.format(version_number)

tf.saved_model.simple_save(sess, export_path,
        inputs=dict(input_image=model.input),
        outputs={t.name:t for t in model.outputs}
)