Tensorflow 2.X h5转pb

Posted 2022-02-18 AI浩

模型选择昨天训练的EfficientNetB0模型，具体的训练方法详见：
https://wanghao.blog.csdn.net/article/details/122807559
将训练出来的模型转为pb文件，转化过程比较简单，如下：
新建h5_2_pb.py,插入下面的代码：

from tensorflow.keras import models
from tensorflow.keras.applications import EfficientNetB0
from tensorflow.python.keras.layers import Dense
from tensorflow.python.keras.models import Sequential

pb_outpath='model_pb'
model = Sequential()
model.add(EfficientNetB0(input_shape=(224, 224, 3), include_top=False, pooling='avg', weights='imagenet'))
model.add(Dense(12, activation='softmax'))
models.load_model('my_model.h5')
models.save_model(model, pb_outpath)

运行结果：