from 手势识别.generate_data import data_reader
from 手势识别.build_model import FullNet_Model
import paddle.fluid as fluid
import paddle as paddle
import numpy as np
test_reader = paddle.batch(reader=data_reader('./test_data.list'), batch_size=32)
with fluid.dygraph.guard():
    accs=[]
    model_dict,_=fluid.load_dygraph('FullNet_Model')
    model = FullNet_Model()
    model.load_dict(model_dict) #加载模型
    model.eval() #开启评估模式
    for batch_id,data in enumerate(test_reader()):
        images = np.array([x[0].reshape(3,100,100) for x in data],np.float32)
        labels = np.array([x[1] for x in data]).astype("int64")
        labels = labels[:,np.newaxis]

        image = fluid.dygraph.to_variable(images)
        label = fluid.dygraph.to_variable(labels)

        predict = model(image)
        acc=fluid.layers.accuracy(predict,label)
        accs.append(acc.numpy()[0])
        avg_acc=np.mean(accs)

    print('平均准确率为',avg_acc)