Paddlepaddle:一般任务最佳实践

图像分类

1.优化器使用Adam

opt = fluid.optimizer.Adam(learning_rate=0.001, parameter_list=model.parameters())

2.激活函数使用relu

self.conv1 = Conv2D(num_channels=1, num_filters=6, filter_size=5, act='relu')

3.求loss使用softmax

loss = fluid.layers.softmax_with_cross_entropy(logits, label)