ac368987658176719fab4039d5d0e04942eead71,python/mxnet_benchmarks/models/alexnet.py,AlexNet,__init__,#AlexNet#Any#,30

Before Change

        relu4 = mx.symbol.Activation(name="relu4", data=conv4, act_type="relu")

        conv5 = mx.symbol.Convolution(name="conv5", data=relu4, kernel=(3, 3), pad=(1, 1), num_filter=256, num_group=1)
        relu5 = mx.symbol.Activation(name="relu5", data=conv5, act_type="relu")
        pool5 = mx.symbol.Pooling(name="pool5", data=relu5, kernel=(3, 3), stride=(2, 2), pool_type="max")

        flatten = mx.symbol.Flatten(data=pool5)
        fc6 = mx.symbol.FullyConnected(name="fc6", data=flatten, num_hidden=4096)
        relu6 = mx.symbol.Activation(name="relu6", data=fc6, act_type="relu")
        drop6 = mx.symbol.Dropout(name="drop6", data=relu6, p=0.5) if training else relu6

        fc7 = mx.symbol.FullyConnected(name="fc7", data=drop6, num_hidden=4096)
        relu7 = mx.symbol.Activation(name="relu7", data=fc7, act_type="relu")
        drop7 = mx.symbol.Dropout(name="drop7", data=relu7, p=0.5) if training else relu7

        self.__output = self.add_head_nodes(drop7)

After Change

        layers = Layers(params)

        data = self.add_data_node()
        data = Layers.conv_transform_layout(data, params["input_layout"], params["model_layout"])

        conv1 = layers.Convolution(name="conv1", data=data, kernel=(11, 11), stride=(4, 4), num_filter=96)
        relu1 = layers.Activation(name="relu1", data=conv1, act_type="relu")

Instances