MNN Batch_size 模型推理
参考 MNN 源码中 demo\exec\pictureRecognition.cpp文件，进行修改实现 Batch_size >= 1 进行模型推理。
模型导出时需要导出为动态 Batch_size
model.eval()
torch.onnx.export(model, img,
export_file, verbose=True, opset_version=12,
training=torch.onnx.TrainingMode.EVAL,
do_constant_folding=True,
input_names=['images'],
output_names=['output'],
dynamic_axes={
'images': {0: 'batch_size'}, # variable lenght axes
'output': {0: 'batch_size'}
}
)其他需要注意：网络结构中 Reshape 或者 View操作不能出现 batchsize , 需要换为 -1
def channel_shuffle(x, groups):
# type: (torch.Tensor, int) -> torch.Tensor
batchsize, num_channels, height, width = x.data.size()
channels_per_group = num_channels // groups
# reshape
x = x.view(-1, groups, channels_per_group, height, width)
x = torch.transpose(x, 1, 2).contiguous()
# flatten
_, c, num , w, h = x.shape
# x = x.view(batchsize, -1 , height, width) error
x = x.view(-1, c*num , height, width) # correct
return x用一个简单地多标签分类为示例进行测试。
参考