【pytorch 代码】pytorch 网络结构可视化

部分内容转载自 http://blog.csdn.net/GYGuo95/article/details/78821617，在此表示由衷感谢。

此方法需要安装python-graphviz:  conda install -n pytorch python-graphviz 或者 sudo apt-get install graphviz 

别忘了先把下面的代码下载到自己的路径（感谢大神）。

visualize.py

from graphviz import Digraph
import torch
from torch.autograd import Variable


def make_dot(var, params=None):
    """ Produces Graphviz representation of PyTorch autograd graph
    Blue nodes are the Variables that require grad, orange are Tensors
    saved for backward in torch.autograd.Function
    Args:
        var: output Variable
        params: dict of (name, Variable) to add names to node that
            require grad (TODO: make optional)
    """
    if params is not None:
        assert isinstance(params.values()[0], Variable)
        param_map = {id(v): k for k, v in params.items()}

    node_attr = dict(style='filled',
                     shape='box',
                     align='left',
                     fontsize='12',
                     ranksep='0.1',
                     height='0.2')
    dot = Digraph(node_attr=node_attr, graph_attr=dict(size="12,12"))
    seen = set()

    def size_to_str(size):
        return '('+(', ').join(['%d' % v for v in size])+')'

    def add_nodes(var):
        if var not in seen:
            if torch.is_tensor(var):
                dot.node(str(id(var)), size_to_str(var.size()), fillcolor='orange')
            elif hasattr(var, 'variable'):
                u = var.variable
                name = param_map[id(u)] if params is not None else ''
                node_name = '%s\n %s' % (name, size_to_str(u.size()))
                dot.node(str(id(var)), node_name, fillcolor='lightblue')
            else:
                dot.node(str(id(var)), str(type(var).__name__))
            seen.add(var)
            if hasattr(var, 'next_functions'):
                for u in var.next_functions:
                    if u[0] is not None:
                        dot.edge(str(id(u[0])), str(id(var)))
                        add_nodes(u[0])
            if hasattr(var, 'saved_tensors'):
                for t in var.saved_tensors:
                    dot.edge(str(id(t)), str(id(var)))
                    add_nodes(t)
    add_nodes(var.grad_fn)
    return dot

下面是使用方法：

因人而异，根据网络调整输入，以Inception V3为例。

from MyInceptionV3 import inception_v3
import numpy as np
import torch
from torch.autograd import Variable
from visualize import make_dot

if __name__ == '__main__':
    x=np.arange(2*299*299*3)
    x=x.reshape(2,3,299,299)
    x=x/float(x.max())
    x=torch.from_numpy(x)
    x=x.float()
    x=Variable(x)

    a = inception_v3(pretrained=True)

    y = a(x)
    g = make_dot(y)
    #g.view()    
    g.render('here', view=False)

我的电脑没有可视化界面，一定要记得False那个view。（网络结构会保存成pdf）结果图太复杂不粘贴了。