计算机视觉可视化/调参工具

模型参数计算
- flops-counter.pytorch
- thop
模型可视化
训练可视化
- pytorch-cnn-visualizations

模型参数计算

https://www.cnblogs.com/xuanyuyt/p/12653041.html#_label4

flops-counter.pytorch

地址：https://github.com/sovrasov/flops-counter.pytorch

import torch
from ptflops import get_model_complexity_info
from models import nin
shape = (3, 32, 32)
# 加载训练完毕的模型
with torch.cuda.device(0):  
    model = nin.Net() 
    model.load_state_dict(torch.load("./models_save/nin.pth")["state_dict"])
    flops, params = get_model_complexity_info(model, shape, as_strings=True, print_per_layer_stat=True)
    print('{:<30}  {:<8}'.format('Computational complexity: ', flops))
    print('{:<30}  {:<8}'.format('Number of parameters: ', params))

乘积累加运算（英语：Multiply Accumulate, MAC）
参考：https://www.zhihu.com/question/65305385

thop

https://pypi.org/project/thop/

pip install --upgrade git+https://github.com/Lyken17/pytorch-OpCounter.git

from torchvision.models import resnet50
from thop import profile
model = resnet50()
input = torch.randn(1, 3, 224, 224)
macs, params = profile(model, inputs=(input, ))
"""
# 加载训练的模型
import torch
from torchvision.models import resnet50
from thop import profile
from models import nin
model = nin.Net()
model.load_state_dict(torch.load("./models_save/nin.pth")["state_dict"])
input = torch.randn(1, 3, 32, 32)
macs, params = profile(model, inputs=(input, ))
print("macs", macs)
print("params", params)
"""

模型可视化

代码生成

from graphviz import Digraph
import torch
from torch.autograd import Variable
import torchvision
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
model  = torchvision.models.resnet18()
data = torch.rand(1, 3, 128, 128)
y = model(data)
g = make_dot(y)
g.view() # 会生成一个 pdf 文件