随笔分类 -  模型压缩和加速

摘要：import torch input = torch.randn(4,requires_grad = True) output = torch.sign(input) loss = output.mean() loss.backward() print(input) print(input.grad 阅读全文

摘要：import torch.nn as nn import torch import torch.nn.functional as F class FP_Conv2d(nn.Module): def __init__(self, input_channels, output_channels, ker 阅读全文