FasterAI: A repository for making smaller and faster models with the FastAI library.
License: MIT License
I'm a postdoc researcher based in Belgium. I am mostly interested in making neural networks smaller and faster ✂️
Line 26 in ea24687
I dont know what is the reason behind this error, I tried to debug it but couldnt get to the core of it
@nathanhubens Can you please take a look into this! I tried the earlier version of pruner with my previous experiments and they seemed to work well. I think this has to do something with my architecture as well. For reference, I am trying to build a super-resolution architecture which looks like this
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.nn.init as init
class depthwise_conv(nn.Module):
def __init__(self, nin, kernel_size, padding, stride=1, dilation=1):
super(depthwise_conv, self).__init__()
self.depthwise = nn.Conv2d(nin, nin, kernel_size=kernel_size, stride=stride, padding=padding, dilation=dilation, groups=nin)
def forward(self, x):
out = self.depthwise(x)
return out
class dw_block(nn.Module):
def __init__(self, nin, kernel_size, padding=1, stride=1, dilation=1):
super(dw_block, self).__init__()
self.dw_block = nn.Sequential(
depthwise_conv(nin, kernel_size, stride, padding, dilation),
)
def forward(self, x):
out = self.dw_block(x)
return out
class pointwise_conv(nn.Module):
def __init__(self, nin, nout, padding=0, stride=1):
super(pointwise_conv, ####qweqwe##self).__init__()
self.pointwise_block = nn.Sequential(
nn.Conv2d(nin, nout, kernel_size=1, stride=stride, padding=padding),
)
def forward(self, x):
out = self.pointwise_block(x)
return out
class SuperRes(nn.Module):
def __init__(self, scale_factor=3, num_channels=1, d=32, s=12, m=4):
super(SuperRes, self).__init__()
self.first_part = nn.Sequential(
nn.Conv2d(num_channels, d, kernel_size=5, padding=5//2),
nn.PReLU(d)
)
self.mid_part = [nn.Conv2d(d, s, kernel_size=1), nn.PReLU(s)]
for _ in range(m):
self.mid_part.extend([nn.Conv2d(s, s, kernel_size=3, padding=2, dilation=2), nn.PReLU(s)])
self.mid_part.extend([nn.Conv2d(s, d, kernel_size=1), nn.PReLU(d)])
self.mid_part = nn.Sequential(*self.mid_part)
#self.last_part = nn.ConvTranspose2d(d, num_channels, kernel_size=9, stride=scale_factor, padding=9//2,
#output_padding=scale_factor-1)
self.dp1 = nn.Sequential(
dw_block(32, kernel_size=3, dilation=2),
nn.PReLU(32),
pointwise_conv(nin = 32, nout = 24),
nn.PReLU(24),
dw_block(24, kernel_size=3, dilation=2),
nn.PReLU(24),
pointwise_conv(nin = 24, nout = 16),
nn.PReLU(16),
dw_block(16, kernel_size=3),
nn.PReLU(16),
pointwise_conv(nin = 16, nout = 8),
nn.PReLU(8),
dw_block(8, kernel_size=3),
nn.PReLU(8),
pointwise_conv(nin = 8, nout = 16),
nn.PReLU(16),
dw_block(16, kernel_size=5),
nn.PReLU(16),
pointwise_conv(nin = 16, nout = 24, padding=2),
nn.PReLU(24),
dw_block(24, kernel_size=5),
nn.PReLU(24),
pointwise_conv(nin = 24, nout = 32, padding=2), # PADDING = 2 here
nn.PReLU(32),
)
self.conv = nn.Conv2d(32, 9, 3, 1, 1)
self.last_part = nn.PixelShuffle(scale_factor)
self._initialize_weights()
def _initialize_weights(self):
for m in self.first_part:
if isinstance(m, nn.Conv2d):
nn.init.normal_(m.weight.data, mean=0.0, std=math.sqrt(2/(m.out_channels*m.weight.data[0][0].numel())))
nn.init.zeros_(m.bias.data)
for m in self.mid_part:
if isinstance(m, nn.Conv2d):
nn.init.normal_(m.weight.data, mean=0.0, std=math.sqrt(2/(m.out_channels*m.weight.data[0][0].numel())))
nn.init.zeros_(m.bias.data)
for m in self.dp1:
if isinstance(m, nn.Conv2d):
nn.init.normal_(m.weight.data, mean=0.0, std=math.sqrt(2/(m.out_channels*m.weight.data[0][0].numel())))
nn.init.zeros_(m.bias.data)
def forward(self, x):
global_residual = x
x1 = self.first_part(x)
x2 = self.mid_part(x1)
x3 = self.dp1(x2)
x4 = x3 + x1
x = self.conv(x4)
x = x + global_residual
x = self.last_part(x)
return x
if __name__ == "__main__":
model = SuperRes()
print(model)
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.