tdevries / enas_pytorch Goto Github PK
View Code? Open in Web Editor NEWPyTorch port of "Efficient Neural Architecture Search via Parameters Sharing"
PyTorch port of "Efficient Neural Architecture Search via Parameters Sharing"
There seems to be some problems in the file ‘models/shared_cnn.py’
class ENASLayer(nn.Module):
def forward(self, x, prev_layers, sample_arc):
layer_type = sample_arc[0]
if self.layer_id > 0:
skip_indices = sample_arc[1]
else:
skip_indices = []
.......
for i, skip in enumerate(skip_indices):
if skip == 1:
out += prev_layers[i]
out = self.bn(out)
return out
In the above code, when layer_id = 0, an error occurs when running to enumerate (skip_indices). Is this part of the code not complete enough?
Looking forward to and thank you for your answer !
Hi,
Thanks for the code sharing!
I tried to read it and run it. 10 epochs in the main function take one Nividia Tesla P100 12G 15h to run. Am I wrong or it just takes this amount of time?
RT, is this a complete implementation? Do you get the reported performance on CIFAR and Imagenet?
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 is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
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.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.