-
Notifications
You must be signed in to change notification settings - Fork 18
New correct model, gpu functionality, overfit of a single image #12
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: master
Are you sure you want to change the base?
Conversation
Network modelling prototype 1
new model in line with python code
|
Dup of #9? |
Almost complete rewrite. Use what you like. |
| y = reshape(y, rsize..., 1, 1) | ||
| x, y | ||
|
|
||
| @with_kw mutable struct Dataset |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This may not scale to how every dataset is stored, depending on the directory layout
| return value.i | ||
| end | ||
|
|
||
| function target_to_onehot(target, nfeatures) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This should be using Flux.onehot otherwise it would be slow.
| Chain(x->expand_dims(x,2), | ||
| BatchNorm(out_ch), | ||
| x->squeeze(x)) | ||
| DoubleConv(in_channels, out_channels; conv_kernel = (3,3), conv_pad = (1,1)) = |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Might be more idiomatic to match the Flux.Conv API
This actually works