Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Conv1D Student Teacher Model #37

Open
wants to merge 33 commits into
base: main
Choose a base branch
from
Open

Conv1D Student Teacher Model #37

wants to merge 33 commits into from

Conversation

ChilloutCharles
Copy link
Owner

Changes

  • Added pretrained decoder and encoder as a teacher
  • Using pieces of the pretrained encoder surrounded by custom first layer and spatial attention as a student
  • Added visualizations to evaluate autoencoder and classifier performance

Be sure to pull requirements.txt again to make sure you have the correct dependencies. This will work with existing recordings.

ChilloutCharles and others added 30 commits September 20, 2024 01:30
@ChilloutCharles
hard work
ef29b73
@ChilloutCharles
dilation from paper
b8fde60
@ChilloutCharles
additional layer and GAP2D
9e50110
@ChilloutCharles
Additional Linear Layer
0a26e09
@ChilloutCharles
better graph
c297b15
@ChilloutCharles
revert to single dense last layer
4c0acf5
@ChilloutCharles
DAE
c91f041
@ChilloutCharles
expand classify tuning
937e214
@ChilloutCharles
expand classify tuning
e68735b
@ChilloutCharles
Merge branch 'efficient-conv2d-autoencoder' of https://github.com/Chi…
be47377 
…lloutCharles/BrainFlowsIntoVRChat into efficient-conv2d-autoencoder
@ChilloutCharles
increase noise
e3eb770
@ChilloutCharles
Added Dropout Back
23ae513
@ChilloutCharles
DAE with residuals
ea89f1c
@ChilloutCharles
shrink kernel pointwise conv update
c0feae8
@ChilloutCharles
Replaced Noise with Channel Dropout
a0abae1
@ChilloutCharles
dilated expander and noise for classifier
d0ebd72
@ChilloutCharles
causal padding for classifier dialated convs
fff110c
@ChilloutCharles
cleanup
7ac5684
@ChilloutCharles
back to conv1d
89b1765 
switch back to conv1d, use standard scaler
@ChilloutCharles
standard scaling for user data
5cae478
@ChilloutCharles
Reduce Training Patience
daccf90
@ChilloutCharles
block change, perceptual loss, mse loss
8d411e9
@ChilloutCharles
Merge branch 'conv1d-improvement' of https://github.com/ChilloutCharl…
e40d8ad 
…es/BrainFlowsIntoVRChat into conv1d-improvement
@ChilloutCharles
adding layer norms to final layer
32fa327
@ChilloutCharles
remove last layer norm
80c97de
@ChilloutCharles
lora added
e2f637e
@ChilloutCharles
lora layers, perceptual classifier
d969249
@ChilloutCharles
student teacher model
ccf0245
@ChilloutCharles
decoder fix
5a4e59e
@ChilloutCharles
cleanup
1e230ac
@ChilloutCharles ChilloutCharles added the enhancement New feature or request label Oct 14, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@RantiMess RantiMess Awaiting requested review from RantiMess

@DeliciousSalad DeliciousSalad Awaiting requested review from DeliciousSalad

At least 1 approving review is required to merge this pull request.

Assignees
No one assigned
Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@ChilloutCharles