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

PHATE #210

Open
wants to merge 41 commits into
base: ntxent_loss
Choose a base branch
from
Open

PHATE #210

wants to merge 41 commits into from
+751 −165

Conversation

edyoshikun
Copy link
Contributor

  • Adding support for PHATE
  • Adding reviewed Alishba's scripts for ALFI.

ziw-liu and others added 22 commits November 8, 2024 09:47
@ziw-liu
translation: fix validation loss aggregation (#202)
db80819
@edyoshikun
exposing prefetch and persistent worker (#203)
820c805
@alishbaimran
metrics for dynamic, smoothness and docstrings
a73b9a0
@alishbaimran
updated metrics and plots for distance
79db1e4
@alishbaimran
fixed CI test cases
42cf756
@edyoshikun
nexnt loss prototype
0a23fc4
@edyoshikun
fix bug with z_scale_range in hcs datamodule. If the value is an int …
f2d75cd 
…this does not work.
@edyoshikun
exclude the negative pair from dataloader and forward pass
aa74eff
@edyoshikun
adding option using pytorch-metric-learning implementation
771d33d 
and modifying previous to match same input args
@edyoshikun
removing our implementation of NTXentLoss and using pytorch metric
46974bc
@edyoshikun
ruff
1fde4ab
@edyoshikun
prototype for phate and umap plot
3f8363d
@edyoshikun
- proofreading the calculations
26ebe74 
- removing unecessary calls to ALFI script
- simplifying code to re-use functions
@ziw-liu @edyoshikun
methods to rank nearest neighbors in embeddings
7a3b036
@ziw-liu @edyoshikun
example script to plot state change of a single track
3992d9d
@ziw-liu @edyoshikun
test using scaled features
cae8290
@edyoshikun
phate embeddings
30f4246
@edyoshikun
removing dataframe from the compute_phate
356c15d 
adding docstring
@edyoshikun
adding phate to the prediction writer and moving it as dependency.
57abb19
@edyoshikun
changing the phate defaults in the prediction writer.
19fdcf2
@edyoshikun
ruff
1ee8af1
@edyoshikun
fixing bug in phate in predict writer
14b7368
ziw-liu
ziw-liu reviewed Nov 21, 2024
View reviewed changes
pyproject.toml Outdated Show resolved Hide resolved
@edyoshikun edyoshikun changed the title Phate PHATE Nov 27, 2024
@Soorya19Pradeep
Copy link
Contributor

@edyoshikun , can we add the calculation and storage of PHATE in the predictions in this PR?

edyoshikun and others added 12 commits December 18, 2024 16:32
@edyoshikun
nexnt loss prototype
b75ddd1
@edyoshikun
exclude the negative pair from dataloader and forward pass
98b36de
@edyoshikun
adding option using pytorch-metric-learning implementation
5bc2ff6 
and modifying previous to match same input args
@edyoshikun
removing our implementation of NTXentLoss and using pytorch metric
472be4e
@edyoshikun
ruff
66b3899
@ziw-liu @edyoshikun
remove blank line diff
9cc35cf
@ziw-liu @edyoshikun
remove blank line diff
3c96f13
@edyoshikun
simplying the engine
d3d4715
@ziw-liu @edyoshikun
explicit target shape argument in the HCS data module
70dfb5d
@ziw-liu @edyoshikun
Revert "explicit target shape argument in the HCS data module"
a845847 
This reverts commit 464d4c9.
@ziw-liu @edyoshikun
Explicit target shape argument in the HCS data module (#212)
a6e2318 
* explicit target shape argument in the HCS data module

* update docstring

* update test cases
@edyoshikun
Gradio example (#158)
fa74b0d 
* initial demo

* using the predict_step

* modifying paths to chkpt and example pngs

* updating gradio as the one on Huggingface
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@ziw-liu ziw-liu ziw-liu left review comments

@Soorya19Pradeep Soorya19Pradeep Awaiting requested review from Soorya19Pradeep

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@edyoshikun @Soorya19Pradeep @ziw-liu @alishbaimran