Live forgetting visualizer for continual learning — watch your neural network forget in real time
Project description
NeuroFade
Live Forgetting Visualizer — see your neural network forget in real time.
What
Tracks per-neuron activation retention across training tasks and renders the forgetting process as animated heatmaps. Watch your model lose knowledge in real time.
Install
pip install neurofade
Quick Start
from neurofade import ForgettingVisualizer
viz = ForgettingVisualizer(model, baseline_task=train_task_1_loader)
# Wraps your training loop
with viz.watch():
trainer.fit(task_2_loader)
viz.export("forgetting.mp4")
viz.share() # Upload to public URL
Why
Catastrophic forgetting is invisible until it tanks your model. NeuroFade makes it undeniable.
- Layers fade in real-time — see which neurons die first
- Export as GIF/MP4 — shareable artifacts
- Per-task baselines — measures retention against any previous task
- Framework-agnostic — works with PyTorch, TensorFlow, JAX
The Vibe
Your model is dying. Now you can see it.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file neurofade-0.1.0.tar.gz.
File metadata
- Download URL: neurofade-0.1.0.tar.gz
- Upload date:
- Size: 5.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
a71ef66185698507d8da850c99b34f6faad00cc1955a60bfb36d11c8fc3d98f9
|
|
| MD5 |
8c13dd97da358690ef7745584d814dff
|
|
| BLAKE2b-256 |
5903ddeaa97a94d2460db9af50ba5c71c510750cd47b7e0d011d4a649e365668
|
File details
Details for the file neurofade-0.1.0-py3-none-any.whl.
File metadata
- Download URL: neurofade-0.1.0-py3-none-any.whl
- Upload date:
- Size: 6.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f4083297b544eb48c135cc7dd5a4ba18762a8fbf4b011bad60d95a48306e1f9c
|
|
| MD5 |
e4b3833b65c0346f920953284ebe8702
|
|
| BLAKE2b-256 |
1bfdf04845adf3085831496d8165112505ad2b3ead7fab1b0c36cd027729ee03
|