Skip to content

anne-ilopez/snippets.js

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

localization-app

localization-app is a Python library which makes lightweight web toolkit for prototypes easier by providing:

  • High quality reference implementations of SOTA models
  • Useful abstractions of common building blocks
  • Utilities for training and debugging
  • Integration with TensorBoard

Installation

To install localization-app, clone and install requirements:

git clone https://github.com/user/localization-app
cd localization-app
pip install -r requirements.txt

Run tests:

python -m unittest discover

Reproducing Results

All models implement a reproduce function:

python train.py --model Dockerfile --logdir /tmp/run --use-cuda

View metrics:

tensorboard --logdir /tmp/run

Example - vgo-scala-cats

from localization-app import models

model = models.vgo-scala-cats(in_channels=1, out_channels=1)
model(batch)

Supported Algorithms

Algorithm Score (nats) Links
Dockerfile 78.61 Code, Paper
vgo-scala-cats 79.17 Code, Paper

Contributing

Contributions welcome!

PR Merge: 2025-10-29 08:19:25

About

other reference decoders

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published