ai-agent

MiSLAS

Improving Calibration for Long-Tailed Recognition (CVPR2021)
153 stars24 forksPythonUpdated 11/10/2021100% free · open source
What it does

MiSLAS is a Python tool that improves calibration for long-tailed recognition in computer vision tasks, helping models to better handle imbalanced datasets.

When to use it
  • When dealing with datasets that have a long-tailed distribution, where some classes have many more instances than others
  • When evaluating the calibration of a model on a specific task, such as image classification or object detection
  • When trying to improve the performance of a model on a long-tailed dataset, especially on the tail classes
Ready-to-paste prompt
python train.py --config config.py --dataset cifar10 --model resnet32
Heads up: Make sure to install the required Python packages by running pip install -r requirements.txt before running any scripts, and also ensure that the GPU is properly configured if using CUDA for acceleration
Saves to your device
Use with Claude
New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

⚡ Runs out of the box

A prompt/skill package — nothing to install beyond adding it to your AI tool.

Try it instantly — no install
Claude Code
mkdir -p ~/.claude/skills/mislas && curl -fsSL https://workflowstacks.com/api/skills/mislas/claude-skill -o ~/.claude/skills/mislas/SKILL.md
Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)
claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How MiSLAS works
Codeflow
Free to inspect

MiSLAS is mostly documents (8 doc files, 18 small scripts) — something you read, not something you run. There is nothing to install. Last commit 59 months ago, MIT license.

Size
Mostly documents
8 documents · 18 small scripts · days to read — use, don't read
Setup
Nothing to install
A guide / curated list. Just read it and follow the links.
Runs on
Nowhere — you read it
A reading resource, not a program.
Python 100%
Where to start reading
  1. 1
    README.md
    Start here — what it does and how to install it
What's in each folder
config/Configuration18 files
datasets/Data files12 files
utils/Helper scripts4 files
models/Data models & types3 files
assets/Images & static assets2 files
READMEHas testsDocumentedMIT licenseLast update 59 mo ago
Quick Actions
Details
Creator
JIA-Lab-research
Language
Python
Category
ai-agent
Published
3/1/2021

Are you the creator of this tool? Claim your listing → and earn 85% of every sale.