ML Application · Python

sfc128-emotion

Compact 792K-parameter emotion classifier with fast FP32 CUDA Graph inference; 86.33% held-out test accuracy and 0.8202 macro-F1 on DAIR-AI Emotion.

Daniel Strandt@DanielStrandt · Atlas 71.0 · #2= builder

View on GitHub2 Open Daniel Strandt's Atlas

What it is

The repository provides a fully fine-tuned 128-layer emotion classification model with CLI, API, and deployment tools. It includes training, evaluation, and inference optimization scripts, achieving 86.33% test accuracy on the DAIR.AI Emotion dataset.

Scored atccfabfd8 Nov 2026Changes since

Merit describes the code at this commit. Later pushes count once the owner rescans. How merit is scored

Where it counts

Machine Learning90%
NLP & LLMs80%
MLOps30%
Data Science20%

How much of the project is about each field, as scoring counts it. It ranks in the field it is most about.

Around it on the Machine Learning board

Whole board

RankProjectBuilderFieldMerit

04celiumsnnArchitecture Modification / Research · PythonThe repository implements Mycelium, a verification-first spiking neural network architecture derived from the constraint set of the CeliumNeUR neuromorphic chip. It features integer LIF neurons, block-sparse synapses with ternary weights, and L0 topology learning, validated on…72.1/100AdvancedMario GutierrezMachine Learning
05datarustLibrary / Framework · Rustdatarust is a Rust Cargo workspace providing scikit-learn‑style preprocessing and classical machine learning components together with a data‑profiling and data‑quality reporting library, all designed with zero external dependencies by default and a type‑safe API.71.0/100AdvancedMurat GençMachine Learning
06sfc128-emotionML Application · Python69.8/100AdvancedDaniel StrandtMachine Learning
07VoiceManagerInference System · PythonVoiceManager is a self-hosted, full-duplex streaming speech-to-text and text-to-speech server over WebSocket, built around a Python FastAPI server and a C++ sidecar process that runs STT/TTS engines. It ships a Python client, an auth/CLI layer, model-download setup, and an…68.9/100AdvancedTacoTakumiMachine Learning
08AECS-schedulerLibrary / Framework · PythonAECS is a PyTorch learning rate scheduler that adjusts the learning rate in real time based on training signals such as gradient norm and loss, switching between four modes (BASELINE, RECOVERY, EXPLORE, STABILIZE) using event-driven logic.67.6/100AdvancedRickMachine Learning

What it is

Where it counts

Around it on the Machine Learning board

More from Daniel Strandt