Skip to content
Catégories de projets

Apprentissage automatique · GitHub

Explorez les outils pour entraîner, évaluer et exécuter des modèles, et comparez leur popularité.

Projets GitHub utiles

12,832 projets

xla

openxla

4,681étoiles+1in 24h
xla preview

Un compilateur d’apprentissage automatique pour les GPU, les CPU et les accélérateurs de machine learning

LMFlow

OptimalScale

8,483étoiles+1in 24h
LMFlow preview

Boîte à outils extensible pour le fine-tuning et l’inférence de grands modèles de fondation. De grands modèles pour tous.

  • LoRA LoRA is a parameter-efficient finetuning algorithm and is more efficient than full finetuning. Check out finetuning-lora for more details.
  • FlashAttention LMFlow supports both FlashAttention-1 and the latest FlashAttention-2. Check out flash_attention for more details.

optuna

optuna

15,071étoiles+1in 24h
optuna preview

Framework d’optimisation des hyperparamètres

  • Handle a wide variety of tasks with a simple installation that has few requirements.
  • Define search spaces using familiar Python syntax including conditionals and loops.

whatsmybitrate

oren-cohen

33étoiles
whatsmybitrate preview

Whatsmybitrate analyzes audio files for quality metrics such as bit rate, frequency, and codec type in bulk. It also generates spectrograms for visual representation of the audio spectrum. It supports a variety of audio formats, including MP3, FLAC, WAV, AAC, M4A, and more.

  • Detailed Quality Analysis: Reports codec, sample rate, stated bit rate, and more.
  • Perceptual Quality Estimation: Estimates the equivalent bitrate (e.g., 128, 192, 256, 320 kbps) based on spectral frequency cutoffs.

Oxen

Oxen-AI

1,211étoiles+1in 24h
Oxen preview

Lightning fast data version control system for large repositories of data. Feels like git, pushes and pulls like oxen.

  • 🔥 Fast (efficient indexing and syncing of data)
  • 🧠 Easy to learn (same commands as git)

Paddle

PaddlePaddle

24,149étoiles+1in 24h

Apprentissage profond parallèle et distribué : framework de machine learning issu de la pratique industrielle (framework principal de Paddle pour l’entraînement haute performance sur une seule machine et en environnement distribué, ainsi que pour le déploiement multiplateforme d’applications de deep learning et de machine learning)

what-if-tool

PAIR-code

1,029étoiles+1in 24h
what-if-tool preview

Source code/webpage/demos for the What-If Tool

  • For classification models, this includes inferred label, confidence of inferred label, and inference correctness.
  • For regression models, this includes inferred score and amount of error (including absolute or squared error) in the inference.

diffrax

patrick-kidger

2,130étoiles+1in 24h

Solveurs numériques d’équations différentielles dans JAX. Différentiables automatiquement et compatibles avec les GPU. https://docs.kidger.site/diffrax/

NeuralFoil

peterdsharpe

510étoiles
NeuralFoil preview

NeuralFoil is a practical airfoil aerodynamics analysis tool using physics-informed machine learning, exposed to end-users in pure Python/NumPy.

  • Post-stall aerodynamics (i.e., truly 360 degree range of $\alpha$). This is useful for applications like wind turbine blades or propeller roots, where the airfoil may be operating at high angles of attack.
  • Control surface deflections. Currently only trailing-edge control surface deflections are supported in AeroSandbox's NeuralFoil interface.

gensim

piskvorky

16,518étoiles+1in 24h
gensim preview

Modélisation de thèmes pour les humains

  • All algorithms are memory-independent w.r.t. the corpus size (can process input larger than RAM, streamed, out-of-core),
  • easy to plug in your own input corpus/datastream (trivial streaming API)