Entrenar

C 69 completed
Library
monorepo / rust · medium
1,523
Files
186,041
LOC
1
Frameworks
11
Languages

Pipeline State

completed
Run ID
#368632
Phase
done
Progress
1%
Started
Finished
2026-04-13 01:31:02
LLM tokens
0

Pipeline Metadata

Stage
Cataloged
Decision
proceed
Novelty
61.67
Framework unique
Isolation
Last stage change
2026-05-10 03:35:24
Deduplication group #48993
Member of a group with 10 similar repo(s) — canonical #57648 view group →
Top concepts (2)
Project DescriptionWeb Backend
Citation: Repobility (2026). State of AI-Generated Code. https://repobility.com/research/

AI Prompt

Create a production-grade neural network training library in pure Rust, similar to Entrenar. I need it to include core features like an autograd engine, various optimizers (like Adam), and support for parameter-efficient fine-tuning methods such as LoRA/QLoRA. Please ensure the architecture supports quantization (QAT, PTQ) and model merging algorithms. The library should also provide utilities for monitoring, including real-time metrics and explainability tools like SHAP.
rust machine-learning neural-network autograd optimization quantization deep-learning
Generated by gemma4:latest

Catalog Information

Entrenar is a production-grade neural network training library in pure Rust, providing features for autograd engine, optimizers, LoRA/QLoRA, quantization, and more.

Description

Entrenar provides everything needed to train neural networks in Rust. It includes an autograd engine, various optimizers, LoRA/QLoRA for parameter-efficient fine-tuning, and quantization methods. Additionally, it offers model merging, knowledge distillation, training loop features, monitoring tools, and explainability techniques.

الوصف

توفير entrenar كل ما يحتاجه لتدريب الشبكات العصبية في الرس. يتضمن ذلك محرك التأريخ الآلي، ووظائف الباحثين المختلفة، و LoRA/QLoRA للتدريب الفعال للمعلمات، ومهارات التحفيز. بالإضافة إلى ذلك، يحتوي على دمج النماذج، وتدريب المعرفة، وخصائص حلقة التدريب، وأدوات المراقبة، والتقنيات التفسيرية.

Novelty

9/10

Tags

neural-network-training rust-library autograd-engine optimizers quantization model-merging knowledge-distillation

Technologies

axum serde tokio

Claude Models

claude-opus-4.6

Quality Score

C
69.1/100
Structure
73
Code Quality
65
Documentation
85
Testing
65
Practices
56
Security
74
Dependencies
60

Strengths

  • CI/CD pipeline configured (github_actions)
  • Consistent naming conventions (snake_case)
  • Properly licensed project

Weaknesses

  • Potential hardcoded secrets in 1 files
  • 21524 duplicate lines detected \u2014 consider DRY refactoring
  • 14 'god files' with >500 LOC need decomposition

Recommendations

  • Add a linter configuration to enforce code style consistency
  • Move hardcoded secrets to environment variables or a secrets manager

Security & Health

11.1h
Tech Debt (A)
A
OWASP (100%)
PASS
Quality Gate
A
Risk (0)
All rows scored by the Repobility analyzer (https://repobility.com)
MIT
License
9.4%
Duplication
Full Security Report AI Fix Prompts SARIF SBOM

Languages

rust
73.4%
markdown
17.3%
json
4.2%
yaml
3.7%
toml
0.5%
shell
0.4%
python
0.2%
text
0.1%
typescript
0.1%
javascript
0.0%
html
0.0%

Frameworks

Axum

Concepts (2)

Repobility · code-quality intelligence · https://repobility.com
CategoryNameDescriptionConfidence
Repobility — the code-quality scanner for AI-generated software · https://repobility.com
auto_descriptionProject Description- What is Entrenar? - Installation - Getting Started80%
auto_categoryWeb Backendweb-backend70%

Quality Timeline

1 quality score recorded.

View File Metrics

Embed Badge

Add to your README:

![Quality](https://repos.aljefra.com/badge/92833.svg)
Quality BadgeSecurity Badge
Export Quality CSVDownload SBOMExport Findings CSV