Main Project

F 46 completed
Web App
unknown / python · tiny
48
Files
11,751
LOC
2
Frameworks
3
Languages

Pipeline State

completed
Run ID
#345601
Phase
done
Progress
1%
Started
Finished
2026-04-13 01:31:02
LLM tokens
0

Pipeline Metadata

Stage
Cataloged
Decision
proceed
Novelty
65.00
Framework unique
Isolation
Last stage change
2026-05-10 03:34:14
Deduplication group #47951
Member of a group with 14 similar repo(s) — canonical #85543 view group →
Top concepts (2)
Project DescriptionWeb Backend
Methodology: Repobility · https://repobility.com/research/state-of-ai-code-2026/

AI Prompt

Create an automated, end-to-end multimodal machine learning pipeline called AutoVision+. I need it to handle data ingestion from CSV, Parquet, image directories, and ZIP archives. The system must support multimodal preprocessing, selecting vision backbones like ConvNeXt-Tiny or ResNet-50 for images, text encoders like DeBERTa-v3-base for text, and using an MLP encoder for tabular data. Crucially, it needs built-in hyperparameter tuning, drift detection, and explainability features. The entire process should be managed within a single, opinionated pipeline structure.
python fastapi machine-learning multimodal automl pytorch image-processing nlp tabular-data pipeline
Generated by gemma4:latest

Catalog Information

AutoVision+ automates end‑to‑end multimodal machine learning, handling image, text, and tabular data with built‑in hyperparameter tuning, drift detection, and explainability.

Description

AutoVision+ is an end‑to‑end pipeline that automatically trains models on fused image, text, and tabular data. It selects appropriate backbones for each modality, applies feature engineering, and builds a fusion head without manual intervention. Built‑in hyperparameter optimization runs efficiently by sharing pretrained encoders across trials, reducing GPU memory waste. The system monitors data drift and provides explainable predictions through a web interface. It targets data scientists and ML engineers who need a reliable, repeatable workflow for multimodal projects.

الوصف

يُسَهل هذا النظام عملية التعلم الآلي المتعدد الوسائط من البداية إلى النهاية، مع دعم الصور والنصوص والبيانات الجداولية، ويشمل تحسين المعلمات، واكتشاف الانحراف، وشرح النماذج. يختار بنى أساسية مناسبة لكل وسيلة، ويطبق هندسة الميزات، ويُبني رأس دمج دون تدخل يدوي. يضمن تحسين المعلمات عبر مشاركة المشفرات المدربة مسبقاً بين التجارب، ما يقلل استهلاك الذاكرة الرسومية. يراقب الانحراف في البيانات ويُقدّم توقعات قابلة للشرح عبر واجهة ويب. يهدف إلى علماء البيانات ومهندسي التعلم الآلي الذين يحتاجون إلى سير عمل موثوق ومتكرر لمشاريع متعددة الوسائط.

Novelty

8/10

Tags

multimodal-machine-learning automated-model-training hyperparameter-optimization data-drift-monitoring model-explainability end‑to‑end-pipeline image‑text‑tabular-fusion

Technologies

fastapi huggingface numpy pandas pydantic pytorch scikit-learn scipy streamlit uvicorn

Claude Models

claude (unknown version) claude-opus-4.6

Quality Score

F
46.1/100
Structure
38
Code Quality
54
Documentation
65
Testing
0
Practices
60
Security
64
Dependencies
60

Strengths

  • Consistent naming conventions (snake_case)

Weaknesses

  • No LICENSE file \u2014 legal ambiguity for contributors
  • No tests found \u2014 high risk of regressions
  • No CI/CD configuration \u2014 manual testing and deployment
  • 1294 duplicate lines detected \u2014 consider DRY refactoring
  • 5 'god files' with >500 LOC need decomposition

Recommendations

  • Add a test suite \u2014 start with critical path integration tests
  • Set up CI/CD (GitHub Actions recommended) to automate testing and deployment
  • Add a linter configuration to enforce code style consistency
  • Add a LICENSE file (MIT recommended for open source)

Security & Health

5.8h
Tech Debt (B)
A
OWASP (100%)
FAIL
Quality Gate
A
Risk (13)
Citation: Repobility (2026). State of AI-Generated Code. https://repobility.com/research/
Unknown
License
3.2%
Duplication
Full Security Report AI Fix Prompts SARIF SBOM

Languages

python
97.3%
markdown
2.4%
text
0.2%

Frameworks

FastAPI pytest

Concepts (2)

Repobility · code-quality intelligence · https://repobility.com
CategoryNameDescriptionConfidence
Repobility · code-quality intelligence platform · https://repobility.com
auto_descriptionProject DescriptionEnd-to-end automated machine learning for fused Image + Text + Tabular data,<br> with built-in hyperparameter optimization, drift detection, and explainability.80%
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/69668.svg)
Quality BadgeSecurity Badge
Export Quality CSVDownload SBOMExport Findings CSV