maheshc4742/ss

How maheshc4742/ss works

ScriptSense evaluates handwritten or JSON student answers, returning rubric-based marks, confidence, and feedback. The principal workflow runs from the web dashboard through evaluation endpoints and an orchestrator that optionally extracts answers with vision OCR, scores semantic alignment, applies rubric grading, and either saves a report or stages answers for human review. Approved annotations feed a threshold-gated, versioned fine-tuning pipeline; model management supports checkpoint activation. The README documents major capabilities not fully visible in excerpts, so those relationships are cited there; other edges use indexed source references.

Main entry points: evaluator.py Evaluation Orchestrator · hitl_service.py HITL Workflow · app.js Web Dashboard · main.py FastAPI Application

Components 25

Web Experience

Evaluate API backend/api/evaluate.py
Web Dashboard frontend/js/app.js

Evaluation

Evaluation Orchestrator backend/services/evaluator.py
Rubric Grading backend/services/grading_module.py
Semantic Scoring backend/services/semantic_module.py
OCR Provider Factory backend/ocr/ocr_factory.py
OCR Interface backend/ocr/base_ocr.py
Gemini OCR backend/ocr/gemini_ocr.py
Qwen OCR backend/ocr/qwen_ocr.py
PDF Rendering backend/ocr/pdf_utils.py

Human Review

Review API backend/api/hitl.py
HITL Workflow backend/services/hitl_service.py

Model Improvement

Training API backend/api/training.py
Model API backend/api/models_api.py
Verified Annotations data/verified_training.jsonl
Dataset Preparation backend/training/prepare_dataset.py
Fine-Tuning backend/training/train.py
Versioned Checkpoints models

Platform

Configuration API backend/api/config_api.py
Application Database backend/core/database.py
Evaluation Records backend/models/database_models.py
Runtime Settings backend/core/config.py
Evaluation Reports reports
FastAPI Application backend/main.py

Other components

Human Reviewer

Connections 38

Web Dashboard → Evaluate API: submits evaluations README.md
Web Dashboard → Review API: reviews answers README.md
Web Dashboard → Training API: triggers training README.md
Web Dashboard → Model API: activates models README.md
Web Dashboard → Configuration API: configures settings README.md
FastAPI Application → Evaluate API: registers routes backend/main.py
FastAPI Application → Review API: registers routes backend/main.py
FastAPI Application → Training API: registers routes backend/main.py
FastAPI Application → Model API: registers routes backend/main.py
FastAPI Application → Configuration API: registers routes backend/main.py
Evaluate API → Evaluation Orchestrator: invokes evaluation backend/api/evaluate.py
Evaluation Orchestrator → OCR Provider Factory: selects provider backend/services/evaluator.py
OCR Provider Factory → Gemini OCR: creates provider backend/ocr/ocr_factory.py
OCR Provider Factory → Qwen OCR: creates provider backend/ocr/ocr_factory.py
Gemini OCR → OCR Interface: implements interface backend/ocr/gemini_ocr.py
Qwen OCR → OCR Interface: implements interface backend/ocr/qwen_ocr.py
Gemini OCR → PDF Rendering: renders PDF pages backend/ocr/gemini_ocr.py
Qwen OCR → PDF Rendering: renders PDF pages backend/ocr/qwen_ocr.py
Evaluation Orchestrator → Semantic Scoring: scores alignment backend/services/evaluator.py
Evaluation Orchestrator → Rubric Grading: grades rubric backend/services/evaluator.py
Evaluation Orchestrator → HITL Workflow: stages review backend/services/evaluator.py
Evaluation Orchestrator → Evaluation Reports: writes report README.md
Evaluation Orchestrator → Application Database: records evaluation backend/api/evaluate.py
Review API → Evaluation Orchestrator: starts train evaluation backend/api/hitl.py
Review API → HITL Workflow: handles review actions backend/api/hitl.py
Human Reviewer → Review API: submits decisions README.md
HITL Workflow → Evaluation Records: updates review records backend/services/hitl_service.py
HITL Workflow → Rubric Grading: quantizes marks backend/services/hitl_service.py
HITL Workflow → Verified Annotations: appends approvals README.md
HITL Workflow → Application Database: persists review state README.md
Application Database → Evaluation Records: creates tables backend/core/database.py
Training API → Fine-Tuning: triggers fine-tuning backend/api/training.py
Fine-Tuning → Dataset Preparation: prepares splits backend/training/train.py
Dataset Preparation → Verified Annotations: loads annotations README.md
Fine-Tuning → Versioned Checkpoints: saves versions README.md
Model API → Evaluation Records: manages registry backend/api/models_api.py
Model API → Runtime Settings: updates active model README.md
Configuration API → Runtime Settings: reads settings backend/api/config_api.py

Drawn by GitDiagram on September 30, 2026 from the repository's file tree, README and source files. Read as Markdown ·