How pritamdeka/sbert2s1 works
sbert2s1 packages a public typed-decision inference model and the research pipeline used to build, train, calibrate, evaluate, and analyse it. The user workflow is to load an exported encoder model, submit text plus typed questions, and receive calibrated option probabilities. Research additionally builds biomedical datasets, runs encoder and generative-LLM experiments, and produces paper analyses. The graph follows documented capabilities and indexed references; unsampled implementation details are not treated as verified calls.
Main entry points: train.py Training Orchestrator · evaluate.py Prediction Evaluation · data.py Tokenized Batches · llm_baseline.py Generative Baseline
Components 26
Public Inference
- Typed Request
- S1 Predictor
inference/sbert2s1.py
Biomedical Data
- Public Task Builder
research/data_build/build_public.py - Clinical Builder
research/data_build/build_clinical.py - MEDLINE Builder
research/data_build/build_medline.py - Abstract Hydration
research/data_build/hydrate_medline.py - Decision Schema
research/s1/schema.py - Prepared Splits
Research Runtime
- Tokenized Batches
research/s1/data.py - Decision Architectures
research/s1/model.py - Model Resolution
research/s1/models_io.py - Training Objectives
research/s1/objectives.py - Training Orchestrator
research/s1/train.py - Temperature Fitting
research/s1/calibrate.py - Temperature Lookup
research/s1/temperature_lookup.py - Prediction Evaluation
research/s1/evaluate.py - Decision Metrics
research/s1/metrics.py
Experiments And Analysis
- Experiment Runner
research/experiments.py - Generative Baseline
research/s1/llm_baseline.py - Data Audit
research/analysis/audit_data.py - Run Summaries
research/analysis/summarize.py - Paper Analysis
research/paper/scripts/analysis_common.py - Confidence Cascade
research/paper/scripts/cascade_v2.py - Results Tables
research/paper/scripts/results_tables.py
Other components
- Caller
- Model Hub
Connections 34
- Caller → Typed Request: submits
README.md - Typed Request → S1 Predictor: predicts
README.md - S1 Predictor → Caller: returns probabilities
README.md - Model Hub → S1 Predictor: provides model
README.md - Public Task Builder → Decision Schema: formats records
research/data_build/build_public.py - Clinical Builder → Decision Schema: formats records
research/data_build/build_clinical.py - MEDLINE Builder → Decision Schema: formats records
research/data_build/build_medline.py - Abstract Hydration → MEDLINE Builder: hydrates abstracts
README.md - Public Task Builder → Prepared Splits: writes splits
research/data_build/build_public.py - Clinical Builder → Prepared Splits: writes splits
README.md - MEDLINE Builder → Prepared Splits: writes splits
README.md - Abstract Hydration → Prepared Splits: adds abstracts
README.md - Prepared Splits → Tokenized Batches: tokenizes
research/s1/data.py - Tokenized Batches → Decision Schema: uses record format
research/s1/data.py - Tokenized Batches → Model Resolution: resolves tokenizer
research/s1/data.py - Training Orchestrator → Tokenized Batches: assembles batches
research/s1/train.py - Training Orchestrator → Decision Architectures: builds and trains
research/s1/train.py - Training Orchestrator → Training Objectives: optimizes with
research/s1/train.py - Training Orchestrator → Temperature Fitting: fits temperatures
research/s1/train.py - Training Orchestrator → Prediction Evaluation: evaluates model
research/s1/train.py - Training Orchestrator → Model Resolution: resolves encoder
research/s1/train.py - Temperature Fitting → Temperature Lookup: exports lookup
research/s1/calibrate.py - Prediction Evaluation → Tokenized Batches: collates batches
research/s1/evaluate.py - Prediction Evaluation → Temperature Fitting: uses calibration
research/s1/evaluate.py - Prediction Evaluation → Decision Metrics: computes metrics
research/s1/evaluate.py - Experiment Runner → Training Orchestrator: launches runs
README.md - Experiment Runner → Generative Baseline: launches baseline
README.md - Generative Baseline → Model Resolution: resolves model
research/s1/llm_baseline.py - Generative Baseline → Temperature Fitting: fits temperatures
research/s1/llm_baseline.py - Generative Baseline → Prediction Evaluation: reuses evaluation
research/s1/llm_baseline.py - Data Audit → Paper Analysis: supplies audit data
research/paper/scripts/analysis_common.py - Paper Analysis → Temperature Lookup: applies temperatures
research/paper/scripts/analysis_common.py - Confidence Cascade → Paper Analysis: reads predictions
research/paper/scripts/cascade_v2.py - Results Tables → Run Summaries: summarizes results
README.md