---
title: SPACEMIND
colorFrom: red
colorTo: indigo
sdk: gradio
sdk_version: 6.29.1
python_version: '3.12'
app_file: app.py
pinned: false
short_description: Behavioural health signals in astronaut oral histories
---

# SPACEMIND


Paste or upload an astronaut oral history transcript and the app runs my pipeline step by step: word count, cleaning, removing the interviewer, segmenting, then all three of my models (VADER baseline, TF-IDF + logistic regression, BERT + LoRA) and a SHAP explanation for one segment. 

Files: 
`app.py` (everything)
my LoRA adapter (`adapter_config.json`,`adapter_model.safetensors`)
the tokenizer (`tokenizer.json`, `tokenizer_config.json`)
the TF-IDF models (`tfidf_vectorizer.pkl`, `tfidf_logistic_models.pkl`) 

Research prototype: it flags passages for a person to read and does not diagnose anyone.