Commit 83a6d599 authored by Shengpu Tang (tangsp)'s avatar Shengpu Tang (tangsp)
Browse files

init experiments

parent e364178c
## Overview
- Requires: the raw csv files of [MIMIC-III database](
- Extract and format data from structured tables in MIMIC-III as input to FIDDLE
- Goal: using data from all tables, generate Time-Invariant features **s** and Time Series features __X__.
## Usage
0. Modify `config.yaml` to specify `mimic3_path` and `data_path`.
1. Data Extraction
- Execute `python -c "from extract_data import *; check_nrows();"` to verify the integrity of raw csv files.
- Run `python`.
1. Labels & Analyses
- Run `python` to generate the event onset time and labels for two outcomes: ARF and shock.
- Run the following notebooks in order: `LabelDistribution.ipynb`, `InclusionExclusion.ipynb` and `PopulationSummary.ipynb`.
1. Generate features
- Run `python --task={task} --T={T} --dt={dt}`
- Run `python --task={task} --T={T} --dt={dt}`
Note: a bash script is provided for generating features.
The generated features and associated metadata are located in `{data_path}/features/task={task}.T={T}.dt={dt}/`:
- `s.npz`: a sparse array of shape (N, d)
- `X.npz`: a sparse tensor of shape (N, L, D)
- `s.feature_names.txt`: names of _d_ time-invariant features
- `X.feature_names.txt`: names of _D_ time-series features
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment