Skip to content

File conversion — Python tools UI (web-based)

Browser-based converters — no installation required to use the hosted versions below. Each conversion tool is its own standalone app.

Live apps

App URL File
Convert .mat → NumPy mds-app-matnpy.streamlit.app app_matnpy.py
Convert NumPy → .mat appnumpytomatpy-dcahmfdqvuqdzyyza3ddyy.streamlit.app app_numpy_to_mat.py
OOO → Triadic ooo-triadic.streamlit.app app_ooo.py

Note: apps may be asleep if inactive — click the link and wait a moment for them to wake up. The hosted versions run on limited, shared resources — best for small datasets and a quick look, not a final fit.

Running locally

Faster than the hosted versions, and not limited by their shared resources:

cd rs-software
pip install streamlit plotly pandas matplotlib scikit-learn
streamlit run app_matnpy.py       # or app_numpy_to_mat.py, app_ooo.py

Convert .mat → NumPy

Convert a .mat file to a NumPy array. Works with either a choices file (raw pairwise judgments) or a coordinates file (the output of a model fit) — the tool detects which one you uploaded automatically, so you don't need to tell it which kind of file it is.

Choices file → output columns: ref, s1, s2, N(s1 chosen), N_repeats (1-indexed)

Coordinates file → output: an (n_stimuli, dim) array of the fitted coordinates. If the file has more than one fitted dimensionality (e.g. both a 2D and a 3D fit saved together), you'll be asked which one to export.

  1. Upload a .mat file (choices or coordinates)
  2. Preview the array in the browser
  3. Download as .npy

Convert NumPy → .mat

Convert a 5-column NumPy choice array back into a .mat choice file — the reverse of the tool above.

Input columns: ref, s1, s2, N(s1 chosen), N_repeats (1-indexed)

  1. Upload a .npy choice array
  2. Optionally provide stimulus names (typed, uploaded as a text file, or skipped — falls back to generic names like stim_01)
  3. Preview the array in the browser
  4. Download as .mat

OOO → Triadic

Convert an odd-one-out .mat file to standard triadic choice format.

Each OOO judgment generates exactly 2 triadic entries: if X is chosen as odd one out from (X, Y, Z), this yields ref=Y, chosen=Z and ref=Z, chosen=Y.

Input format: s1, s2, s3, N(s1 odd out), N(s2 odd out), N(s3 odd out)

Output format: ref, s1, s2, N(s1 chosen), N_repeats

  1. Upload an OOO .mat file
  2. Preview the converted triadic table
  3. Download as .mat

The output file can be used directly in the Comparing RS or Choice → Coordinates tools.

Choice → Coordinates has moved

Fitting coordinates from a choice file is now documented under Creating RS → Choice → Coordinates, alongside the other ways to create a representational space from perceptual judgments.