Optimize clustering, add Whisper, CI workflow

.github/workflows/tests.yml

@@ -0,0 +1,66 @@
+name: Tests
+
+on:
+  push:
+    branches: [ master, main, optimize-quail ]
+  pull_request:
+    branches: [ master, main ]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ['3.9', '3.10', '3.11', '3.12']
+      fail-fast: false
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v5
+      with:
+        python-version: ${{ matrix.python-version }}
+
+    - name: Install system dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y ffmpeg
+
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install pytest pytest-cov
+        pip install torch --index-url https://download.pytorch.org/whl/cpu
+        pip install openai-whisper
+        pip install -e .
+
+    - name: Run tests
+      run: |
+        pytest tests/ -v --tb=short
+
+    - name: Run tests with coverage
+      run: |
+        pytest tests/ --cov=quail --cov-report=xml --cov-report=term-missing
+
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Set up Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: '3.11'
+
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install flake8
+
+    - name: Lint with flake8
+      run: |
+        # Stop the build if there are Python syntax errors or undefined names
+        flake8 quail --count --select=E9,F63,F7,F82 --show-source --statistics
+        # Exit-zero treats all errors as warnings
+        flake8 quail --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics

.gitignore

@@ -12,3 +12,4 @@ dist/*
 venv/
 docs/_build
 docs/sg_execution_times.rst
+CLAUDE.md

.readthedocs.yaml

@@ -0,0 +1,18 @@
+# Read the Docs configuration file for Sphinx projects
+# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
+
+version: 2
+
+build:
+  os: ubuntu-22.04
+  tools:
+    python: "3.11"
+
+sphinx:
+  configuration: docs/conf.py
+
+python:
+  install:
+    - requirements: docs/doc_requirements.txt
+    - method: pip
+      path: .

README.md

@@ -1,5 +1,7 @@
 [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.1003184.svg)](https://doi.org/10.5281/zenodo.1003184)
 [![JOSS](http://joss.theoj.org/papers/3fb5123eb2538e06f6a25ded0a088b73/status.svg)](http://joss.theoj.org/papers/10.21105/joss.00424)
+[![Tests](https://github.com/ContextLab/quail/actions/workflows/tests.yml/badge.svg)](https://github.com/ContextLab/quail/actions/workflows/tests.yml)
+[![Documentation Status](https://readthedocs.org/projects/cdl-quail/badge/?version=latest)](https://cdl-quail.readthedocs.io/en/latest/?badge=latest)
 
 ![Quail logo](images/Quail_Logo_small.png)
 
@@ -12,7 +14,7 @@ Quail is a Python package that facilitates analyses of behavioral data from memo
 - Clustering metrics (e.g. single-number summaries of how often participants transition from recalling a word to another related word, where "related" can be user-defined.)
 - Many nice plotting functions
 - Convenience functions for loading in data
-- Automatically parse speech data (audio files) using wrappers for the Google Cloud Speech to Text API
+- Automatically parse speech data (audio files) using OpenAI Whisper
 
 The intended user of this toolbox is a memory researcher who seeks an easy way to analyze and visualize data from free recall psychology experiments.
 

docs/doc_requirements.txt

@@ -1,18 +1,17 @@
-sphinx==1.5.5
-sphinx_bootstrap_theme==0.4.13
+sphinx>=4.0
+sphinx_bootstrap_theme
 sphinx-gallery
 numpydoc
 nbsphinx
-seaborn>=0.7.1
-matplotlib>=1.5.1
-scipy>=0.17.1
-numpy>=1.10.4
-pandas==0.18.1
-future
+seaborn>=0.12.0
+matplotlib>=3.5.0
+scipy>=1.10.0
+numpy>=2.0.0
+pandas>=2.0.0
+joblib>=1.3.0
 sqlalchemy
-dill
 requests
 pydub
-multiprocessing
 pathos
 jupyter_client
+ipykernel

quail/analysis/clustering.py

@@ -96,9 +96,8 @@ def _get_weight_exact(egg, feature, distdict, permute, n_perms):
     rec = list(egg.get_rec_items().values[0])
 
     if len(rec) <= 2:
-        warnings.warn('Not enough recalls to compute fingerprint, returning default'
-              'fingerprint.. (everything is .5)')
-        return 0.5
+        warnings.warn('Not enough recalls to compute fingerprint, returning NaN')
+        return np.nan
 
     distmat = get_distmat(egg, feature, distdict)
 
@@ -174,14 +173,11 @@ def _get_weight_best(egg, feature, distdict, permute, n_perms, distance):
 
     rec = list(egg.get_rec_items().values[0])
     if len(rec) <= 2:
-        warnings.warn('Not enough recalls to compute fingerprint, returning default'
-              'fingerprint.. (everything is .5)')
+        warnings.warn('Not enough recalls to compute fingerprint, returning NaN')
         return np.nan
 
     distmat = get_distmat(egg, feature, distdict)
     matchmat = get_match(egg, feature, distdict)
-    print(f"DEBUG: matchmat.shape={matchmat.shape}, len(rec)={len(rec)}")
-    print(f"DEBUG: distmat.shape={distmat.shape}")
 
     ranks = []
     for i in range(len(rec)-1):

quail/fingerprint.py

@@ -370,13 +370,6 @@ def compute_feature_stick(features, weights, alpha):
         return feature_stick
 
     def reorder_list(egg, feature_stick, dist_dict, tau):
-
-        def compute_stimulus_stick(s, tau):
-            '''create a 'stick' of feature weights'''
-
-            feature_stick = [[weights[feature]]*round(weights[feature]**alpha)*100 for feature in w]
-            return [item for sublist in feature_stick for item in sublist]
-
         # parse egg
         pres, rec, features, dist_funcs = parse_egg(egg)
 

quail/load.py

@@ -387,7 +387,7 @@ def getFeatures(stimDict):
 
     # add custom filters
     if filters:
-        filter_func = [adaptive_filter, experimeter_filter, experiments_filter] + filters
+        filter_func = [adaptive_filter, experimenter_filter, experiments_filter] + filters
     else:
         filter_func = [adaptive_filter, experimenter_filter, experiments_filter]
 

quail/simulate.py

@@ -10,9 +10,8 @@ def simulate_list(nwords=16, nrec=10, ncats=4):
     path = os.path.join(os.path.dirname(__file__), 'data/cut_wordpool.csv')
     wp = pd.read_csv(path)
 
-    # get one list
-    # logic seems to pick a group random
-    wp = wp[wp['GROUP']==np.random.choice(list(range(16)), 1)[0]].sample(16)
+    # get one list - pick a random group (groups are 1-16)
+    wp = wp[wp['GROUP']==np.random.choice(list(range(1, 17)), 1)[0]].sample(16)
 
     wp['COLOR'] = [[int(np.random.rand() * 255) for i in range(3)] for i in range(16)]
 

requirements.txt

@@ -4,3 +4,4 @@ matplotlib>=3.5.0
 seaborn>=0.12.0
 pandas>=2.0.0
 joblib>=1.3.0
+openai-whisper

setup.py

@@ -1,5 +1,6 @@
 # -*- coding: utf-8 -*-
 
+import os
 from setuptools import setup, find_packages
 
 DESCRIPTION = 'A python toolbox for analyzing and plotting free recall data'
@@ -12,11 +13,15 @@
 - Clustering metrics (e.g. single-number summaries of how often participants transition from recalling a word to another related word, where "related" can be user-defined.)
 - Many nice plotting functions
 - Convenience functions for loading in data
-- Automatically parse speech data (audio files) using wrappers for the Google Cloud Speech to Text API
+- Automatically parse speech data (audio files) using OpenAI Whisper
 
 The intended user of this toolbox is a memory researcher who seeks an easy way to analyze and visualize data from free recall psychology experiments.
 """
 
+# Read requirements from requirements.txt
+here = os.path.abspath(os.path.dirname(__file__))
+with open(os.path.join(here, 'requirements.txt')) as f:
+    requirements = [line.strip() for line in f if line.strip() and not line.startswith('#')]
 
 EXTRAS_REQUIRE={
     'speech-decoding': ["pydub", "openai-whisper"],
@@ -35,13 +40,6 @@
     license='MIT',
     packages=find_packages(exclude=('tests', 'docs', 'paper')),
     include_package_data=True,
-    install_requires=[
-        'numpy>=2.0.0',
-        'scipy>=1.10.0',
-        'matplotlib>=3.5.0',
-        'seaborn>=0.12.0',
-        'pandas>=2.0.0',
-        'joblib>=1.3.0',
-    ],
+    install_requires=requirements,
     extras_require=EXTRAS_REQUIRE,
 )