Open database for scientific gaze data.

Akseli Palén Infant Cognition Laboratory University of Tampere

$ virtualenv gazelledb
$ cd gazelledb
$ source bin/activate
(env)$ pip install flask
(env)$ pip install pymongo
(env)$ pip install numpy
(env)$ pip install coloredlogs

First, start MongoDB database server:

$ cd project/root/path
$ mongod --config data/mongodb.conf

Then, activate the virtual Python environment and start the web server:

$ source bin/activate
(env)$ python gazelledb/server.py

If you want to shut down the web server, close it and exit the environment:

[press ctrl-c]
(env)$ deactivate

Optionally you can also open Mongo console:

$ mongo

Trial gaze points

{ 'name': name_root + '/', 'timestamp': 123456789, 'format': 'gazelle/v1/trial', 'meta': { 'date': gazedata.get_trial_date(trial), 'participant_id': meta['participant_id'], 'method_version': meta['method_version'], 'participant_age_months': meta['participant_age_months'], 'calibration_successful': meta['calibration_successful'], 'trial_configuration_id': meta['trial_configuration_id'], 'trial_number': trial_num, 'aoi_x_rel': aoi_xy[0], 'aoi_y_rel': aoi_xy[1], 'source_file': file, 'tags': { 'target': { 'range': [100, -1] 'meta': { ... } }, 'first_and_second': { 'include': [0, 1] }, 'last': { 'include': [-1] }, 'saccade': { 'range': [678, 789] } } }, 'function': [], 'input': [], 'input_timestamp': [], 'output': norm_g }

Trial analysis

{ 'name': icl/trial/analysis, 'timestamp': 1233456789, 'format': 'gazelle/v1/saccade', // Description 'meta': {}, 'function': ['analysis.py'], 'input': string or list or pattern ['icl/name@123456789', '...@...', 'name@time'] 'input_timestamp': { // Check if newer 'icl/name@123456789': 123456789, 'icl/name2': 123456788 }// or list, keys are in the input 'output': a list in format gazelle/v1/saccade }

Trial analysis merging

{ 'name': 'icl/sequence', 'timestamp': 1233456789, 'format': 'gazelle/v1/saccades', 'meta': {}, 'function': 'merge-analysis.py', 'input': ['icl/trial', 'icl/trial2'], 'output': list in format gazelle/v1/saccades }

Prediction analysis

{ 'name': ..., 'timestamp': 11111111111, 'format': 'gazelle/v1/prediction', 'meta': {}, 'function': 'prediction-analysis.py', 'input': ['icl/sequence'], 'input_timestamp': [1234567678], 'output': { 'probabilities': ... how they affect reaction time, LATER model } }