You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Direction 1: In the ConfigDialog class, the methods ConfigDialog and createCenterPanel() both call the static methods checkPythonEnvironment(), checkEyeTracker(), getEyeTrackerName(), getFrequencies() from the AvailabilityChecker class. Each of these methods invokes a new Python process, which is time-consuming and causes the IDE to freeze. Can we combine these methods into one to invoke the Python process just once? I understand that we use print() in Python to transmit results to Java, and there might be annoying error messages. One solution could be to use the try...catch syntax to filter them.
Direction 2: We use round(time.time() * 1000) to get the timestamp for each gaze point. However, possibly due to Python's execution mechanism, there are duplicate timestamps at high frequencies. Although they don't severely impact the results, avoiding them is preferable. Interpreting the timestamps directly from the eye-tracking device, i.e., gaze_data['device_time_stamp'] and gaze_data['system_time_stamp'], is challenging. I don't fully understand their relationship with the Unix timestamp.
Direction 1: In the
ConfigDialogclass, the methodsConfigDialogandcreateCenterPanel()both call the static methodscheckPythonEnvironment(),checkEyeTracker(),getEyeTrackerName(),getFrequencies()from theAvailabilityCheckerclass. Each of these methods invokes a new Python process, which is time-consuming and causes the IDE to freeze. Can we combine these methods into one to invoke the Python process just once? I understand that we useprint()in Python to transmit results to Java, and there might be annoying error messages. One solution could be to use thetry...catchsyntax to filter them.Direction 2: We use
round(time.time() * 1000)to get the timestamp for each gaze point. However, possibly due to Python's execution mechanism, there are duplicate timestamps at high frequencies. Although they don't severely impact the results, avoiding them is preferable. Interpreting the timestamps directly from the eye-tracking device, i.e.,gaze_data['device_time_stamp']andgaze_data['system_time_stamp'], is challenging. I don't fully understand their relationship with the Unix timestamp.