Started===== MAIN: evaluate performance =====
=== START program3: ./run interact ../program1/run ../dataset2/raw ../program1/predictions
StartedWARNING: path already exists, overwriting: '../program1/predictions'
=== END program3: ./run interact ../program1/run ../dataset2/raw ../program1/predictions --- OK [14s]
===== MAIN: evaluate performance =====
=== START program4: ./run evaluate ../dataset2/raw ../program1/predictions
=== END program4: ./run evaluate ../dataset2/raw ../program1/predictions --- OK [0s]
real 0m14.294s
user 0m5.348s
sys 0m0.064s
Run specification
interactive-learning: Main entry for interactive learning tasks (online learning), where is two phases.
(interactivelearner:Program) general-perceptron: General perceptron implementation with standard perceptron loss function.
(dataset:Dataset) NY Traffic R11 2008 Weekday/Weekend popular roads normalized: Classes are 1 for Mon-Fri, 2 for Sat-Sun. Each feature vector is a combination of the readings for an entire day. This is 1 years worth of data. Each feature index uniquely identifies a combination of (road, direction, and hour) where the hour can also be a combination of all hours in that day. Feature values are the counts of cars sensed at that road/direction/hour divided by the average counts for that road/direction/hour across all days, or the sums of counts of cars across all hours of a day sensed at that road/direction divided by the number of readings for that day and by the average value of that across all days Only road/direction/hour tuples with at least 350 datapoints were included. Only road/direction tuples with at least 350*20 datapoints were included. This is all because the consistency of readings varies a lot.
(interactor:Program[Interact]) online-multiclass-utils: Inspects OnlineLearningMulticlass datasets and evaluates OnlineLearningMulticlass performance.
(evaluator:Program[Evaluate]) online-multiclass-utils: Inspects OnlineLearningMulticlass datasets and evaluates OnlineLearningMulticlass performance.
When you generate a run, you can set a time limit for the run (no more than 24 hours). After that point, we will terminate the program.
Your program can use 1.5GB of memory. More information here.
Go to the page for the run and look at the log file for signs of the responsible error.
You can also download the run and run it locally on your machine (a README file should
be included in the download which provides more information).
We said that a run was simply a program/dataset pair, but that's not the full story.
A run actually includes other helper programs such as the evaluation program and
various programs for reductions (e.g., one-versus-all, hyperparameter tuning).
More formally, a run is a given by a run specification,
which can be found on the page for any run.
A run specification is a tree where each internal node represents a program
and its children represents the arguments to be passed into its constructor.
For example, the one-versus-all program takes your binary classification program
as a constructor argument and behaves like a multiclass classification program.
Must be logged in to post comments.