Make plots for understanding the influence of mtry and L on performance

[tyler-tomita]

For benchmark datasets, plot:

1. Ensemble OOB error vs L (avg nonzeros per projection) and mtry, where line width indicates value of mtry.
2. Individual tree OOB errors vs individual tree avg projection density for different values of mtry, with mtry coded by color, for different values of L with each panel corresponding to a particular value of L
3. Tree diversity (variance of OOB error and some other metric) vs mtry and L
4. Tree strength (average of individual tree OOB errors) vs mtry and L

[jiandanjinxin]

hi, ttomita, when try to run the code on your example.pdf . I have set the path to the matlab. The proble accurs that 错误使用 parpool (line 111) Failed to start a parallel pool. (For information in addition to the causing error, validate the profile 'local' in the Cluster Profile Manager.)

出错 rpclassificationforest/growTrees (line 140) parpool('local',NWorkers,'IdleTimeout',360);

出错 rpclassificationforest (line 81) forest = growTrees(forest,nTrees,X,Y,varargin{:});

原因: 错误使用 parallel.internal.pool.InteractiveClient/start (line 329) Failed to start pool. 错误使用 parallel.Job/submit (line 304) 所有维度参数必须大于零

can you tell me how to solve this problem ? Thanks advanced!

[jovo]

Do you have the parallel toolbox?

Tyler - lets update the trade to describe the dependencies?

On Friday, March 25, 2016, jiandanjinxin [email protected] wrote:

hi, ttomita, when try to run the code on your example.pdf . I have set the path to the matlab. The proble accurs that 错误使用 parpool (line 111) Failed to start a parallel pool. (For information in addition to the causing error, validate the profile 'local' in the Cluster Profile Manager.)

出错 rpclassificationforest/growTrees (line 140) parpool('local',NWorkers,'IdleTimeout',360);

出错 rpclassificationforest (line 81) forest = growTrees(forest,nTrees,X,Y,varargin{:});

原因: 错误使用 parallel.internal.pool.InteractiveClient/start (line 329) Failed to start pool. 错误使用 parallel.Job/submit (line 304) 所有维度参数必须大于零

can you tell me how to solve this problem ? Thanks advanced!

— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/ttomita/RandomerForest/issues/79#issuecomment-201179099

the glass is all full: half water, half air. neurodata.io