MLcomp version alpha Statistics: 2043 users, 665 datasets, 389 programs, 18920 runs (0 queued, 4 running), 0 workers

movielens1m

1M MovieLens movie ratings dataset from http://www.grouplens.org/. The original dataset contains 1,000,209 anonymous ratings (1-5 stars) of approximately 3,900 movies made by 6,040 MovieLens users who joined MovieLens in 2000. This sub-dataset includes the ratings of 5,000 randomly selected users. A single rating from each user was withheld to form the test set. See included README.txt for more information.

Format:	CollaborativeFiltering
User:	lester
Size:	32M
Status:	processed
Access:	restricted_access
#train	830307
#test	5000
#rows	5000
#cols	3952
minY	1
maxY	5

Run a program on this dataset

Select a program CollaborativeFiltering 1-25 of 110 < > Filter:

	Name	Created	User	Rating <<	#runs
run	jjariyas-svdj-r	3y191d ago	jjariyas	68	17
run	MyMediaLite-slopeone	1y320d ago	zenogantner	62	15
run	mjanderson-knn-r	3y191d ago	mjanderson	61	8
run	cberner-KNN-r	3y194d ago	cberner	61	8
run	xterm-knnsvd_shrunk_unbias_rev-r	3y187d ago	xterm	61	15
run	MyMediaLite-user-average	1y320d ago	zenogantner	61	8
run	MyMediaLite-biased-matrix-factorization-k-40	1y320d ago	zenogantner	60	12
run	MyMediaLite-user-item-baseline	1y319d ago	zenogantner	60	15
run	MyMediaLite-biased-matrix-factorization-k-10	1y320d ago	zenogantner	60	12
run	MyMediaLite-matrix-factorization-k-10	1y320d ago	zenogantner	60	19
run	MyMediaLite-matrix-factorization-k-20	1y320d ago	zenogantner	60	19
run	MyMediaLite-matrix-factorization-k-5	1y320d ago	zenogantner	59	19
run	MyMediaLite-biased-matrix-factorization-k-5	1y320d ago	zenogantner	59	12
run	akshayk-knn-cosine-distance-r	3y191d ago	akshayk	59	8
run	cberner-KNNImproved-r	3y192d ago	cberner	59	8
run	MyMediaLite-matrix-factorization-k-40	1y320d ago	zenogantner	59	19
run	simple-knn-with-tuning-and-shrunk-column	2y61d ago	yorinasub17	59	9
run	MyMediaLite-global-average	1y320d ago	zenogantner	59	11
run	kenghao-knn_k50-r	3y190d ago	kenghaochang	58	9
run	MyMediaLite-matrix-factorization-k-60	1y320d ago	zenogantner	58	19
run	lester-shrunk_item_mean-r	3y204d ago	lester	57	15
run	lester-shrunk_item_mean-testSunday	3y194d ago	jjariyas	57	15
run	MyMediaLite-biased-matrix-factorization-k-60	1y320d ago	zenogantner	57	12
run	MyMediaLite-biased-matrix-factorization-k-20	1y320d ago	zenogantner	56	13
run	whoburg-ALS_K1-r	3y190d ago	whoburg	56	8

1. pagination_page => 0
2. include => uservresult
3. reverse_sort => true
4. width => 100%
5. pagination_maxed => false
6. limit => 25
7. columns => prg_create_runprg_nameprg_created_atprg_userprg_avg_percentileprg_num_runs
8. model => Program
9. title => Select a program
10. joins =>
11. current_sort_col => avg_percentile
12. name => programs
13. paginate => true
14. show_footer => true
15. filters => is_helperfalseprocess_statussuccesstask_typeCollaborativeFiltering

Existing runs on movielens1m (all) 1-25 of 117 < > Filter:

ID	Program	Dataset	Tuned hyper.	User	Updated	Status	Total time	Memory	Error >>
Run #13964	MML-BMF	movielens1m	no	zenogantner	1y321d ago	done	6m24s	423M	0.884
Run #1215	xterm-knnsvd_shrunk_unbias-r	movielens1m	no	xterm	2y225d ago	done	4h16m	1463M	0.887
Run #2344	xterm-knnsvd_shrunk_unbias_rev-r	movielens1m	no	internal	2y225d ago	done	2h51m	1180M	0.887
Run #14853	MyMediaLite-biased-matrix-factorization-k-60	movielens1m	no	internal	1y311d ago	done	1h18m	423M	0.888
Run #14848	MyMediaLite-matrix-factorization-k-20	movielens1m	no	internal	1y311d ago	done	7m41s	423M	0.888
Run #14727	MyMediaLite-matrix-factorization-k-60	movielens1m	no	internal	1y311d ago	done	22m23s	423M	0.889
Run #14755	MyMediaLite-matrix-factorization-k-10	movielens1m	no	internal	1y311d ago	done	6m16s	423M	0.889
Run #14858	MyMediaLite-biased-matrix-factorization-k-10	movielens1m	no	internal	1y311d ago	done	28m9s	423M	0.890
Run #14868	MyMediaLite-biased-matrix-factorization-k-40	movielens1m	no	internal	1y311d ago	done	1h16m	423M	0.891
Run #14741	MyMediaLite-matrix-factorization-k-5	movielens1m	no	internal	1y311d ago	done	4m46s	417M	0.891
Run #14843	MyMediaLite-biased-matrix-factorization-k-20	movielens1m	no	internal	1y311d ago	done	38m40s	423M	0.892
Run #14863	MyMediaLite-biased-matrix-factorization-k-5	movielens1m	no	internal	1y311d ago	done	26m34s	423M	0.893
Run #14769	MyMediaLite-matrix-factorization-k-40	movielens1m	no	internal	1y311d ago	done	12m36s	423M	0.897
Run #14097	MyMediaLite-matrix-factorization-k-5	movielens1m	yes	zenogantner	1y320d ago	done	16m18s	430M	0.899
Run #14121	MyMediaLite-biased-matrix-factorization-k-5	movielens1m	yes	zenogantner	1y319d ago	done	1h14m	431M	0.904
Run #1336	cberner-KNNImproved-r	movielens1m	no	jake	2y225d ago	done	1h29m	833M	0.905
Run #1199	kenghao-matrix_em_k10_v2-r	movielens1m	no	kenghaochang	2y225d ago	done	2h15m	2021M	0.906
Run #919	mjanderson-matrix-factorization-r	movielens1m	no	mjanderson	2y225d ago	done	20m32s	570M	0.909
Run #14093	MyMediaLite-matrix-factorization-k-10	movielens1m	yes	zenogantner	1y320d ago	done	21m45s	430M	0.909
Run #14089	MyMediaLite-matrix-factorization-k-20	movielens1m	yes	zenogantner	1y320d ago	done	32m56s	432M	0.910
Run #1335	lester-mf_rsgd-r	movielens1m	no	jake	2y225d ago	done	34m14s	425M	0.913
Run #14125	MyMediaLite-biased-matrix-factorization-k-10	movielens1m	yes	zenogantner	1y319d ago	done	1h38m	432M	0.914
Run #14117	MyMediaLite-biased-matrix-factorization-k-40	movielens1m	yes	zenogantner	1y319d ago	done	4h46m	431M	0.916
Run #14081	MyMediaLite-matrix-factorization-k-60	movielens1m	yes	zenogantner	1y320d ago	done	1h3m	430M	0.918
Run #1174	kenghao-knn_k50-r	movielens1m	no	kenghaochang	2y225d ago	done	3h10m	667M	0.921

1. pagination_page => 0
2. reverse_sort => false
3. width => 100%
4. pagination_maxed => false
5. limit => 25
6. columns => run_idrun_core_programrun_core_datasetrun_hyperrun_userrun_updated_atrun_statusrun_timerun_memoryrun_error
7. model => Run
8. title => Existing runs on movielens1m
9. joins => statuscore_programcore_dataset
10. filter_field => programs.namedatasets.nameruns.idrun_statuses.status
11. current_sort_col => error
12. name => runs
13. show_footer => true
14. paginate => true
15. filters => core_dataset_id335programs.is_helperfalse

Processing details

• exitCode: 0
• success: true
• test:
  • maxLabelValue: 5.0
  • minLabelValue: 1.0
  • numCols: 3952
  • numExamples: 5000
  • numRows: 5000
  • success: true
  • time: 2
• train:
  • maxLabelValue: 5.0
  • minLabelValue: 1.0
  • numCols: 3952
  • numExamples: 830307
  • numRows: 5000
  • success: true
  • time: 5

Comments:

• 

  Oriol Vinyals (xterm)
  UC Berkeley
  Posted at 11/14/2009

  Uhmm after a lot of tweaking of my R code, I got my thing to run under 5 hours... 0.887 yay!
• 

  Jacob Abernethy (jake)
  UC Berkeley Comp Sci
  Posted at 11/13/2009

  Wow, that's close

  mjanderson got pretty darn close to cberner's lead
• 

  Jacob Abernethy (jake)
  UC Berkeley Comp Sci
  Posted at 11/12/2009

  Mean movie rating

  I would just like to point out how well using the mean movie rating works. How's that for learning :-)