Commit 957896bc9780a73241cd1ceabcb35ccbfa9f6e9c

Authored by quillotm
1 parent ed89325d5d
Exists in master

Adding an example of data and script to use clustering.

Showing 4 changed files with 156 additions and 0 deletions Inline Diff

receipts/clustering_example.sh
File was created 1 # Don't forget to install volia before or to run it in volia root directory.
2 # You can uncomment "debug" flag parameter to see print dedicated to the debug.
3
4 # Example k-means with (default) euclidian distance
5 python -m volia.clustering \
6 --features "example/feats_example.txt" \
7 --lst "example/example.lst" \
8 -k 4 \
9 --output "tests/kmeans_on_example.pkl" \
10 #--debug
11
12 python -m volia.clustering measure \
13 --measure "entropy" "v-measure" "homogeneity" "completeness" "purity" \
14 --features "example/feats_example.txt" \
15 --lst "example/example.lst" \
16 --truelabels "example/utt2grp_example" \
17 --model "tests/kmeans_on_example.pkl" \
18 --modeltype "k-means"
19
20
21 # Example k-means with mahalanobis distance
22 python -m volia.clustering kmeans \
23 --features "example/feats_example.txt" \
24 --lst "example/example.lst" \
25 -k 4 \
26 --output "tests/kmeans_mahalanobis_on_example.pkl" \
27 --mahalanobis
28 #-- debug
29
30 python -m volia.clustering measure \
31 --measure "entropy" "v-measure" "homogeneity" "completeness" "purity" \
32 --features "example/feats_example.txt" \
33 --lst "example/example.lst" \
34 --truelabels "example/utt2grp_example" \
35 --model "kmeans_mahalanobis_on_example.pkl" \
36 --modeltype "k-means-mahalanobis"
receipts/example/example.lst
File was created 1 00001
2 00002
3 00003
4 00004
5 00005
6 00006
7 00007
8 00008
9 00009
10 00010
11 00011
12 00012
13 00013
14 00014
15 00015
16 00016
17 00017
18 00018
19 00019
20 00020
21 00021
22 00022
23 00023
24 00024
25 00025
26 00026
27 00027
28 00028
29 00029
30 00030
31 00031
32 00032
33 00033
34 00034
35 00035
36 00036
37 00037
38 00038
39 00039
40 00040
41
receipts/example/feats_example.txt
File was created 1 00001 9.9777948745 5.0097253094
2 00002 9.5229160451 5.1065689857
3 00003 10.0683032257 5.0929926845
4 00004 10.4190537483 5.0436588347
5 00005 10.2489220003 4.764469118
6 00006 10.0102482307 5.1295142096
7 00007 10.0132105155 4.975514489
8 00008 9.6582134421 5.3718089569
9 00009 10.3687320788 5.0070660353
10 00010 10.0206559865 5.0782852429
11 00011 5.3790591998 10.5482455864
12 00012 4.8410845425 9.393727279
13 00013 4.7244926286 9.7959140641
14 00014 4.8898252527 9.8588087418
15 00015 4.9001782344 9.8886957353
16 00016 4.7953159294 10.1760344582
17 00017 4.987703639 10.2005482342
18 00018 5.4247526431 9.4028065528
19 00019 5.1492437327 9.6273676282
20 00020 5.1409374793 9.8161364221
21 00021 10.0364010754 10.033142897
22 00022 9.9678777136 10.3401433342
23 00023 9.9490251639 10.048815697
24 00024 10.2081153569 9.7356741508
25 00025 9.9570713519 10.3380561866
26 00026 10.2848058713 10.2612123135
27 00027 10.0367532914 9.881463012
28 00028 9.9394777833 9.8033954281
29 00029 10.3024249623 10.2459711714
30 00030 9.8460800216 9.9275337768
31 00031 0.0411836404 5.4633030866
32 00032 0.0139924297 4.5859670157
33 00033 -0.4310357985 5.1511811554
34 00034 -0.3548044058 4.9971822733
35 00035 -0.2065391374 4.6706733169
36 00036 0.3792866979 5.1858242378
37 00037 0.0223456313 5.0171648356
38 00038 0.4665682914 4.8223001752
39 00039 0.2108448666 5.0367557812
40 00040 -0.3157195455 5.4294636419
receipts/example/utt2grp_example
File was created 1 00001 1
2 00002 1
3 00003 1
4 00004 1
5 00005 1
6 00006 1
7 00007 1
8 00008 1
9 00009 1
10 00010 1
11 00011 2
12 00012 2
13 00013 2
14 00014 2
15 00015 2
16 00016 2
17 00017 2
18 00018 2
19 00019 2
20 00020 2
21 00021 3
22 00022 3
23 00023 3
24 00024 3
25 00025 3
26 00026 3
27 00027 3
28 00028 3
29 00029 3
30 00030 3
31 00031 4
32 00032 4
33 00033 4
34 00034 4
35 00035 4
36 00036 4
37 00037 4
38 00038 4
39 00039 4
40 00040 4
41