dieKarotte commited on
Commit
dd39446
·
verified ·
1 Parent(s): 0032753

Add files using upload-large-folder tool

Browse files
Files changed (20) hide show
  1. __pycache__/train_beats_multilabel_trunk.cpython-311.pyc +0 -0
  2. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__hm3d__00034-6imZUJGRUq4__000000-foa__132991__pred.csv +165 -0
  3. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__hm3d__00034-6imZUJGRUq4__000096-foa__161195__gt.csv +2 -0
  4. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__hm3d__00056-kdw2Uapns3b__000020-foa__89741__pred.csv +201 -0
  5. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov2_000022__gt.csv +28 -0
  6. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov2_000031__gt.csv +68 -0
  7. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov2_000043__gt.csv +44 -0
  8. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov2_000129__gt.csv +8 -0
  9. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov3_000055__pred.csv +161 -0
  10. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov3_000066__pred.csv +201 -0
  11. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov3_000126__pred.csv +49 -0
  12. checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov3_000136__gt.csv +77 -0
  13. docs/README_DOCUMENTATION_INDEX.md +274 -0
  14. docs/train_valid_gap_diagram.txt +204 -0
  15. docs/v13e_spatial_beats_design.md +78 -0
  16. eval_voxaudio_ood_results/flow2gan/per_sample.json +2161 -0
  17. eval_voxaudio_ood_results/flow2gan/summary.json +25 -0
  18. eval_voxaudio_ood_results/stable_audio_vae/per_sample.json +2342 -0
  19. eval_voxaudio_ood_results/stable_audio_vae/summary.json +25 -0
  20. eval_voxaudio_ood_results/summary_all.json +127 -0
__pycache__/train_beats_multilabel_trunk.cpython-311.pyc ADDED
Binary file (53.6 kB). View file
 
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__hm3d__00034-6imZUJGRUq4__000000-foa__132991__pred.csv ADDED
@@ -0,0 +1,165 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,41,singing,128.476,-43.128,2.421,0.4824
3
+ 1,0.4041,0,41,singing,128.361,-43.859,2.45,0.7852
4
+ 2,0.8083,0,41,singing,128.274,-44.241,2.464,0.8867
5
+ 3,1.2124,0,41,singing,128.041,-44.48,2.464,0.9219
6
+ 4,1.6166,0,41,singing,128.073,-44.729,2.464,0.9414
7
+ 5,2.0207,0,41,singing,127.925,-44.976,2.464,0.9492
8
+ 6,2.4248,0,41,singing,127.66,-45.334,2.464,0.957
9
+ 7,2.829,0,41,singing,127.809,-45.462,2.464,0.9609
10
+ 8,3.2331,0,41,singing,127.546,-45.445,2.464,0.9648
11
+ 9,3.6372,0,41,singing,127.694,-45.572,2.464,0.9648
12
+ 10,4.0414,0,41,singing,127.432,-45.555,2.464,0.9648
13
+ 11,4.4455,0,41,singing,127.58,-45.681,2.464,0.9688
14
+ 12,4.8497,0,41,singing,127.58,-45.681,2.464,0.9688
15
+ 13,5.2538,0,41,singing,127.283,-45.795,2.464,0.9688
16
+ 14,5.6579,0,41,singing,127.283,-45.795,2.464,0.9688
17
+ 15,6.0621,0,41,singing,127.283,-45.795,2.464,0.9648
18
+ 16,6.4662,0,41,singing,127.283,-45.977,2.464,0.9648
19
+ 17,6.8704,0,41,singing,127.283,-45.977,2.464,0.9609
20
+ 18,7.2745,0,41,singing,127.02,-45.958,2.464,0.9609
21
+ 19,7.6786,0,41,singing,127.02,-45.958,2.478,0.9531
22
+ 20,8.0828,0,41,singing,127.057,-45.828,2.478,0.9492
23
+ 21,8.4869,0,41,singing,126.907,-46.064,2.478,0.9414
24
+ 22,8.8911,0,41,singing,126.907,-46.064,2.478,0.9336
25
+ 23,9.2952,0,41,singing,126.907,-46.064,2.478,0.9219
26
+ 24,9.6993,0,41,singing,126.757,-46.12,2.478,0.9062
27
+ 25,10.1035,0,41,singing,126.757,-45.94,2.478,0.8828
28
+ 26,10.5076,0,41,singing,126.757,-45.94,2.478,0.8555
29
+ 27,10.9117,0,41,singing,126.719,-46.071,2.464,0.7969
30
+ 28,11.3159,0,41,singing,126.681,-46.021,2.464,0.7344
31
+ 29,11.72,0,41,singing,126.681,-45.838,2.45,0.6367
32
+ 30,12.1242,0,41,singing,126.489,-45.843,2.45,0.4863
33
+ 31,12.5283,0,41,singing,126.409,-45.739,2.435,0.3555
34
+ 32,12.9324,0,41,singing,126.368,-45.496,2.421,0.252
35
+ 33,13.3366,0,41,singing,126.517,-45.543,2.407,0.1602
36
+ 34,13.7407,0,41,singing,126.275,-45.294,2.393,0.0981
37
+ 35,14.1449,0,41,singing,126.146,-45.118,2.364,0.0566
38
+ 36,14.549,0,41,singing,125.967,-44.873,2.336,0.0303
39
+ 37,14.9531,0,41,singing,125.828,-44.681,2.308,0.0175
40
+ 38,15.3573,0,41,singing,125.712,-44.431,2.266,0.0092
41
+ 39,15.7614,0,41,singing,125.64,-44.124,2.224,0.0051
42
+ 40,16.1656,0,41,singing,125.848,-43.885,2.196,0.0037
43
+ 0,0.0,1,41,singing,127.097,-42.292,2.421,0.5117
44
+ 1,0.4041,1,41,singing,126.87,-43.091,2.435,0.8047
45
+ 2,0.8083,1,41,singing,126.943,-43.419,2.45,0.9023
46
+ 3,1.2124,1,41,singing,126.578,-43.902,2.45,0.9375
47
+ 4,1.6166,1,41,singing,126.578,-44.091,2.45,0.9531
48
+ 5,2.0207,1,41,singing,126.469,-44.392,2.45,0.9648
49
+ 6,2.4248,1,41,singing,126.431,-44.705,2.45,0.9688
50
+ 7,2.829,1,41,singing,126.469,-44.577,2.45,0.9727
51
+ 8,3.2331,1,41,singing,126.215,-44.742,2.45,0.9766
52
+ 9,3.6372,1,41,singing,126.323,-44.816,2.45,0.9766
53
+ 10,4.0414,1,41,singing,126.215,-44.925,2.464,0.9766
54
+ 11,4.4455,1,41,singing,126.215,-44.925,2.464,0.9766
55
+ 12,4.8497,1,41,singing,126.361,-44.871,2.464,0.9766
56
+ 13,5.2538,1,41,singing,126.323,-45.181,2.464,0.9766
57
+ 14,5.6579,1,41,singing,126.175,-45.235,2.464,0.9766
58
+ 15,6.0621,1,41,singing,126.283,-45.308,2.464,0.9766
59
+ 16,6.4662,1,41,singing,126.175,-45.415,2.478,0.9727
60
+ 17,6.8704,1,41,singing,126.175,-45.415,2.478,0.9727
61
+ 18,7.2745,1,41,singing,126.068,-45.521,2.478,0.9688
62
+ 19,7.6786,1,41,singing,126.068,-45.521,2.478,0.9648
63
+ 20,8.0828,1,41,singing,126.068,-45.521,2.478,0.957
64
+ 21,8.4869,1,41,singing,126.068,-45.521,2.478,0.9453
65
+ 22,8.8911,1,41,singing,126.068,-45.521,2.478,0.9297
66
+ 23,9.2952,1,41,singing,125.92,-45.575,2.478,0.9102
67
+ 24,9.6993,1,41,singing,125.92,-45.575,2.478,0.8828
68
+ 25,10.1035,1,41,singing,125.772,-45.449,2.478,0.832
69
+ 26,10.5076,1,41,singing,125.772,-45.449,2.478,0.7695
70
+ 27,10.9117,1,41,singing,125.838,-45.471,2.464,0.6406
71
+ 28,11.3159,1,41,singing,125.796,-45.418,2.464,0.5273
72
+ 29,11.72,1,41,singing,125.711,-45.125,2.45,0.3633
73
+ 30,12.1242,1,41,singing,125.472,-45.067,2.421,0.2119
74
+ 31,12.5283,1,41,singing,125.337,-45.085,2.407,0.127
75
+ 32,12.9324,1,41,singing,125.515,-44.798,2.379,0.0781
76
+ 33,13.3366,1,41,singing,125.492,-44.695,2.364,0.0466
77
+ 34,13.7407,1,41,singing,125.352,-44.506,2.336,0.028
78
+ 35,14.1449,1,41,singing,125.208,-44.311,2.308,0.0175
79
+ 36,14.549,1,41,singing,125.128,-44.124,2.28,0.0103
80
+ 37,14.9531,1,41,singing,125.268,-43.718,2.252,0.0065
81
+ 38,15.3573,1,41,singing,125.01,-43.792,2.224,0.0039
82
+ 39,15.7614,1,41,singing,124.974,-43.421,2.182,0.0026
83
+ 40,16.1656,1,41,singing,125.175,-43.17,2.168,0.002
84
+ 0,0.0,2,41,singing,129.13,-40.517,2.478,0.707
85
+ 1,0.4041,2,41,singing,129.121,-41.206,2.493,0.8984
86
+ 2,0.8083,2,41,singing,129.174,-41.57,2.493,0.9492
87
+ 3,1.2124,2,41,singing,129.058,-41.906,2.493,0.9688
88
+ 4,1.6166,2,41,singing,128.944,-42.037,2.493,0.9766
89
+ 5,2.0207,2,41,singing,128.944,-42.436,2.507,0.9805
90
+ 6,2.4248,2,41,singing,128.944,-42.633,2.507,0.9844
91
+ 7,2.829,2,41,singing,128.83,-42.759,2.507,0.9844
92
+ 8,3.2331,2,41,singing,128.688,-42.816,2.507,0.9883
93
+ 9,3.6372,2,41,singing,128.83,-42.954,2.507,0.9883
94
+ 10,4.0414,2,41,singing,128.575,-42.94,2.507,0.9883
95
+ 11,4.4455,2,41,singing,128.716,-43.077,2.507,0.9883
96
+ 12,4.8497,2,41,singing,128.716,-43.077,2.507,0.9883
97
+ 13,5.2538,2,41,singing,128.575,-43.326,2.521,0.9883
98
+ 14,5.6579,2,41,singing,128.688,-43.204,2.521,0.9883
99
+ 15,6.0621,2,41,singing,128.688,-43.397,2.521,0.9883
100
+ 16,6.4662,2,41,singing,128.575,-43.518,2.521,0.9883
101
+ 17,6.8704,2,41,singing,128.575,-43.518,2.536,0.9844
102
+ 18,7.2745,2,41,singing,128.462,-43.447,2.536,0.9844
103
+ 19,7.6786,2,41,singing,128.575,-43.518,2.536,0.9805
104
+ 20,8.0828,2,41,singing,128.462,-43.447,2.536,0.9766
105
+ 21,8.4869,2,41,singing,128.35,-43.566,2.55,0.9688
106
+ 22,8.8911,2,41,singing,128.32,-43.693,2.55,0.957
107
+ 23,9.2952,2,41,singing,128.066,-43.678,2.55,0.9414
108
+ 24,9.6993,2,41,singing,128.066,-43.678,2.55,0.9141
109
+ 25,10.1035,2,41,singing,127.923,-43.544,2.55,0.8672
110
+ 26,10.5076,2,41,singing,127.891,-43.671,2.55,0.8047
111
+ 27,10.9117,2,41,singing,127.714,-43.471,2.55,0.668
112
+ 28,11.3159,2,41,singing,127.761,-43.609,2.536,0.5391
113
+ 29,11.72,2,41,singing,127.694,-43.478,2.521,0.3711
114
+ 30,12.1242,2,41,singing,127.326,-43.459,2.507,0.2109
115
+ 31,12.5283,2,41,singing,127.063,-43.312,2.493,0.1299
116
+ 32,12.9324,2,41,singing,127.26,-43.191,2.464,0.0825
117
+ 33,13.3366,2,41,singing,127.147,-42.974,2.45,0.0527
118
+ 34,13.7407,2,41,singing,126.991,-42.89,2.435,0.0337
119
+ 35,14.1449,2,41,singing,126.747,-42.797,2.421,0.0223
120
+ 36,14.549,2,41,singing,126.87,-42.614,2.393,0.014
121
+ 37,14.9531,2,41,singing,126.743,-42.484,2.364,0.0097
122
+ 38,15.3573,2,41,singing,126.567,-42.381,2.336,0.0065
123
+ 39,15.7614,2,41,singing,126.431,-42.12,2.308,0.0043
124
+ 40,16.1656,2,41,singing,126.693,-41.931,2.294,0.0036
125
+ 0,0.0,3,41,singing,125.854,-41.759,2.564,0.6719
126
+ 1,0.4041,3,41,singing,125.725,-42.357,2.579,0.9141
127
+ 2,0.8083,3,41,singing,125.725,-42.938,2.579,0.9609
128
+ 3,1.2124,3,41,singing,125.372,-43.036,2.579,0.9766
129
+ 4,1.6166,3,41,singing,125.433,-43.423,2.564,0.9844
130
+ 5,2.0207,3,41,singing,125.287,-43.664,2.564,0.9883
131
+ 6,2.4248,3,41,singing,125.14,-43.716,2.564,0.9922
132
+ 7,2.829,3,41,singing,124.992,-43.955,2.564,0.9922
133
+ 8,3.2331,3,41,singing,124.844,-44.007,2.564,0.9922
134
+ 9,3.6372,3,41,singing,125.052,-44.155,2.564,0.9922
135
+ 10,4.0414,3,41,singing,124.902,-44.207,2.564,0.9922
136
+ 11,4.4455,3,41,singing,124.902,-44.207,2.564,0.9922
137
+ 12,4.8497,3,41,singing,124.902,-44.207,2.564,0.9922
138
+ 13,5.2538,3,41,singing,124.753,-44.446,2.564,0.9922
139
+ 14,5.6579,3,41,singing,124.753,-44.259,2.579,0.9922
140
+ 15,6.0621,3,41,singing,124.962,-44.596,2.564,0.9922
141
+ 16,6.4662,3,41,singing,124.753,-44.446,2.564,0.9883
142
+ 17,6.8704,3,41,singing,124.753,-44.446,2.579,0.9883
143
+ 18,7.2745,3,41,singing,124.603,-44.498,2.579,0.9883
144
+ 19,7.6786,3,41,singing,124.603,-44.498,2.579,0.9844
145
+ 20,8.0828,3,41,singing,124.603,-44.683,2.579,0.9805
146
+ 21,8.4869,3,41,singing,124.603,-44.683,2.579,0.9766
147
+ 22,8.8911,3,41,singing,124.603,-44.683,2.593,0.9727
148
+ 23,9.2952,3,41,singing,124.452,-44.55,2.593,0.9648
149
+ 24,9.6993,3,41,singing,124.452,-44.735,2.593,0.957
150
+ 25,10.1035,3,41,singing,124.452,-44.55,2.593,0.9453
151
+ 26,10.5076,3,41,singing,124.452,-44.55,2.593,0.9297
152
+ 27,10.9117,3,41,singing,124.452,-44.55,2.593,0.9023
153
+ 28,11.3159,3,41,singing,124.301,-44.415,2.593,0.8672
154
+ 29,11.72,3,41,singing,124.356,-44.431,2.593,0.8125
155
+ 30,12.1242,3,41,singing,124.203,-44.294,2.593,0.7109
156
+ 31,12.5283,3,41,singing,124.0,-44.285,2.593,0.6094
157
+ 32,12.9324,3,41,singing,124.104,-44.17,2.579,0.5039
158
+ 33,13.3366,3,41,singing,124.266,-44.263,2.579,0.377
159
+ 34,13.7407,3,41,singing,124.059,-43.861,2.564,0.2754
160
+ 35,14.1449,3,41,singing,123.956,-43.728,2.55,0.1787
161
+ 36,14.549,3,41,singing,123.853,-43.739,2.536,0.0996
162
+ 37,14.9531,3,41,singing,123.69,-43.319,2.507,0.0566
163
+ 38,15.3573,3,41,singing,123.577,-43.105,2.478,0.0284
164
+ 39,15.7614,3,41,singing,123.286,-42.825,2.45,0.015
165
+ 40,16.1656,3,41,singing,123.573,-42.436,2.435,0.0097
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__hm3d__00034-6imZUJGRUq4__000096-foa__161195__gt.csv ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,11,war_sound,43.236,-30.502,3.012,1.0
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__hm3d__00056-kdw2Uapns3b__000020-foa__89741__pred.csv ADDED
@@ -0,0 +1,201 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,5,water,-162.55,-10.769,1.196,0.1973
3
+ 1,0.4,0,5,water,-162.525,-10.887,1.207,0.4219
4
+ 2,0.8,0,5,water,-162.58,-10.932,1.215,0.5586
5
+ 3,1.2,0,5,water,-162.502,-11.0,1.215,0.6328
6
+ 4,1.6,0,5,water,-162.568,-10.961,1.215,0.6797
7
+ 5,2.0,0,5,water,-162.698,-10.936,1.221,0.707
8
+ 6,2.4,0,5,water,-162.594,-11.035,1.215,0.7305
9
+ 7,2.8,0,5,water,-162.801,-11.099,1.215,0.75
10
+ 8,3.2,0,5,water,-162.659,-11.048,1.218,0.7578
11
+ 9,3.6,0,5,water,-162.889,-10.977,1.218,0.7695
12
+ 10,4.0,0,5,water,-162.85,-10.984,1.218,0.7773
13
+ 11,4.4,0,5,water,-162.85,-11.036,1.221,0.7812
14
+ 12,4.8,0,5,water,-162.874,-10.902,1.218,0.7852
15
+ 13,5.2,0,5,water,-162.874,-10.902,1.226,0.7891
16
+ 14,5.6,0,5,water,-162.874,-10.851,1.229,0.793
17
+ 15,6.0,0,5,water,-162.874,-10.902,1.232,0.793
18
+ 16,6.4,0,5,water,-162.998,-10.828,1.232,0.7969
19
+ 17,6.8,0,5,water,-163.099,-10.782,1.237,0.7969
20
+ 18,7.2,0,5,water,-163.099,-10.833,1.243,0.7969
21
+ 19,7.6,0,5,water,-163.221,-10.709,1.243,0.793
22
+ 20,8.0,0,5,water,-163.321,-10.765,1.251,0.793
23
+ 21,8.4,0,5,water,-163.44,-10.641,1.254,0.793
24
+ 22,8.8,0,5,water,-163.44,-10.541,1.265,0.7891
25
+ 23,9.2,0,5,water,-163.44,-10.591,1.265,0.7891
26
+ 24,9.6,0,5,water,-163.558,-10.47,1.271,0.7891
27
+ 25,10.0,0,5,water,-163.558,-10.42,1.279,0.7852
28
+ 26,10.4,0,5,water,-163.558,-10.32,1.285,0.7852
29
+ 27,10.8,0,5,water,-163.674,-10.201,1.291,0.7812
30
+ 28,11.2,0,5,water,-163.674,-10.201,1.302,0.7812
31
+ 29,11.6,0,5,water,-163.674,-10.052,1.308,0.7773
32
+ 30,12.0,0,5,water,-163.789,-9.935,1.319,0.7734
33
+ 31,12.4,0,5,water,-163.593,-9.826,1.325,0.7695
34
+ 32,12.8,0,5,water,-163.707,-9.663,1.336,0.7656
35
+ 33,13.2,0,5,water,-163.707,-9.663,1.342,0.7617
36
+ 34,13.6,0,5,water,-163.61,-9.511,1.359,0.7539
37
+ 35,14.0,0,5,water,-163.514,-9.359,1.365,0.7461
38
+ 36,14.4,0,5,water,-163.417,-9.305,1.383,0.7344
39
+ 37,14.8,0,5,water,-163.514,-9.162,1.394,0.7227
40
+ 38,15.2,0,5,water,-163.32,-9.005,1.418,0.7031
41
+ 39,15.6,0,5,water,-163.223,-8.853,1.436,0.6758
42
+ 40,16.0,0,5,water,-163.223,-8.804,1.454,0.6562
43
+ 41,16.4,0,5,water,-163.009,-8.61,1.472,0.625
44
+ 42,16.8,0,5,water,-162.912,-8.457,1.484,0.5859
45
+ 43,17.2,0,5,water,-162.912,-8.308,1.508,0.5469
46
+ 44,17.6,0,5,water,-162.695,-8.16,1.532,0.4922
47
+ 45,18.0,0,5,water,-162.597,-8.006,1.551,0.4434
48
+ 46,18.4,0,5,water,-162.671,-7.867,1.569,0.4082
49
+ 47,18.8,0,5,water,-162.35,-7.861,1.582,0.3809
50
+ 48,19.2,0,5,water,-162.547,-7.717,1.607,0.3496
51
+ 49,19.6,0,5,water,-162.224,-7.71,1.607,0.334
52
+ 0,0.0,1,5,water,-162.897,-10.867,1.172,0.1953
53
+ 1,0.4,1,5,water,-162.822,-10.938,1.182,0.4277
54
+ 2,0.8,1,5,water,-162.807,-11.076,1.188,0.5781
55
+ 3,1.2,1,5,water,-162.726,-11.144,1.191,0.6562
56
+ 4,1.6,1,5,water,-162.962,-11.07,1.188,0.7031
57
+ 5,2.0,1,5,water,-162.922,-11.131,1.193,0.7383
58
+ 6,2.4,1,5,water,-162.881,-11.138,1.188,0.7656
59
+ 7,2.8,1,5,water,-162.986,-11.249,1.188,0.7852
60
+ 8,3.2,1,5,water,-162.841,-11.197,1.191,0.793
61
+ 9,3.6,1,5,water,-163.009,-11.216,1.191,0.8086
62
+ 10,4.0,1,5,water,-162.905,-11.21,1.191,0.8164
63
+ 11,4.4,1,5,water,-163.032,-11.183,1.193,0.8242
64
+ 12,4.8,1,5,water,-163.095,-11.092,1.191,0.8281
65
+ 13,5.2,1,5,water,-163.095,-11.092,1.199,0.8359
66
+ 14,5.6,1,5,water,-162.992,-11.138,1.201,0.8398
67
+ 15,6.0,1,5,water,-162.952,-11.093,1.204,0.8438
68
+ 16,6.4,1,5,water,-163.055,-11.099,1.201,0.8477
69
+ 17,6.8,1,5,water,-163.055,-11.099,1.21,0.8516
70
+ 18,7.2,1,5,water,-163.179,-11.073,1.212,0.8516
71
+ 19,7.6,1,5,water,-163.28,-11.028,1.215,0.8516
72
+ 20,8.0,1,5,water,-163.402,-11.003,1.221,0.8516
73
+ 21,8.4,1,5,water,-163.301,-10.997,1.223,0.8516
74
+ 22,8.8,1,5,water,-163.402,-10.85,1.232,0.8516
75
+ 23,9.2,1,5,water,-163.421,-10.871,1.232,0.8516
76
+ 24,9.6,1,5,water,-163.421,-10.77,1.24,0.8516
77
+ 25,10.0,1,5,water,-163.54,-10.697,1.248,0.8516
78
+ 26,10.4,1,5,water,-163.54,-10.596,1.254,0.8516
79
+ 27,10.8,1,5,water,-163.44,-10.541,1.26,0.8477
80
+ 28,11.2,1,5,water,-163.558,-10.47,1.268,0.8477
81
+ 29,11.6,1,5,water,-163.558,-10.37,1.274,0.8438
82
+ 30,12.0,1,5,water,-163.36,-10.259,1.288,0.8398
83
+ 31,12.4,1,5,water,-163.477,-10.091,1.293,0.8398
84
+ 32,12.8,1,5,water,-163.281,-10.031,1.302,0.8359
85
+ 33,13.2,1,5,water,-163.281,-9.982,1.31,0.8281
86
+ 34,13.6,1,5,water,-163.281,-9.783,1.33,0.8203
87
+ 35,14.0,1,5,water,-163.085,-9.724,1.336,0.8125
88
+ 36,14.4,1,5,water,-163.106,-9.506,1.354,0.8008
89
+ 37,14.8,1,5,water,-163.106,-9.456,1.365,0.7852
90
+ 38,15.2,1,5,water,-162.912,-9.2,1.389,0.7578
91
+ 39,15.6,1,5,water,-162.815,-9.096,1.406,0.7266
92
+ 40,16.0,1,5,water,-162.815,-8.998,1.424,0.6992
93
+ 41,16.4,1,5,water,-162.5,-8.849,1.442,0.6602
94
+ 42,16.8,1,5,water,-162.306,-8.691,1.454,0.6133
95
+ 43,17.2,1,5,water,-162.403,-8.496,1.478,0.5664
96
+ 44,17.6,1,5,water,-162.181,-8.348,1.508,0.498
97
+ 45,18.0,1,5,water,-161.957,-8.248,1.526,0.4395
98
+ 46,18.4,1,5,water,-162.055,-8.101,1.545,0.3984
99
+ 47,18.8,1,5,water,-161.729,-8.045,1.557,0.3672
100
+ 48,19.2,1,5,water,-161.797,-7.904,1.576,0.3301
101
+ 49,19.6,1,5,water,-161.598,-7.844,1.582,0.3164
102
+ 0,0.0,2,5,water,-163.121,-10.087,1.161,0.2734
103
+ 1,0.4,2,5,water,-163.082,-10.115,1.166,0.5078
104
+ 2,0.8,2,5,water,-163.129,-10.172,1.172,0.6328
105
+ 3,1.2,2,5,water,-163.045,-10.249,1.172,0.6914
106
+ 4,1.6,2,5,water,-163.28,-10.127,1.172,0.7227
107
+ 5,2.0,2,5,water,-163.174,-10.228,1.174,0.7461
108
+ 6,2.4,2,5,water,-163.237,-10.245,1.169,0.7617
109
+ 7,2.8,2,5,water,-163.301,-10.314,1.166,0.7734
110
+ 8,3.2,2,5,water,-163.259,-10.219,1.169,0.7773
111
+ 9,3.6,2,5,water,-163.426,-10.241,1.172,0.7852
112
+ 10,4.0,2,5,water,-163.322,-10.235,1.169,0.7852
113
+ 11,4.4,2,5,water,-163.384,-10.199,1.172,0.7891
114
+ 12,4.8,2,5,water,-163.446,-10.11,1.169,0.7891
115
+ 13,5.2,2,5,water,-163.507,-10.127,1.177,0.7891
116
+ 14,5.6,2,5,water,-163.507,-10.022,1.18,0.7891
117
+ 15,6.0,2,5,water,-163.465,-10.034,1.18,0.7891
118
+ 16,6.4,2,5,water,-163.465,-10.034,1.182,0.7852
119
+ 17,6.8,2,5,water,-163.791,-9.922,1.188,0.7852
120
+ 18,7.2,2,5,water,-163.689,-9.917,1.193,0.7812
121
+ 19,7.6,2,5,water,-163.909,-9.802,1.196,0.7812
122
+ 20,8.0,2,5,water,-163.909,-9.802,1.204,0.7773
123
+ 21,8.4,2,5,water,-164.011,-9.755,1.21,0.7734
124
+ 22,8.8,2,5,water,-164.127,-9.534,1.218,0.7695
125
+ 23,9.2,2,5,water,-164.127,-9.586,1.221,0.7695
126
+ 24,9.6,2,5,water,-164.127,-9.534,1.229,0.7617
127
+ 25,10.0,2,5,water,-164.141,-9.363,1.237,0.7617
128
+ 26,10.4,2,5,water,-164.241,-9.265,1.248,0.7578
129
+ 27,10.8,2,5,water,-164.254,-9.147,1.257,0.7539
130
+ 28,11.2,2,5,water,-164.254,-9.096,1.265,0.7539
131
+ 29,11.6,2,5,water,-164.254,-8.894,1.274,0.7461
132
+ 30,12.0,2,5,water,-164.267,-8.779,1.288,0.7422
133
+ 31,12.4,2,5,water,-164.168,-8.624,1.299,0.7383
134
+ 32,12.8,2,5,water,-164.168,-8.574,1.308,0.7344
135
+ 33,13.2,2,5,water,-164.069,-8.52,1.319,0.7305
136
+ 34,13.6,2,5,water,-164.069,-8.318,1.336,0.7227
137
+ 35,14.0,2,5,water,-163.871,-8.159,1.348,0.7109
138
+ 36,14.4,2,5,water,-163.871,-8.109,1.365,0.6953
139
+ 37,14.8,2,5,water,-163.871,-8.008,1.383,0.6836
140
+ 38,15.2,2,5,water,-163.674,-7.749,1.406,0.6562
141
+ 39,15.6,2,5,water,-163.674,-7.598,1.43,0.625
142
+ 40,16.0,2,5,water,-163.576,-7.544,1.442,0.5977
143
+ 41,16.4,2,5,water,-163.36,-7.388,1.466,0.5625
144
+ 42,16.8,2,5,water,-163.36,-7.185,1.478,0.5195
145
+ 43,17.2,2,5,water,-163.341,-7.136,1.502,0.4805
146
+ 44,17.6,2,5,water,-163.241,-6.928,1.532,0.4258
147
+ 45,18.0,2,5,water,-163.02,-6.817,1.551,0.3789
148
+ 46,18.4,2,5,water,-163.12,-6.718,1.569,0.3477
149
+ 47,18.8,2,5,water,-162.897,-6.657,1.582,0.3223
150
+ 48,19.2,2,5,water,-162.874,-6.577,1.607,0.2949
151
+ 49,19.6,2,5,water,-162.671,-6.517,1.607,0.2832
152
+ 0,0.0,3,5,water,-164.118,-10.265,1.072,0.1445
153
+ 1,0.4,3,5,water,-164.116,-10.304,1.097,0.3594
154
+ 2,0.8,3,5,water,-164.252,-10.416,1.113,0.5273
155
+ 3,1.2,3,5,water,-164.219,-10.347,1.121,0.6289
156
+ 4,1.6,3,5,water,-164.278,-10.31,1.132,0.6953
157
+ 5,2.0,3,5,water,-164.337,-10.378,1.14,0.7422
158
+ 6,2.4,3,5,water,-164.395,-10.393,1.142,0.7773
159
+ 7,2.8,3,5,water,-164.395,-10.445,1.148,0.8086
160
+ 8,3.2,3,5,water,-164.407,-10.315,1.156,0.8242
161
+ 9,3.6,3,5,water,-164.51,-10.32,1.164,0.8438
162
+ 10,4.0,3,5,water,-164.624,-10.248,1.169,0.8555
163
+ 11,4.4,3,5,water,-164.521,-10.243,1.174,0.8672
164
+ 12,4.8,3,5,water,-164.634,-10.172,1.18,0.875
165
+ 13,5.2,3,5,water,-164.634,-10.121,1.191,0.8828
166
+ 14,5.6,3,5,water,-164.634,-10.018,1.201,0.8906
167
+ 15,6.0,3,5,water,-164.431,-10.009,1.207,0.8945
168
+ 16,6.4,3,5,water,-164.543,-9.889,1.212,0.8984
169
+ 17,6.8,3,5,water,-164.543,-9.838,1.221,0.9023
170
+ 18,7.2,3,5,water,-164.442,-9.833,1.232,0.9023
171
+ 19,7.6,3,5,water,-164.654,-9.72,1.237,0.9023
172
+ 20,8.0,3,5,water,-164.454,-9.711,1.248,0.9023
173
+ 21,8.4,3,5,water,-164.454,-9.61,1.257,0.9062
174
+ 22,8.8,3,5,water,-164.354,-9.505,1.268,0.9023
175
+ 23,9.2,3,5,water,-164.465,-9.39,1.274,0.9023
176
+ 24,9.6,3,5,water,-164.366,-9.336,1.279,0.9023
177
+ 25,10.0,3,5,water,-164.267,-9.281,1.291,0.8984
178
+ 26,10.4,3,5,water,-164.267,-9.181,1.299,0.8945
179
+ 27,10.8,3,5,water,-164.168,-9.076,1.308,0.8906
180
+ 28,11.2,3,5,water,-164.168,-9.026,1.319,0.8906
181
+ 29,11.6,3,5,water,-164.069,-8.871,1.325,0.8828
182
+ 30,12.0,3,5,water,-164.083,-8.707,1.342,0.875
183
+ 31,12.4,3,5,water,-163.985,-8.653,1.348,0.8672
184
+ 32,12.8,3,5,water,-163.887,-8.549,1.354,0.8594
185
+ 33,13.2,3,5,water,-163.789,-8.495,1.359,0.8516
186
+ 34,13.6,3,5,water,-163.789,-8.346,1.377,0.8398
187
+ 35,14.0,3,5,water,-163.691,-8.192,1.383,0.8242
188
+ 36,14.4,3,5,water,-163.691,-8.142,1.4,0.8047
189
+ 37,14.8,3,5,water,-163.691,-8.043,1.412,0.7812
190
+ 38,15.2,3,5,water,-163.379,-7.888,1.43,0.7461
191
+ 39,15.6,3,5,water,-163.379,-7.788,1.448,0.7031
192
+ 40,16.0,3,5,water,-163.379,-7.687,1.46,0.668
193
+ 41,16.4,3,5,water,-163.163,-7.583,1.478,0.6211
194
+ 42,16.8,3,5,water,-163.163,-7.381,1.484,0.5664
195
+ 43,17.2,3,5,water,-163.261,-7.232,1.508,0.5117
196
+ 44,17.6,3,5,water,-163.142,-7.129,1.532,0.4395
197
+ 45,18.0,3,5,water,-163.02,-7.075,1.545,0.3848
198
+ 46,18.4,3,5,water,-163.12,-6.873,1.563,0.3418
199
+ 47,18.8,3,5,water,-162.797,-6.861,1.569,0.3145
200
+ 48,19.2,3,5,water,-162.874,-6.707,1.594,0.2754
201
+ 49,19.6,3,5,water,-162.671,-6.674,1.594,0.2637
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov2_000022__gt.csv ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,21,footsteps,134.847,0.876,1.823,1.0
3
+ 1,0.3965,0,21,footsteps,134.847,0.876,1.823,1.0
4
+ 2,0.7929,0,21,footsteps,134.847,0.876,1.823,1.0
5
+ 3,1.1894,0,21,footsteps,134.847,0.876,1.823,1.0
6
+ 4,1.5859,0,21,footsteps,134.847,0.876,1.823,1.0
7
+ 5,1.9823,0,21,footsteps,134.847,0.876,1.823,1.0
8
+ 6,2.3788,0,21,footsteps,134.847,0.876,1.823,1.0
9
+ 7,2.7753,0,21,footsteps,134.847,0.876,1.823,1.0
10
+ 8,3.1718,0,21,footsteps,134.847,0.876,1.823,1.0
11
+ 9,3.5682,0,21,footsteps,134.847,0.876,1.823,1.0
12
+ 10,3.9647,0,21,footsteps,134.847,0.876,1.823,1.0
13
+ 11,4.3612,0,21,footsteps,134.847,0.876,1.823,1.0
14
+ 12,4.7576,0,21,footsteps,134.847,0.876,1.823,1.0
15
+ 13,5.1541,0,21,footsteps,134.847,0.876,1.823,1.0
16
+ 12,4.7576,1,18,dog,68.702,-3.533,1.157,1.0
17
+ 13,5.1541,1,18,dog,68.702,-3.533,1.157,1.0
18
+ 14,5.5506,1,18,dog,68.702,-3.533,1.157,1.0
19
+ 15,5.947,1,18,dog,68.702,-3.533,1.157,1.0
20
+ 16,6.3435,1,18,dog,68.702,-3.533,1.157,1.0
21
+ 17,6.74,1,18,dog,68.702,-3.533,1.157,1.0
22
+ 18,7.1364,1,18,dog,68.702,-3.533,1.157,1.0
23
+ 19,7.5329,1,18,dog,68.702,-3.533,1.157,1.0
24
+ 20,7.9294,1,18,dog,68.702,-3.533,1.157,1.0
25
+ 21,8.3258,1,18,dog,68.702,-3.533,1.157,1.0
26
+ 22,8.7223,1,18,dog,68.702,-3.533,1.157,1.0
27
+ 23,9.1188,1,18,dog,68.702,-3.533,1.157,1.0
28
+ 24,9.5153,1,18,dog,68.702,-3.533,1.157,1.0
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov2_000031__gt.csv ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,49,paper,123.534,-0.801,4.953,1.0
3
+ 1,0.4017,0,49,paper,123.534,-0.801,4.953,1.0
4
+ 2,0.8034,0,49,paper,123.534,-0.801,4.953,1.0
5
+ 3,1.2051,0,49,paper,123.534,-0.801,4.953,1.0
6
+ 4,1.6068,0,49,paper,123.534,-0.801,4.953,1.0
7
+ 5,2.0085,0,49,paper,123.534,-0.801,4.953,1.0
8
+ 6,2.4102,0,49,paper,123.534,-0.801,4.953,1.0
9
+ 7,2.8119,0,49,paper,123.534,-0.801,4.953,1.0
10
+ 8,3.2137,0,49,paper,123.534,-0.801,4.953,1.0
11
+ 9,3.6154,0,49,paper,123.534,-0.801,4.953,1.0
12
+ 10,4.0171,0,49,paper,123.534,-0.801,4.953,1.0
13
+ 11,4.4188,0,49,paper,123.534,-0.801,4.953,1.0
14
+ 12,4.8205,0,49,paper,123.534,-0.801,4.953,1.0
15
+ 13,5.2222,0,49,paper,123.534,-0.801,4.953,1.0
16
+ 14,5.6239,0,49,paper,123.534,-0.801,4.953,1.0
17
+ 15,6.0256,0,49,paper,123.534,-0.801,4.953,1.0
18
+ 16,6.4273,0,49,paper,123.534,-0.801,4.953,1.0
19
+ 17,6.829,0,49,paper,123.534,-0.801,4.953,1.0
20
+ 18,7.2307,0,49,paper,123.534,-0.801,4.953,1.0
21
+ 19,7.6324,0,49,paper,123.534,-0.801,4.953,1.0
22
+ 20,8.0341,0,49,paper,123.534,-0.801,4.953,1.0
23
+ 21,8.4358,0,49,paper,123.534,-0.801,4.953,1.0
24
+ 22,8.8375,0,49,paper,123.534,-0.801,4.953,1.0
25
+ 23,9.2393,0,49,paper,123.534,-0.801,4.953,1.0
26
+ 24,9.641,0,49,paper,123.534,-0.801,4.953,1.0
27
+ 25,10.0427,0,49,paper,123.534,-0.801,4.953,1.0
28
+ 26,10.4444,0,49,paper,123.534,-0.801,4.953,1.0
29
+ 27,10.8461,0,49,paper,123.534,-0.801,4.953,1.0
30
+ 28,11.2478,0,49,paper,123.534,-0.801,4.953,1.0
31
+ 29,11.6495,0,49,paper,123.534,-0.801,4.953,1.0
32
+ 30,12.0512,0,49,paper,123.534,-0.801,4.953,1.0
33
+ 31,12.4529,0,49,paper,123.534,-0.801,4.953,1.0
34
+ 32,12.8546,0,49,paper,123.534,-0.801,4.953,1.0
35
+ 33,13.2563,0,49,paper,123.534,-0.801,4.953,1.0
36
+ 34,13.658,0,49,paper,123.534,-0.801,4.953,1.0
37
+ 35,14.0597,0,49,paper,123.534,-0.801,4.953,1.0
38
+ 36,14.4614,0,49,paper,123.534,-0.801,4.953,1.0
39
+ 37,14.8631,0,49,paper,123.534,-0.801,4.953,1.0
40
+ 1,0.4017,1,30,door,-120.011,-39.052,2.313,1.0
41
+ 2,0.8034,1,30,door,-120.011,-39.052,2.313,1.0
42
+ 3,1.2051,1,30,door,-120.011,-39.052,2.313,1.0
43
+ 4,1.6068,1,30,door,-120.011,-39.052,2.313,1.0
44
+ 5,2.0085,1,30,door,-120.011,-39.052,2.313,1.0
45
+ 6,2.4102,1,30,door,-120.011,-39.052,2.313,1.0
46
+ 7,2.8119,1,30,door,-120.011,-39.052,2.313,1.0
47
+ 9,3.6154,2,18,dog,-140.274,-10.193,4.696,1.0
48
+ 10,4.0171,2,18,dog,-140.274,-10.193,4.696,1.0
49
+ 11,4.4188,2,18,dog,-140.274,-10.193,4.696,1.0
50
+ 12,4.8205,2,18,dog,-140.274,-10.193,4.696,1.0
51
+ 13,5.2222,2,18,dog,-140.274,-10.193,4.696,1.0
52
+ 14,5.6239,2,18,dog,-140.274,-10.193,4.696,1.0
53
+ 15,6.0256,2,18,dog,-140.274,-10.193,4.696,1.0
54
+ 16,6.4273,2,18,dog,-140.274,-10.193,4.696,1.0
55
+ 28,11.2478,3,48,appliance,-161.588,15.486,3.834,1.0
56
+ 29,11.6495,3,48,appliance,-161.588,15.486,3.834,1.0
57
+ 30,12.0512,3,48,appliance,-161.588,15.486,3.834,1.0
58
+ 31,12.4529,3,48,appliance,-161.588,15.486,3.834,1.0
59
+ 32,12.8546,3,48,appliance,-161.588,15.486,3.834,1.0
60
+ 33,13.2563,3,48,appliance,-161.588,15.486,3.834,1.0
61
+ 34,13.658,3,48,appliance,-161.588,15.486,3.834,1.0
62
+ 35,14.0597,3,48,appliance,-161.588,15.486,3.834,1.0
63
+ 36,14.4614,3,48,appliance,-161.588,15.486,3.834,1.0
64
+ 37,14.8631,3,48,appliance,-161.588,15.486,3.834,1.0
65
+ 38,15.2649,3,48,appliance,-161.588,15.486,3.834,1.0
66
+ 39,15.6666,3,48,appliance,-161.588,15.486,3.834,1.0
67
+ 40,16.0683,3,48,appliance,-161.588,15.486,3.834,1.0
68
+ 41,16.47,3,48,appliance,-161.588,15.486,3.834,1.0
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov2_000043__gt.csv ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 1,0.4056,0,41,singing,135.995,6.412,4.488,1.0
3
+ 2,0.8111,0,41,singing,135.995,6.412,4.488,1.0
4
+ 3,1.2167,0,41,singing,135.995,6.412,4.488,1.0
5
+ 4,1.6222,0,41,singing,135.995,6.412,4.488,1.0
6
+ 5,2.0278,0,41,singing,135.995,6.412,4.488,1.0
7
+ 6,2.4334,0,41,singing,135.995,6.412,4.488,1.0
8
+ 7,2.8389,0,41,singing,135.995,6.412,4.488,1.0
9
+ 8,3.2445,0,41,singing,135.995,6.412,4.488,1.0
10
+ 9,3.6501,0,41,singing,135.995,6.412,4.488,1.0
11
+ 10,4.0556,0,41,singing,135.995,6.412,4.488,1.0
12
+ 11,4.4612,0,41,singing,135.995,6.412,4.488,1.0
13
+ 12,4.8667,0,41,singing,135.995,6.412,4.488,1.0
14
+ 13,5.2723,0,41,singing,135.995,6.412,4.488,1.0
15
+ 14,5.6779,0,41,singing,135.995,6.412,4.488,1.0
16
+ 15,6.0834,0,41,singing,135.995,6.412,4.488,1.0
17
+ 16,6.489,0,41,singing,135.995,6.412,4.488,1.0
18
+ 17,6.8946,0,41,singing,135.995,6.412,4.488,1.0
19
+ 18,7.3001,0,41,singing,135.995,6.412,4.488,1.0
20
+ 19,7.7057,0,41,singing,135.995,6.412,4.488,1.0
21
+ 20,8.1112,0,41,singing,135.995,6.412,4.488,1.0
22
+ 21,8.5168,0,41,singing,135.995,6.412,4.488,1.0
23
+ 22,8.9224,0,41,singing,135.995,6.412,4.488,1.0
24
+ 1,0.4056,1,53,crackle,75.337,16.016,3.096,1.0
25
+ 2,0.8111,1,53,crackle,75.337,16.016,3.096,1.0
26
+ 3,1.2167,1,53,crackle,75.337,16.016,3.096,1.0
27
+ 4,1.6222,1,53,crackle,75.337,16.016,3.096,1.0
28
+ 5,2.0278,1,53,crackle,75.337,16.016,3.096,1.0
29
+ 6,2.4334,1,53,crackle,75.337,16.016,3.096,1.0
30
+ 7,2.8389,1,53,crackle,75.337,16.016,3.096,1.0
31
+ 8,3.2445,1,53,crackle,75.337,16.016,3.096,1.0
32
+ 9,3.6501,1,53,crackle,75.337,16.016,3.096,1.0
33
+ 10,4.0556,1,53,crackle,75.337,16.016,3.096,1.0
34
+ 11,4.4612,1,53,crackle,75.337,16.016,3.096,1.0
35
+ 12,4.8667,1,53,crackle,75.337,16.016,3.096,1.0
36
+ 13,5.2723,1,53,crackle,75.337,16.016,3.096,1.0
37
+ 14,5.6779,1,53,crackle,75.337,16.016,3.096,1.0
38
+ 15,6.0834,1,53,crackle,75.337,16.016,3.096,1.0
39
+ 16,6.489,1,53,crackle,75.337,16.016,3.096,1.0
40
+ 17,6.8946,1,53,crackle,75.337,16.016,3.096,1.0
41
+ 18,7.3001,1,53,crackle,75.337,16.016,3.096,1.0
42
+ 19,7.7057,1,53,crackle,75.337,16.016,3.096,1.0
43
+ 20,8.1112,1,53,crackle,75.337,16.016,3.096,1.0
44
+ 21,8.5168,1,53,crackle,75.337,16.016,3.096,1.0
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov2_000129__gt.csv ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,1,string_instrument,127.331,-55.972,1.811,1.0
3
+ 1,0.395,0,1,string_instrument,127.331,-55.972,1.811,1.0
4
+ 2,0.7899,0,1,string_instrument,127.331,-55.972,1.811,1.0
5
+ 3,1.1849,0,1,string_instrument,127.331,-55.972,1.811,1.0
6
+ 4,1.5799,0,1,string_instrument,127.331,-55.972,1.811,1.0
7
+ 3,1.1849,1,33,cat,23.369,-3.236,3.69,1.0
8
+ 4,1.5799,1,33,cat,23.369,-3.236,3.69,1.0
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov3_000055__pred.csv ADDED
@@ -0,0 +1,161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,57,printer,-106.058,9.509,4.203,0.6836
3
+ 1,0.4043,0,57,printer,-106.348,9.086,4.203,0.8281
4
+ 2,0.8086,0,57,printer,-106.328,9.008,4.172,0.875
5
+ 3,1.2129,0,57,printer,-106.423,9.102,4.172,0.8906
6
+ 4,1.6172,0,57,printer,-106.47,9.236,4.141,0.8984
7
+ 5,2.0215,0,57,printer,-106.308,9.225,4.11,0.8984
8
+ 6,2.4257,0,57,printer,-106.26,9.443,4.11,0.9023
9
+ 7,2.83,0,57,printer,-106.28,9.405,4.11,0.8984
10
+ 8,3.2343,0,57,printer,-106.163,9.352,4.08,0.8984
11
+ 9,3.6386,0,57,printer,-106.231,9.508,4.08,0.8945
12
+ 10,4.0429,0,57,printer,-106.182,9.552,4.049,0.8945
13
+ 11,4.4472,0,57,printer,-106.182,9.612,4.049,0.8906
14
+ 12,4.8515,0,57,printer,-106.25,9.591,4.049,0.8867
15
+ 13,5.2558,0,57,printer,-106.031,9.747,4.018,0.8828
16
+ 14,5.6601,0,57,printer,-105.989,9.653,4.003,0.8789
17
+ 15,6.0644,0,57,printer,-105.876,9.764,3.987,0.875
18
+ 16,6.4687,0,57,printer,-105.84,9.837,3.972,0.8672
19
+ 17,6.8729,0,57,printer,-105.874,9.828,3.957,0.8633
20
+ 18,7.2772,0,57,printer,-105.701,9.864,3.941,0.8555
21
+ 19,7.6815,0,57,printer,-105.481,9.98,3.911,0.8516
22
+ 20,8.0858,0,57,printer,-104.976,9.864,3.896,0.8438
23
+ 21,8.4901,0,57,printer,-104.733,9.878,3.865,0.8359
24
+ 22,8.8944,0,57,printer,-104.683,10.119,3.85,0.8281
25
+ 23,9.2987,0,57,printer,-104.259,10.021,3.834,0.8242
26
+ 24,9.703,0,57,printer,-104.27,10.008,3.819,0.8164
27
+ 25,10.1073,0,44,car,-104.2,10.16,3.789,0.8125
28
+ 26,10.5116,0,44,car,-104.123,10.435,3.758,0.8047
29
+ 27,10.9158,0,44,car,-104.036,10.515,3.727,0.7969
30
+ 28,11.3201,0,44,car,-103.842,10.729,3.712,0.793
31
+ 29,11.7244,0,44,car,-104.3,11.021,3.697,0.7852
32
+ 30,12.1287,0,44,car,-104.328,11.268,3.651,0.7773
33
+ 31,12.533,0,44,car,-104.74,11.897,3.621,0.7734
34
+ 32,12.9373,0,44,car,-105.486,12.329,3.606,0.7695
35
+ 33,13.3416,0,44,car,-106.021,12.99,3.575,0.7578
36
+ 34,13.7459,0,44,car,-106.598,13.971,3.56,0.7461
37
+ 35,14.1502,0,44,car,-107.331,14.518,3.53,0.7305
38
+ 36,14.5545,0,44,car,-108.228,15.864,3.499,0.6992
39
+ 37,14.9588,0,44,car,-109.398,17.569,3.484,0.6641
40
+ 38,15.363,0,44,car,-110.692,19.977,3.439,0.5938
41
+ 39,15.7673,0,44,car,-111.459,22.995,3.424,0.5078
42
+ 0,0.0,1,57,printer,-102.162,7.59,4.295,0.5352
43
+ 1,0.4043,1,57,printer,-102.396,7.104,4.295,0.7617
44
+ 2,0.8086,1,57,printer,-102.367,6.933,4.264,0.8398
45
+ 3,1.2129,1,57,printer,-102.325,6.934,4.264,0.8711
46
+ 4,1.6172,1,57,printer,-102.498,7.042,4.233,0.8867
47
+ 5,2.0215,1,57,printer,-102.27,7.048,4.233,0.8945
48
+ 6,2.4257,1,57,printer,-102.374,7.214,4.233,0.9023
49
+ 7,2.83,1,57,printer,-102.304,7.076,4.203,0.9062
50
+ 8,3.2343,1,57,printer,-102.095,6.997,4.203,0.9102
51
+ 9,3.6386,1,57,printer,-102.248,7.191,4.203,0.9102
52
+ 10,4.0429,1,57,printer,-102.297,7.191,4.172,0.9102
53
+ 11,4.4472,1,57,printer,-102.079,7.196,4.172,0.9102
54
+ 12,4.8515,1,57,printer,-102.289,7.221,4.172,0.9102
55
+ 13,5.2558,1,57,printer,-102.103,7.286,4.141,0.9062
56
+ 14,5.6601,1,57,printer,-101.913,7.119,4.141,0.9062
57
+ 15,6.0644,1,57,printer,-101.797,7.238,4.11,0.9023
58
+ 16,6.4687,1,57,printer,-101.7,7.275,4.11,0.8984
59
+ 17,6.8729,1,57,printer,-101.609,7.161,4.11,0.8945
60
+ 18,7.2772,1,57,printer,-101.467,7.166,4.08,0.8945
61
+ 19,7.6815,1,57,printer,-101.223,7.238,4.08,0.8906
62
+ 20,8.0858,1,57,printer,-100.667,7.098,4.049,0.8867
63
+ 21,8.4901,1,57,printer,-100.275,7.078,4.049,0.8789
64
+ 22,8.8944,1,57,printer,-100.048,7.118,4.018,0.875
65
+ 23,9.2987,1,57,printer,-99.556,6.922,4.003,0.8711
66
+ 24,9.703,1,57,printer,-99.255,6.814,3.987,0.8672
67
+ 25,10.1073,1,57,printer,-98.938,6.81,3.957,0.8633
68
+ 26,10.5116,1,57,printer,-98.637,6.923,3.941,0.8555
69
+ 27,10.9158,1,57,printer,-98.384,6.877,3.926,0.8516
70
+ 28,11.3201,1,57,printer,-98.045,6.873,3.896,0.8477
71
+ 29,11.7244,1,57,printer,-97.98,7.007,3.88,0.8398
72
+ 30,12.1287,1,57,printer,-97.765,7.033,3.865,0.832
73
+ 31,12.533,1,57,printer,-98.016,7.483,3.834,0.8203
74
+ 32,12.9373,1,57,printer,-98.292,7.745,3.819,0.8125
75
+ 33,13.3416,1,57,printer,-98.407,8.173,3.804,0.7969
76
+ 34,13.7459,1,57,printer,-98.385,8.857,3.789,0.7734
77
+ 35,14.1502,1,57,printer,-98.504,9.244,3.773,0.7422
78
+ 36,14.5545,1,44,car,-98.403,10.34,3.743,0.6797
79
+ 37,14.9588,1,44,car,-98.52,11.588,3.727,0.6133
80
+ 38,15.363,1,44,car,-98.307,13.599,3.697,0.4863
81
+ 39,15.7673,1,44,car,-97.712,15.799,3.682,0.3633
82
+ 0,0.0,2,57,printer,-102.674,9.053,4.264,0.6367
83
+ 1,0.4043,2,57,printer,-103.041,8.622,4.264,0.7852
84
+ 2,0.8086,2,57,printer,-102.858,8.439,4.233,0.8359
85
+ 3,1.2129,2,57,printer,-102.99,8.5,4.203,0.8516
86
+ 4,1.6172,2,57,printer,-103.055,8.63,4.203,0.8633
87
+ 5,2.0215,2,57,printer,-102.918,8.676,4.172,0.8633
88
+ 6,2.4257,2,57,printer,-102.913,8.851,4.172,0.8633
89
+ 7,2.83,2,57,printer,-102.835,8.805,4.141,0.8633
90
+ 8,3.2343,2,57,printer,-102.626,8.652,4.141,0.8633
91
+ 9,3.6386,2,57,printer,-102.755,8.758,4.11,0.8594
92
+ 10,4.0429,2,57,printer,-102.879,8.841,4.11,0.8555
93
+ 11,4.4472,2,57,printer,-102.81,8.877,4.11,0.8477
94
+ 12,4.8515,2,57,printer,-102.936,8.823,4.08,0.8438
95
+ 13,5.2558,2,57,printer,-102.779,8.963,4.08,0.8398
96
+ 14,5.6601,2,57,printer,-102.553,8.814,4.049,0.8359
97
+ 15,6.0644,2,57,printer,-102.402,8.936,4.049,0.8281
98
+ 16,6.4687,2,57,printer,-102.431,8.951,4.018,0.8203
99
+ 17,6.8729,2,57,printer,-102.32,8.886,4.003,0.8125
100
+ 18,7.2772,2,57,printer,-102.215,8.883,3.987,0.8086
101
+ 19,7.6815,2,57,printer,-101.926,8.948,3.972,0.8008
102
+ 20,8.0858,2,57,printer,-101.294,8.823,3.957,0.793
103
+ 21,8.4901,2,57,printer,-101.181,8.929,3.941,0.7812
104
+ 22,8.8944,2,57,printer,-100.954,8.998,3.926,0.7734
105
+ 23,9.2987,2,57,printer,-100.55,8.906,3.911,0.7656
106
+ 24,9.703,2,57,printer,-100.644,8.913,3.88,0.7578
107
+ 25,10.1073,2,57,printer,-100.49,8.98,3.865,0.75
108
+ 26,10.5116,2,44,car,-100.481,9.216,3.85,0.7461
109
+ 27,10.9158,2,44,car,-100.381,9.271,3.819,0.7383
110
+ 28,11.3201,2,44,car,-100.079,9.487,3.804,0.7344
111
+ 29,11.7244,2,44,car,-100.016,9.594,3.789,0.7266
112
+ 30,12.1287,2,44,car,-99.949,9.84,3.758,0.7227
113
+ 31,12.533,2,44,car,-99.943,10.464,3.743,0.7188
114
+ 32,12.9373,2,44,car,-100.483,10.804,3.712,0.7188
115
+ 33,13.3416,2,44,car,-100.544,11.46,3.697,0.7109
116
+ 34,13.7459,2,44,car,-100.479,12.071,3.682,0.7109
117
+ 35,14.1502,2,44,car,-100.838,12.594,3.651,0.707
118
+ 36,14.5545,2,44,car,-100.637,13.534,3.636,0.6953
119
+ 37,14.9588,2,44,car,-101.113,14.961,3.621,0.6797
120
+ 38,15.363,2,44,car,-101.03,17.164,3.575,0.6406
121
+ 39,15.7673,2,44,car,-100.351,19.766,3.56,0.5859
122
+ 0,0.0,3,57,printer,-102.236,12.45,3.926,0.5703
123
+ 1,0.4043,3,57,printer,-102.299,11.479,3.911,0.8008
124
+ 2,0.8086,3,57,printer,-102.095,10.988,3.88,0.875
125
+ 3,1.2129,3,57,printer,-102.106,10.927,3.865,0.9062
126
+ 4,1.6172,3,57,printer,-102.106,10.927,3.85,0.9219
127
+ 5,2.0215,3,57,printer,-101.703,10.772,3.819,0.9297
128
+ 6,2.4257,3,57,printer,-101.596,10.975,3.804,0.9375
129
+ 7,2.83,3,57,printer,-101.407,10.791,3.789,0.9414
130
+ 8,3.2343,3,57,printer,-100.923,10.571,3.773,0.9453
131
+ 9,3.6386,3,57,printer,-100.949,10.754,3.758,0.9453
132
+ 10,4.0429,3,57,printer,-100.85,10.702,3.743,0.9453
133
+ 11,4.4472,3,57,printer,-100.451,10.738,3.727,0.9453
134
+ 12,4.8515,3,57,printer,-100.383,10.64,3.712,0.9453
135
+ 13,5.2558,3,57,printer,-99.908,10.814,3.697,0.9453
136
+ 14,5.6601,3,57,printer,-99.615,10.609,3.682,0.9453
137
+ 15,6.0644,3,57,printer,-99.211,10.809,3.667,0.9453
138
+ 16,6.4687,3,57,printer,-99.004,10.877,3.651,0.9414
139
+ 17,6.8729,3,57,printer,-98.829,10.848,3.621,0.9414
140
+ 18,7.2772,3,57,printer,-98.787,10.941,3.606,0.9375
141
+ 19,7.6815,3,57,printer,-98.508,11.232,3.59,0.9336
142
+ 20,8.0858,3,44,car,-97.986,11.226,3.575,0.9297
143
+ 21,8.4901,3,44,car,-98.121,11.436,3.56,0.9258
144
+ 22,8.8944,3,44,car,-98.37,11.895,3.56,0.9219
145
+ 23,9.2987,3,44,car,-98.115,11.87,3.545,0.918
146
+ 24,9.703,3,44,car,-98.569,12.141,3.545,0.9141
147
+ 25,10.1073,3,44,car,-98.7,12.415,3.53,0.9102
148
+ 26,10.5116,3,44,car,-99.0,13.122,3.515,0.9062
149
+ 27,10.9158,3,44,car,-99.369,13.59,3.515,0.9023
150
+ 28,11.3201,3,44,car,-98.967,14.249,3.515,0.8945
151
+ 29,11.7244,3,44,car,-99.408,15.087,3.499,0.8906
152
+ 30,12.1287,3,44,car,-99.284,15.779,3.484,0.8828
153
+ 31,12.533,3,44,car,-99.266,17.126,3.484,0.875
154
+ 32,12.9373,3,44,car,-99.766,18.1,3.484,0.8672
155
+ 33,13.3416,3,44,car,-99.834,19.624,3.469,0.8555
156
+ 34,13.7459,3,44,car,-99.619,21.162,3.469,0.8398
157
+ 35,14.1502,3,44,car,-99.964,22.421,3.454,0.8242
158
+ 36,14.5545,3,44,car,-99.497,24.867,3.454,0.793
159
+ 37,14.9588,3,44,car,-100.299,27.966,3.454,0.7617
160
+ 38,15.363,3,44,car,-100.187,32.442,3.424,0.6992
161
+ 39,15.7673,3,44,car,-98.913,37.71,3.424,0.6289
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov3_000066__pred.csv ADDED
@@ -0,0 +1,201 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,19,vehicle,143.711,-18.468,1.174,0.6406
3
+ 1,0.4,0,19,vehicle,142.994,-19.144,1.174,0.7305
4
+ 2,0.8,0,19,vehicle,142.398,-19.849,1.169,0.7461
5
+ 3,1.2,0,19,vehicle,141.246,-20.778,1.164,0.7461
6
+ 4,1.6,0,19,vehicle,140.11,-21.653,1.161,0.7383
7
+ 5,2.0,0,19,vehicle,137.161,-23.748,1.158,0.7305
8
+ 6,2.4,0,19,vehicle,130.772,-27.163,1.156,0.7266
9
+ 7,2.8,0,19,vehicle,119.545,-31.795,1.161,0.7266
10
+ 8,3.2,0,19,vehicle,99.569,-36.736,1.166,0.7344
11
+ 9,3.6,0,19,vehicle,67.207,-36.88,1.169,0.7383
12
+ 10,4.0,0,45,rain,44.481,-31.601,1.177,0.7422
13
+ 11,4.4,0,45,rain,28.501,-24.865,1.182,0.75
14
+ 12,4.8,0,45,rain,19.676,-19.63,1.185,0.7539
15
+ 13,5.2,0,45,rain,15.079,-16.511,1.188,0.7578
16
+ 14,5.6,0,45,rain,11.696,-13.696,1.193,0.7617
17
+ 15,6.0,0,45,rain,9.734,-12.329,1.196,0.7656
18
+ 16,6.4,0,45,rain,8.717,-11.087,1.199,0.7695
19
+ 17,6.8,0,45,rain,7.803,-10.091,1.196,0.7695
20
+ 18,7.2,0,45,rain,7.427,-9.549,1.191,0.7695
21
+ 19,7.6,0,45,rain,7.272,-9.099,1.196,0.7695
22
+ 20,8.0,0,45,rain,6.744,-8.538,1.196,0.7734
23
+ 21,8.4,0,45,rain,6.749,-8.333,1.201,0.7734
24
+ 22,8.8,0,45,rain,6.658,-8.142,1.207,0.7734
25
+ 23,9.2,0,45,rain,6.571,-7.947,1.207,0.7773
26
+ 24,9.6,0,45,rain,6.848,-7.897,1.207,0.7773
27
+ 25,10.0,0,45,rain,6.667,-7.713,1.212,0.7812
28
+ 26,10.4,0,45,rain,6.758,-7.621,1.215,0.7812
29
+ 27,10.8,0,45,rain,6.85,-7.529,1.212,0.7852
30
+ 28,11.2,0,45,rain,6.67,-7.484,1.215,0.7891
31
+ 29,11.6,0,45,rain,6.804,-7.439,1.218,0.793
32
+ 30,12.0,0,45,rain,6.758,-7.349,1.223,0.7969
33
+ 31,12.4,0,45,rain,6.804,-7.257,1.223,0.8047
34
+ 32,12.8,0,45,rain,6.575,-7.261,1.226,0.8086
35
+ 33,13.2,0,45,rain,6.667,-7.259,1.223,0.8086
36
+ 34,13.6,0,45,rain,6.617,-7.17,1.229,0.8125
37
+ 35,14.0,0,45,rain,6.479,-7.126,1.226,0.8164
38
+ 36,14.4,0,45,rain,6.661,-7.124,1.234,0.8164
39
+ 37,14.8,0,45,rain,6.707,-7.123,1.237,0.8203
40
+ 38,15.2,0,45,rain,6.658,-7.124,1.24,0.8164
41
+ 39,15.6,0,45,rain,6.702,-7.124,1.237,0.8164
42
+ 40,16.0,0,45,rain,6.654,-7.124,1.243,0.8086
43
+ 41,16.4,0,45,rain,6.604,-7.219,1.248,0.8086
44
+ 42,16.8,0,45,rain,6.556,-7.126,1.246,0.8047
45
+ 43,17.2,0,45,rain,6.699,-7.124,1.257,0.7969
46
+ 44,17.6,0,45,rain,6.696,-7.124,1.26,0.7891
47
+ 45,18.0,0,45,rain,6.648,-7.125,1.262,0.7852
48
+ 46,18.4,0,45,rain,6.6,-7.031,1.271,0.7773
49
+ 47,18.8,0,45,rain,6.505,-7.032,1.271,0.7734
50
+ 48,19.2,0,45,rain,6.645,-7.125,1.279,0.7656
51
+ 49,19.6,0,45,rain,6.741,-7.076,1.282,0.7578
52
+ 0,0.0,1,10,machine,144.551,-16.522,1.313,0.8906
53
+ 1,0.4,1,10,machine,144.23,-16.637,1.296,0.9336
54
+ 2,0.8,1,10,machine,144.318,-16.714,1.288,0.9492
55
+ 3,1.2,1,10,machine,144.118,-16.787,1.282,0.9531
56
+ 4,1.6,1,10,machine,144.063,-16.693,1.276,0.957
57
+ 5,2.0,1,10,machine,144.009,-16.711,1.271,0.957
58
+ 6,2.4,1,10,machine,143.865,-16.99,1.262,0.957
59
+ 7,2.8,1,10,machine,144.008,-16.823,1.262,0.957
60
+ 8,3.2,1,10,machine,143.812,-16.783,1.254,0.957
61
+ 9,3.6,1,10,machine,143.812,-16.894,1.251,0.9531
62
+ 10,4.0,1,10,machine,143.812,-16.894,1.248,0.9531
63
+ 11,4.4,1,10,machine,143.759,-16.911,1.243,0.9531
64
+ 12,4.8,1,10,machine,143.759,-16.8,1.237,0.9492
65
+ 13,5.2,1,10,machine,143.759,-16.8,1.237,0.9492
66
+ 14,5.6,1,10,machine,143.759,-16.8,1.237,0.9453
67
+ 15,6.0,1,10,machine,143.759,-16.911,1.229,0.9414
68
+ 16,6.4,1,10,machine,143.759,-16.911,1.223,0.9414
69
+ 17,6.8,1,10,machine,143.902,-16.857,1.221,0.9375
70
+ 18,7.2,1,10,machine,143.759,-16.911,1.215,0.9375
71
+ 19,7.6,1,10,machine,143.902,-16.857,1.212,0.9336
72
+ 20,8.0,1,10,machine,143.707,-16.818,1.21,0.9297
73
+ 21,8.4,1,10,machine,143.565,-16.982,1.204,0.9258
74
+ 22,8.8,1,10,machine,143.707,-16.818,1.201,0.9219
75
+ 23,9.2,1,10,machine,143.707,-16.928,1.193,0.9219
76
+ 24,9.6,1,10,machine,143.707,-16.818,1.191,0.918
77
+ 25,10.0,1,10,machine,143.707,-16.818,1.188,0.9141
78
+ 26,10.4,1,10,machine,143.707,-16.818,1.185,0.9102
79
+ 27,10.8,1,10,machine,143.707,-16.818,1.177,0.9102
80
+ 28,11.2,1,10,machine,143.707,-16.818,1.174,0.9062
81
+ 29,11.6,1,10,machine,143.759,-16.8,1.172,0.9023
82
+ 30,12.0,1,10,machine,143.565,-16.65,1.169,0.8984
83
+ 31,12.4,1,10,machine,143.759,-16.689,1.164,0.8945
84
+ 32,12.8,1,10,machine,143.759,-16.689,1.166,0.8945
85
+ 33,13.2,1,10,machine,143.616,-16.632,1.158,0.8906
86
+ 34,13.6,1,10,machine,143.812,-16.56,1.158,0.8906
87
+ 35,14.0,1,10,machine,143.865,-16.541,1.153,0.8828
88
+ 36,14.4,1,10,machine,143.574,-16.649,1.15,0.8789
89
+ 37,14.8,1,10,machine,143.574,-16.536,1.145,0.875
90
+ 38,15.2,1,10,machine,143.531,-16.552,1.134,0.8711
91
+ 39,15.6,1,10,machine,143.434,-16.471,1.132,0.8633
92
+ 40,16.0,1,10,machine,143.486,-16.451,1.129,0.8594
93
+ 41,16.4,1,10,machine,143.339,-16.382,1.124,0.8516
94
+ 42,16.8,1,10,machine,143.13,-16.27,1.118,0.8438
95
+ 43,17.2,1,10,machine,142.86,-16.177,1.118,0.832
96
+ 44,17.6,1,10,machine,142.296,-15.981,1.118,0.8125
97
+ 45,18.0,1,10,machine,141.698,-15.903,1.118,0.7969
98
+ 46,18.4,1,10,machine,141.203,-15.651,1.118,0.7773
99
+ 47,18.8,1,10,machine,140.848,-15.729,1.121,0.7578
100
+ 48,19.2,1,10,machine,140.036,-15.463,1.129,0.7344
101
+ 49,19.6,1,10,machine,139.754,-15.466,1.126,0.7188
102
+ 0,0.0,2,10,machine,144.866,-16.965,1.254,0.5938
103
+ 1,0.4,2,10,machine,144.548,-17.07,1.234,0.6406
104
+ 2,0.8,2,10,machine,144.631,-17.252,1.223,0.6484
105
+ 3,1.2,2,10,machine,144.434,-17.322,1.218,0.6367
106
+ 4,1.6,2,10,machine,144.379,-17.227,1.212,0.625
107
+ 5,2.0,2,10,machine,144.184,-17.297,1.21,0.6055
108
+ 6,2.4,2,10,machine,144.184,-17.297,1.204,0.5859
109
+ 7,2.8,2,10,machine,144.184,-17.297,1.204,0.5703
110
+ 8,3.2,2,10,machine,143.99,-17.148,1.201,0.5547
111
+ 9,3.6,2,10,machine,143.99,-17.148,1.201,0.5312
112
+ 10,4.0,2,10,machine,143.99,-17.039,1.199,0.5117
113
+ 11,4.4,2,10,machine,143.99,-17.039,1.201,0.4961
114
+ 12,4.8,2,10,machine,143.99,-16.93,1.201,0.4746
115
+ 13,5.2,2,10,machine,144.043,-16.913,1.21,0.4609
116
+ 14,5.6,2,10,machine,143.955,-16.84,1.21,0.4414
117
+ 15,6.0,2,10,machine,143.812,-16.783,1.212,0.4238
118
+ 16,6.4,2,10,machine,143.865,-16.653,1.218,0.4043
119
+ 17,6.8,2,10,machine,143.918,-16.521,1.223,0.3926
120
+ 18,7.2,2,10,machine,143.679,-16.497,1.229,0.377
121
+ 19,7.6,2,10,machine,143.434,-16.471,1.237,0.3633
122
+ 20,8.0,2,10,machine,143.233,-16.306,1.248,0.3477
123
+ 21,8.4,2,10,machine,142.866,-16.189,1.26,0.332
124
+ 22,8.8,2,10,machine,142.645,-16.009,1.276,0.3242
125
+ 23,9.2,2,10,machine,142.028,-15.961,1.285,0.3145
126
+ 24,9.6,2,10,machine,141.318,-15.859,1.299,0.3086
127
+ 25,10.0,2,10,machine,140.494,-15.833,1.325,0.3008
128
+ 26,10.4,2,10,machine,139.612,-15.66,1.342,0.2949
129
+ 27,10.8,2,10,machine,138.549,-15.59,1.359,0.293
130
+ 28,11.2,2,10,machine,137.309,-15.524,1.377,0.291
131
+ 29,11.6,2,10,machine,135.951,-15.495,1.394,0.293
132
+ 30,12.0,2,10,machine,133.499,-15.361,1.424,0.293
133
+ 31,12.4,2,10,machine,131.618,-15.426,1.442,0.2988
134
+ 32,12.8,2,10,machine,130.458,-15.466,1.46,0.3027
135
+ 33,13.2,2,10,machine,128.539,-15.428,1.466,0.3066
136
+ 34,13.6,2,10,machine,125.983,-15.345,1.49,0.3145
137
+ 35,14.0,2,10,machine,123.98,-15.419,1.508,0.3223
138
+ 36,14.4,2,10,machine,121.112,-15.394,1.532,0.3379
139
+ 37,14.8,2,10,machine,119.526,-15.391,1.545,0.3496
140
+ 38,15.2,2,10,machine,116.259,-15.377,1.569,0.373
141
+ 39,15.6,2,10,machine,112.047,-15.472,1.582,0.3945
142
+ 40,16.0,2,10,machine,110.925,-15.361,1.594,0.4082
143
+ 41,16.4,2,10,machine,107.998,-15.494,1.607,0.4258
144
+ 42,16.8,2,10,machine,105.716,-15.335,1.613,0.4473
145
+ 43,17.2,2,10,machine,104.036,-15.209,1.632,0.457
146
+ 44,17.6,2,10,machine,101.736,-15.174,1.644,0.4707
147
+ 45,18.0,2,10,machine,99.992,-15.154,1.651,0.4727
148
+ 46,18.4,2,10,machine,98.342,-15.043,1.663,0.4727
149
+ 47,18.8,2,10,machine,96.194,-15.194,1.663,0.4707
150
+ 48,19.2,2,10,machine,94.511,-15.186,1.67,0.4688
151
+ 49,19.6,2,10,machine,92.848,-15.086,1.67,0.4531
152
+ 0,0.0,3,10,machine,144.574,-14.089,1.418,0.6602
153
+ 1,0.4,3,19,vehicle,144.349,-14.148,1.418,0.8203
154
+ 2,0.8,3,19,vehicle,143.881,-14.212,1.442,0.8633
155
+ 3,1.2,3,19,vehicle,143.13,-13.976,1.484,0.8711
156
+ 4,1.6,3,19,vehicle,141.988,-13.679,1.526,0.8711
157
+ 5,2.0,3,19,vehicle,140.262,-13.418,1.6,0.8594
158
+ 6,2.4,3,19,vehicle,138.094,-13.158,1.676,0.8477
159
+ 7,2.8,3,19,vehicle,136.58,-12.904,1.753,0.8359
160
+ 8,3.2,3,19,vehicle,135.201,-12.655,1.818,0.8242
161
+ 9,3.6,3,19,vehicle,133.772,-12.389,1.89,0.8086
162
+ 10,4.0,3,19,vehicle,132.94,-12.192,1.944,0.8008
163
+ 11,4.4,3,19,vehicle,132.51,-11.994,1.977,0.7891
164
+ 12,4.8,3,19,vehicle,132.075,-11.933,2.011,0.7773
165
+ 13,5.2,3,19,vehicle,131.298,-11.788,2.045,0.7695
166
+ 14,5.6,3,19,vehicle,130.96,-11.703,2.072,0.7578
167
+ 15,6.0,3,19,vehicle,130.618,-11.69,2.099,0.75
168
+ 16,6.4,3,19,vehicle,130.272,-11.53,2.12,0.7422
169
+ 17,6.8,3,13,breathing,130.156,-11.476,2.141,0.7383
170
+ 18,7.2,3,13,breathing,129.611,-11.414,2.155,0.7344
171
+ 19,7.6,3,13,breathing,129.688,-11.406,2.168,0.7266
172
+ 20,8.0,3,13,breathing,129.213,-11.407,2.182,0.7227
173
+ 21,8.4,3,13,breathing,129.094,-11.352,2.21,0.7188
174
+ 22,8.8,3,19,vehicle,128.9,-11.232,2.224,0.7188
175
+ 23,9.2,3,19,vehicle,128.733,-11.334,2.224,0.7148
176
+ 24,9.6,3,13,breathing,128.49,-11.297,2.238,0.7148
177
+ 25,10.0,3,13,breathing,128.245,-11.259,2.252,0.7148
178
+ 26,10.4,3,19,vehicle,127.93,-11.155,2.252,0.7148
179
+ 27,10.8,3,19,vehicle,127.999,-11.22,2.266,0.7188
180
+ 28,11.2,3,13,breathing,127.751,-11.182,2.266,0.7188
181
+ 29,11.6,3,19,vehicle,127.751,-11.106,2.28,0.7227
182
+ 30,12.0,3,19,vehicle,127.569,-11.055,2.294,0.7227
183
+ 31,12.4,3,19,vehicle,127.189,-11.033,2.294,0.7266
184
+ 32,12.8,3,19,vehicle,127.189,-11.033,2.294,0.7305
185
+ 33,13.2,3,19,vehicle,127.127,-11.04,2.308,0.7305
186
+ 34,13.6,3,19,vehicle,126.87,-10.999,2.322,0.7344
187
+ 35,14.0,3,19,vehicle,126.543,-11.043,2.322,0.7344
188
+ 36,14.4,3,19,vehicle,126.342,-11.068,2.322,0.7344
189
+ 37,14.8,3,19,vehicle,126.209,-11.007,2.336,0.7344
190
+ 38,15.2,3,19,vehicle,125.868,-11.051,2.336,0.7344
191
+ 39,15.6,3,19,vehicle,125.38,-11.033,2.35,0.7266
192
+ 40,16.0,3,19,vehicle,125.577,-11.085,2.35,0.7227
193
+ 41,16.4,3,19,vehicle,124.935,-11.167,2.364,0.7188
194
+ 42,16.8,3,19,vehicle,124.509,-11.057,2.364,0.7109
195
+ 43,17.2,3,19,vehicle,124.418,-11.148,2.379,0.6992
196
+ 44,17.6,3,19,vehicle,124.034,-11.194,2.393,0.6836
197
+ 45,18.0,3,19,vehicle,123.64,-11.241,2.393,0.668
198
+ 46,18.4,3,19,vehicle,123.389,-11.182,2.407,0.6562
199
+ 47,18.8,3,19,vehicle,122.978,-11.316,2.407,0.6484
200
+ 48,19.2,3,19,vehicle,122.514,-11.241,2.421,0.6367
201
+ 49,19.6,3,19,vehicle,122.358,-11.215,2.421,0.6211
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov3_000126__pred.csv ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,24,glass,44.376,3.79,4.264,0.8633
3
+ 1,0.3983,0,24,glass,44.753,3.097,4.233,0.8477
4
+ 2,0.7965,0,24,glass,44.756,2.358,4.203,0.8125
5
+ 3,1.1948,0,24,glass,44.523,1.529,4.141,0.7773
6
+ 4,1.593,0,24,glass,44.177,0.811,4.08,0.7148
7
+ 5,1.9913,0,24,glass,43.603,-0.025,3.987,0.5391
8
+ 6,2.3896,0,24,glass,43.025,-0.693,3.88,0.293
9
+ 7,2.7878,0,24,glass,42.643,-1.114,3.789,0.1211
10
+ 8,3.1861,0,24,glass,42.354,-1.509,3.697,0.0454
11
+ 9,3.5843,0,24,glass,42.262,-1.726,3.606,0.0179
12
+ 10,3.9826,0,24,glass,42.068,-1.869,3.53,0.0089
13
+ 11,4.3809,0,24,glass,41.864,-1.927,3.454,0.0049
14
+ 0,0.0,1,24,glass,46.068,3.915,4.264,0.8828
15
+ 1,0.3983,1,24,glass,46.169,3.282,4.233,0.9023
16
+ 2,0.7965,1,24,glass,46.037,2.647,4.233,0.8828
17
+ 3,1.1948,1,24,glass,45.794,1.864,4.172,0.8555
18
+ 4,1.593,1,24,glass,45.338,1.224,4.141,0.8086
19
+ 5,1.9913,1,24,glass,45.0,0.422,4.049,0.6641
20
+ 6,2.3896,1,24,glass,44.556,-0.194,3.987,0.416
21
+ 7,2.7878,1,24,glass,44.212,-0.696,3.926,0.1846
22
+ 8,3.1861,1,24,glass,43.625,-1.088,3.85,0.0718
23
+ 9,3.5843,1,24,glass,43.465,-1.315,3.758,0.0276
24
+ 10,3.9826,1,24,glass,43.544,-1.512,3.697,0.0128
25
+ 11,4.3809,1,24,glass,43.257,-1.584,3.621,0.0069
26
+ 0,0.0,2,24,glass,43.963,3.821,4.203,0.8242
27
+ 1,0.3983,2,24,glass,44.365,3.163,4.203,0.8047
28
+ 2,0.7965,2,24,glass,44.373,2.503,4.172,0.7578
29
+ 3,1.1948,2,24,glass,44.265,1.763,4.141,0.7188
30
+ 4,1.593,2,24,glass,43.915,1.119,4.08,0.668
31
+ 5,1.9913,2,24,glass,43.343,0.306,4.003,0.5352
32
+ 6,2.3896,2,24,glass,42.653,-0.34,3.911,0.332
33
+ 7,2.7878,2,24,glass,42.274,-0.817,3.819,0.1523
34
+ 8,3.1861,2,24,glass,41.874,-1.222,3.743,0.0618
35
+ 9,3.5843,2,24,glass,41.648,-1.436,3.651,0.0237
36
+ 10,3.9826,2,24,glass,41.44,-1.606,3.59,0.011
37
+ 11,4.3809,2,24,glass,41.342,-1.673,3.515,0.0061
38
+ 0,0.0,3,24,glass,47.212,2.474,4.018,0.9023
39
+ 1,0.3983,3,24,glass,46.809,1.811,3.957,0.9492
40
+ 2,0.7965,3,24,glass,46.184,1.088,3.896,0.9531
41
+ 3,1.1948,3,24,glass,45.462,0.296,3.819,0.9336
42
+ 4,1.593,3,24,glass,44.886,-0.303,3.758,0.875
43
+ 5,1.9913,3,24,glass,44.428,-0.945,3.682,0.6875
44
+ 6,2.3896,3,24,glass,43.963,-1.334,3.621,0.4141
45
+ 7,2.7878,3,24,glass,43.826,-1.608,3.56,0.1982
46
+ 8,3.1861,3,24,glass,43.684,-1.86,3.515,0.0913
47
+ 9,3.5843,3,24,glass,43.651,-1.983,3.439,0.042
48
+ 10,3.9826,3,24,glass,43.499,-2.099,3.394,0.0226
49
+ 11,4.3809,3,24,glass,43.472,-2.137,3.348,0.0132
checkpoints/spatial_beats_ov1_local_spatial_v7j_ov123_exp/03_ov123_top4/val_predictions/epoch_0001_csv/valid__ov3_000136__gt.csv ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ frame_idx,frame_time_s,src_or_track_idx,class_idx,class_name,azimuth_deg,elevation_deg,distance_m,activity_prob
2
+ 0,0.0,0,51,ocean,-11.172,-1.929,2.073,1.0
3
+ 1,0.4,0,51,ocean,-11.172,-1.929,2.073,1.0
4
+ 2,0.8,0,51,ocean,-11.172,-1.929,2.073,1.0
5
+ 3,1.2,0,51,ocean,-11.172,-1.929,2.073,1.0
6
+ 4,1.6,0,51,ocean,-11.172,-1.929,2.073,1.0
7
+ 5,2.0,0,51,ocean,-11.172,-1.929,2.073,1.0
8
+ 6,2.4,0,51,ocean,-11.172,-1.929,2.073,1.0
9
+ 7,2.8,0,51,ocean,-11.172,-1.929,2.073,1.0
10
+ 8,3.2,0,51,ocean,-11.172,-1.929,2.073,1.0
11
+ 9,3.6,0,51,ocean,-11.172,-1.929,2.073,1.0
12
+ 10,4.0,0,51,ocean,-11.172,-1.929,2.073,1.0
13
+ 11,4.4,0,51,ocean,-11.172,-1.929,2.073,1.0
14
+ 12,4.8,0,51,ocean,-11.172,-1.929,2.073,1.0
15
+ 13,5.2,0,51,ocean,-11.172,-1.929,2.073,1.0
16
+ 14,5.6,0,51,ocean,-11.172,-1.929,2.073,1.0
17
+ 15,6.0,0,51,ocean,-11.172,-1.929,2.073,1.0
18
+ 16,6.4,0,51,ocean,-11.172,-1.929,2.073,1.0
19
+ 17,6.8,0,51,ocean,-11.172,-1.929,2.073,1.0
20
+ 18,7.2,0,51,ocean,-11.172,-1.929,2.073,1.0
21
+ 19,7.6,0,51,ocean,-11.172,-1.929,2.073,1.0
22
+ 20,8.0,0,51,ocean,-11.172,-1.929,2.073,1.0
23
+ 21,8.4,0,51,ocean,-11.172,-1.929,2.073,1.0
24
+ 22,8.8,0,51,ocean,-11.172,-1.929,2.073,1.0
25
+ 23,9.2,0,51,ocean,-11.172,-1.929,2.073,1.0
26
+ 24,9.6,0,51,ocean,-11.172,-1.929,2.073,1.0
27
+ 25,10.0,0,51,ocean,-11.172,-1.929,2.073,1.0
28
+ 26,10.4,0,51,ocean,-11.172,-1.929,2.073,1.0
29
+ 27,10.8,0,51,ocean,-11.172,-1.929,2.073,1.0
30
+ 28,11.2,0,51,ocean,-11.172,-1.929,2.073,1.0
31
+ 29,11.6,0,51,ocean,-11.172,-1.929,2.073,1.0
32
+ 30,12.0,0,51,ocean,-11.172,-1.929,2.073,1.0
33
+ 31,12.4,0,51,ocean,-11.172,-1.929,2.073,1.0
34
+ 32,12.8,0,51,ocean,-11.172,-1.929,2.073,1.0
35
+ 33,13.2,0,51,ocean,-11.172,-1.929,2.073,1.0
36
+ 34,13.6,0,51,ocean,-11.172,-1.929,2.073,1.0
37
+ 35,14.0,0,51,ocean,-11.172,-1.929,2.073,1.0
38
+ 36,14.4,0,51,ocean,-11.172,-1.929,2.073,1.0
39
+ 37,14.8,0,51,ocean,-11.172,-1.929,2.073,1.0
40
+ 38,15.2,0,51,ocean,-11.172,-1.929,2.073,1.0
41
+ 39,15.6,0,51,ocean,-11.172,-1.929,2.073,1.0
42
+ 40,16.0,0,51,ocean,-11.172,-1.929,2.073,1.0
43
+ 41,16.4,0,51,ocean,-11.172,-1.929,2.073,1.0
44
+ 42,16.8,0,51,ocean,-11.172,-1.929,2.073,1.0
45
+ 43,17.2,0,51,ocean,-11.172,-1.929,2.073,1.0
46
+ 44,17.6,0,51,ocean,-11.172,-1.929,2.073,1.0
47
+ 45,18.0,0,51,ocean,-11.172,-1.929,2.073,1.0
48
+ 46,18.4,0,51,ocean,-11.172,-1.929,2.073,1.0
49
+ 47,18.8,0,51,ocean,-11.172,-1.929,2.073,1.0
50
+ 48,19.2,0,51,ocean,-11.172,-1.929,2.073,1.0
51
+ 49,19.6,0,51,ocean,-11.172,-1.929,2.073,1.0
52
+ 7,2.8,1,49,paper,70.448,11.18,4.001,1.0
53
+ 8,3.2,1,49,paper,70.448,11.18,4.001,1.0
54
+ 9,3.6,1,49,paper,70.448,11.18,4.001,1.0
55
+ 10,4.0,1,49,paper,70.448,11.18,4.001,1.0
56
+ 11,4.4,1,49,paper,70.448,11.18,4.001,1.0
57
+ 12,4.8,1,49,paper,70.448,11.18,4.001,1.0
58
+ 13,5.2,1,49,paper,70.448,11.18,4.001,1.0
59
+ 14,5.6,1,49,paper,70.448,11.18,4.001,1.0
60
+ 15,6.0,1,49,paper,70.448,11.18,4.001,1.0
61
+ 16,6.4,1,49,paper,70.448,11.18,4.001,1.0
62
+ 17,6.8,1,49,paper,70.448,11.18,4.001,1.0
63
+ 18,7.2,1,49,paper,70.448,11.18,4.001,1.0
64
+ 19,7.6,1,49,paper,70.448,11.18,4.001,1.0
65
+ 20,8.0,1,49,paper,70.448,11.18,4.001,1.0
66
+ 21,8.4,1,49,paper,70.448,11.18,4.001,1.0
67
+ 22,8.8,1,49,paper,70.448,11.18,4.001,1.0
68
+ 23,9.2,1,49,paper,70.448,11.18,4.001,1.0
69
+ 24,9.6,1,49,paper,70.448,11.18,4.001,1.0
70
+ 25,10.0,1,49,paper,70.448,11.18,4.001,1.0
71
+ 26,10.4,1,49,paper,70.448,11.18,4.001,1.0
72
+ 27,10.8,1,49,paper,70.448,11.18,4.001,1.0
73
+ 28,11.2,1,49,paper,70.448,11.18,4.001,1.0
74
+ 18,7.2,2,60,crack,-108.185,14.847,3.81,1.0
75
+ 19,7.6,2,60,crack,-108.185,14.847,3.81,1.0
76
+ 20,8.0,2,60,crack,-108.185,14.847,3.81,1.0
77
+ 21,8.4,2,60,crack,-108.185,14.847,3.81,1.0
docs/README_DOCUMENTATION_INDEX.md ADDED
@@ -0,0 +1,274 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Documentation Index — Spatial-BEATs Analysis & Reference
2
+
3
+ ## 📋 Quick Navigation
4
+
5
+ ### **New to the codebase?**
6
+ → Start here: [`SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md`](SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md) (5 min read)
7
+
8
+ ### **Debugging a train/validation gap in DOA?**
9
+ → Go here: [`doa_train_valid_gap_analysis.md`](doa_train_valid_gap_analysis.md) → Part 6 + Part 8
10
+
11
+ ### **Need detailed architecture reference?**
12
+ → Read here: [`SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md`](SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md) (deep dive, 30 min)
13
+
14
+ ### **Planning experiments (v11 series)?**
15
+ → Use: [`0427_v11_series.md`](0427_v11_series.md) + `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md` (v11 section)
16
+
17
+ ### **Understanding specific component (e.g., SourceQueryDecoder)?**
18
+ → Use: Search in `SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md` Part 2 + Appendix for line numbers
19
+
20
+ ---
21
+
22
+ ## 📚 Document Reference Table
23
+
24
+ | Document | Lines | Size | Best For | Read Time |
25
+ |----------|-------|------|----------|-----------|
26
+ | **ANALYSIS_COMPLETION_SUMMARY** (this index) | 150 | 6KB | Overview + navigation | 5 min |
27
+ | **SPATIAL_FRAMEWORKS_QUICK_REFERENCE** | 192 | 7KB | Quick lookup, practitioner guide | 5-10 min |
28
+ | **SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS** | 724 | 28KB | Deep technical reference | 30-45 min |
29
+ | **doa_train_valid_gap_analysis** | 434 | 19KB | Diagnostics + fixes | 20-30 min |
30
+ | **0427_v11_series** | 185 | 13KB | Experimental design (v11a/b/c/d) | 15 min |
31
+ | **spatial_beats_ov123_frame_routes** | 512 | 22KB | Routes A/B/C architecture | 25 min |
32
+ | **spatial_beats_training_overview** | 420 | 15KB | Training pipeline + presets | 20 min |
33
+
34
+ ---
35
+
36
+ ## 🎯 Use Case Lookup
37
+
38
+ ### "I need to understand the DOA train/val gap"
39
+ 1. Read: [`doa_train_valid_gap_analysis.md`](doa_train_valid_gap_analysis.md) **Executive Summary** (2 min)
40
+ 2. Identify: Which of 6 mechanisms applies to your case (Part 6)
41
+ 3. Fix: Follow priority order in Part 8
42
+ 4. Reference: Code locations in Appendix
43
+
44
+ **Expected outcome**: Root cause identified + fix strategy
45
+
46
+ ---
47
+
48
+ ### "I'm new and want to understand the architecture"
49
+ 1. Read: [`SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md`](SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md) **Sections 1-3** (5 min)
50
+ 2. Read: [`SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md`](SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md) **Part 1 + Part 2** (15 min)
51
+ 3. Reference: Code locations in Appendix for specific functions
52
+ 4. Cross-check: `spatial_beats_ov123_frame_routes.md` for Routes A/B/C details
53
+
54
+ **Expected outcome**: High-level understanding + ability to navigate code
55
+
56
+ ---
57
+
58
+ ### "I want to run v11a experiment"
59
+ 1. Read: [`0427_v11_series.md`](0427_v11_series.md) **Section 2.2 (v11a)** (5 min)
60
+ 2. Check: `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md` **v11 section** for shell script
61
+ 3. Read: Part 4 (verification method) for how to evaluate results
62
+ 4. Reference: Appendix in `doa_train_valid_gap_analysis.md` for code line numbers if modifying
63
+
64
+ **Expected outcome**: Experiment ready to launch, understanding of what to expect
65
+
66
+ ---
67
+
68
+ ### "What are all the spatial frameworks in this codebase?"
69
+ 1. Read: [`SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md`](SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md) **Part 1** (5 min)
70
+ 2. Summary: Four frameworks implemented (Spatial-AST, DCASE SELD, EINV2, DETR-slots)
71
+ 3. Reference: Part 2 shows how each is implemented as Routes A/B/C
72
+
73
+ **Expected outcome**: Framework inventory + where each is implemented
74
+
75
+ ---
76
+
77
+ ### "How do I compare Routes A/B/C?"
78
+ 1. Go to: [`SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md`](SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md) **Part 2** (15 min)
79
+ 2. Check: Comparison table in `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md`
80
+ 3. Deep dive: `spatial_beats_ov123_frame_routes.md` for architectural details
81
+
82
+ **Expected outcome**: Understanding of paradigm differences, when to use each
83
+
84
+ ---
85
+
86
+ ### "What changed from v7 to v11?"
87
+ 1. Read: [`SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md`](SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md) **Part 3** (10 min)
88
+ 2. Reference: `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md` **version series** for quick compare
89
+ 3. Deep dive: `doa_train_valid_gap_analysis.md` **Part 3** for v9/v10 details
90
+ 4. Experimental: `0427_v11_series.md` **Section 1** for v11 rationale
91
+
92
+ **Expected outcome**: Version history + innovation tracking
93
+
94
+ ---
95
+
96
+ ### "Where exactly is the direction head loss computed?"
97
+ 1. Go to: `SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md` **Appendix** (search "direction loss")
98
+ 2. Result: `spatial_loss.py:1562-1565`
99
+ 3. Read: Part of `doa_train_valid_gap_analysis.md` **Part 2.3** for context
100
+
101
+ **Expected outcome**: Exact code location + understanding of loss formulation
102
+
103
+ ---
104
+
105
+ ## 🔍 Code Navigation Quick Reference
106
+
107
+ ### For Each Major Component
108
+
109
+ | Component | Primary Ref | Backup Ref | Concept |
110
+ |-----------|------------|-----------|---------|
111
+ | **LocalSpatialEncoder** | ANALYSIS Part 6 | QUICK_REF "Key locations" | 7-channel FOA → spatial features |
112
+ | **SourceQueryDecoder** | ROUTES p.20-30 | ANALYSIS Part 2.2b | K track queries → per-frame features |
113
+ | **FrameTrackPredictionHeads** | QUICK_REF Appendix | ANALYSIS Part 2.2b | Predicts act/class/dir/dist per frame |
114
+ | **Hungarian Matching** | DOA_GAP Part 2.2 | ANALYSIS Part 4 | How sources get assigned to slots/queries |
115
+ | **ClassHeadSpectralDemixer** | ANALYSIS Part 3 | QUICK_REF "v9 baseline" | Breaks frequency pooling bottleneck |
116
+ | **ACCDOA heads** | ROUTES p.40+ | ANALYSIS Part 1.2 | Route C: per-class 3D vectors |
117
+ | **SpecAugment** | DOA_GAP Part 1.2 | QUICK_REF "Training" | Spectral masking (train-only!) |
118
+
119
+ ---
120
+
121
+ ## 📊 Experiment Planning Matrix
122
+
123
+ ### To understand **which experiment tests what**:
124
+
125
+ Use: `0427_v11_series.md` Section 2 (detailed specifications)
126
+ + `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md` v11 table (quick reference)
127
+ + `doa_train_valid_gap_analysis.md` Part 6 (root causes)
128
+
129
+ | Problem | Experiment | Mechanism Tested | Expected Impact | Doc Reference |
130
+ |---------|-----------|-----------------|-----------------|----------------|
131
+ | ov2 angle errors (73.9%) | v11a | DOA demixer | ↓ 5pp+ | 0427_v11_series.md:18-40 |
132
+ | ov2/ov3 angles | v11b | IV signal path | Compare vs v11a | 0427_v11_series.md:41-63 |
133
+ | ov3 binding (24.5%) | v11c | ACCDOA paradigm | ↓ 5pp+ | 0427_v11_series.md:64-87 |
134
+ | ov1 ranking (37% loss) | v11d | Post-hoc decoding | ↑ 5pp+ | 0427_v11_series.md:88-112 |
135
+
136
+ ---
137
+
138
+ ## ⚡ Critical Findings Summary
139
+
140
+ ### Three Most Important Things to Know
141
+
142
+ 1. **Zero spatial augmentation (rotations)** is the #1 cause of DOA train/val gap
143
+ - Location: `doa_train_valid_gap_analysis.md` **Executive Summary** + Part 6
144
+ - Impact: 40-60% of variance
145
+ - Fix: See Part 8 recommendation #1
146
+
147
+ 2. **Three parallel routes (A/B/C) coexist in the codebase**
148
+ - Route A: Per-frame slot allocation (unstable)
149
+ - Route B: Learnable track queries (production, v9)
150
+ - Route C: Per-class vectors (prototype, being tested in v11c)
151
+ - Reference: `SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md` Part 2
152
+
153
+ 3. **v10 phase-1 freezes the direction head entirely**
154
+ - Direction head gets no gradients for 10 epochs
155
+ - Then unfrozen with poor initialization
156
+ - Causes 30-40% DOA metric drop on multi-source splits
157
+ - Reference: `doa_train_valid_gap_analysis.md` Part 3.2 + Part 6
158
+
159
+ ---
160
+
161
+ ## 🔗 Cross-Reference Guide
162
+
163
+ ### "I'm reading X, how do I find related content?"
164
+
165
+ | Reading | See Also |
166
+ |---------|----------|
167
+ | ROUTES page 20-30 (Route B) | ANALYSIS Part 2.2b, QUICK_REF "Routes", DOA_GAP Part 2-3 |
168
+ | ANALYSIS Part 1 (Spatial-AST) | QUICK_REF table, ANALYSIS Part 6 (code locations) |
169
+ | DOA_GAP Part 8 (fixes) | ANALYSIS Part 6 (line numbers), QUICK_REF "Training" |
170
+ | 0427_v11_series Section 2 (v11 specs) | QUICK_REF v11 table, DOA_GAP Part 6 (why these experiments) |
171
+ | TRAINING_OVERVIEW | ANALYSIS Part 5 (configs), QUICK_REF "Loss weights" |
172
+
173
+ ---
174
+
175
+ ## 📝 How Documents Were Created
176
+
177
+ All documentation created from **comprehensive codebase analysis** on 2026-04-27:
178
+ - ✅ 4 spatial frameworks identified (Spatial-AST, DCASE SELD, EINV2, DETR-slots)
179
+ - ✅ 3 parallel routes analyzed with full architecture specifications
180
+ - ✅ 6 mechanisms causing DOA train/val gaps discovered
181
+ - ✅ v7→v11 experimental series mapped with root cause tracing
182
+ - ✅ ~10,000+ lines of code reviewed and cross-referenced
183
+ - ✅ All findings tied to exact file:line numbers
184
+
185
+ **Quality Assurance**:
186
+ - Code references verified with actual line numbers
187
+ - Architecture descriptions validated against source
188
+ - Experimental hypotheses cross-checked with docstrings
189
+ - Cross-document consistency checked
190
+
191
+ ---
192
+
193
+ ## 🚀 Recommended Reading Order
194
+
195
+ ### **For Different Roles**
196
+
197
+ #### **Principal Investigator / Project Lead**
198
+ 1. This index + section on "Critical Findings" (5 min)
199
+ 2. `ANALYSIS_COMPLETION_SUMMARY.md` (10 min)
200
+ 3. `0427_v11_series.md` Section 1 (diagnostics review) + Section 5 (order) (10 min)
201
+ → **Decision**: Approve v11 experiments? (Total: 25 min)
202
+
203
+ #### **Researcher / Experimenter**
204
+ 1. `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md` (10 min)
205
+ 2. `0427_v11_series.md` full document (15 min)
206
+ 3. `SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md` Part 2 + 3 (25 min)
207
+ → **Ready**: Design and run experiments (Total: 50 min)
208
+
209
+ #### **New Contributor / Intern**
210
+ 1. `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md` (10 min)
211
+ 2. `SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md` Part 1 + 2 (25 min)
212
+ 3. Pick a component (e.g., "LocalSpatialEncoder") → find in Part 6 → read code
213
+ 4. `spatial_beats_training_overview.md` or `spatial_beats_coding_guide.md` as needed
214
+ → **Goal**: Understand codebase (Total: 1-2 hours)
215
+
216
+ #### **Debugging Train/Val Gap**
217
+ 1. `doa_train_valid_gap_analysis.md` Executive Summary + Part 6 (10 min)
218
+ 2. Part 7 (diagnostics) — check your logs (10 min)
219
+ 3. Part 8 (fixes) — pick priority #1-3 (5 min)
220
+ 4. Appendix — get code locations (5 min)
221
+ → **Goal**: Root cause + fix strategy (Total: 30 min)
222
+
223
+ ---
224
+
225
+ ## 📞 FAQ About Documentation
226
+
227
+ **Q: Where do I find the code for Route A, B, or C?**
228
+ A: See `SPATIAL_AUDIO_FRAMEWORKS_ANALYSIS.md` Part 2.1/2.2/2.3, each has exact file:line references in Part 6 Appendix.
229
+
230
+ **Q: What's the difference between v9 and v10?**
231
+ A: `doa_train_valid_gap_analysis.md` Part 3.1 vs 3.2, or quick summary in `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md`.
232
+
233
+ **Q: Should I implement fix #1, #2, or #3?**
234
+ A: Depends on your problem. See `doa_train_valid_gap_analysis.md` Part 6 — rank root causes by severity against your gap size.
235
+
236
+ **Q: How long will v11 experiments take?**
237
+ A: ~14 days total. See `0427_v11_series.md` Section 5 for recommended order (serial vs parallel).
238
+
239
+ **Q: Can I run v11 without understanding everything?**
240
+ A: Yes! Copy the shell script from `SPATIAL_FRAMEWORKS_QUICK_REFERENCE.md`, follow Part 4 (verification method) in `0427_v11_series.md`.
241
+
242
+ ---
243
+
244
+ ## ✅ Checklist: What You Can Do Now
245
+
246
+ After reading appropriate documentation:
247
+
248
+ - [ ] Understand what spatial frameworks exist in codebase
249
+ - [ ] Identify which route (A/B/C) solves your problem
250
+ - [ ] Diagnose your train/val gap (Part 6 in DOA_GAP)
251
+ - [ ] Plan an experiment (v11 specs + order)
252
+ - [ ] Find exact code locations (Appendix tables)
253
+ - [ ] Understand loss weight patterns (QUICK_REF tables)
254
+ - [ ] Know when to hot-start from which checkpoint
255
+ - [ ] Compare validation metrics across routes
256
+
257
+ ---
258
+
259
+ ## 📄 License & Citation
260
+
261
+ These documents are analysis artifacts for internal research use. They synthesize information from:
262
+ - Source code comments and docstrings
263
+ - Configuration file specifications
264
+ - DCASE challenge documentation (officially referenced in code)
265
+ - Research paper citations in docstrings
266
+
267
+ ---
268
+
269
+ **Last Updated**: 2026-04-27
270
+ **Analysis Completed**: Yes
271
+ **Ready for Use**: Yes
272
+ **Maintenance**: Update after v11 experiments complete
273
+
274
+ For questions or clarifications, refer to exact file:line citations in the appendices of technical documents.
docs/train_valid_gap_diagram.txt ADDED
@@ -0,0 +1,204 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ╔═══════════════════════════════════════════════════════════════════════════════════════════════╗
2
+ ║ SPATIAL-BEATS TRAINING VS VALIDATION DATA FLOW ║
3
+ ║ (SHOWING THE ROOT CAUSES OF DOA GAP) ║
4
+ ╚═══════════════════════════════════════════════════════════════════════════════════════════════╝
5
+
6
+ TRAINING PATH (epoch T) VALIDATION PATH (epoch T)
7
+ ════════════════════════════════════════════════════════════════════════════════════════════════
8
+
9
+ ┌─────────────────────────────────┐ ┌──────────────────────────────┐
10
+ │ FOA Waveform [4, T] │ │ FOA Waveform [4, T] │
11
+ │ (W, X, Y, Z channels) │ │ (W, X, Y, Z channels) │
12
+ └──────────────┬──────────────────┘ └────────────┬─────────────────┘
13
+ │ │
14
+ ▼ ─────────────────────────────────────────► ▼
15
+ ╔══════════════════════╗ ╔═══════════════════╗
16
+ │ PREPROCESSING │ │ PREPROCESSING │
17
+ ├──────────────────────┤ ├───────────────────┤
18
+ │ • Normalize (same) │ │ • Normalize (same)│
19
+ │ • Mel filterbank (OK)│ │ • Mel filterbank │
20
+ │ • FOA features (OK) │ │ (OK) │
21
+ └──────────┬───────────┘ └─────────┬─────────┘
22
+ │ │
23
+ ⚠️ ISSUE #1 ◄──┘─────────────────────────────────────────┘ ⚠️ ISSUE #1
24
+
25
+ SpecAugment: SpecAugment:
26
+ ├─ freq_masks=2×27 ├─ NO MASKING
27
+ ├─ time_masks=2×100 ├─ w_logmel unchanged
28
+ └─ Applied to W channel └─ Returns w_logmel as-is
29
+
30
+ Result: ~20-50% of spec erased Result: Full pristine spectrum
31
+
32
+ │ │
33
+ ▼ ─────────────────────────────────────────► ▼
34
+ ╔══════════════════════╗ ╔═══════════════════╗
35
+ │ ENCODER (Frozen) │ │ ENCODER (Eval) │
36
+ │ • BEATs backbone │ │ • BEATs backbone │
37
+ │ • SpatialFuser │ │ • SpatialFuser │
38
+ │ • LocalSpatialEnc │ │ • LocalSpatialEnc │
39
+ └──────────┬───────────┘ └─────────┬─────────┘
40
+ │ │
41
+ [B, T_s, D] [B, T_s, D]
42
+ │ │
43
+ ▼ ▼
44
+ ╔══════════════════════╗ ╔═══════════════════╗
45
+ │ PREDICTION HEADS │ │ PREDICTION HEADS │
46
+ ├──────────────────────┤ ├───────────────────┤
47
+ │ activity_head │ │ activity_head │
48
+ │ class_head │ │ class_head │
49
+ │ direction_head │ │ direction_head │
50
+ │ distance_head │ │ distance_head │
51
+ └──────────┬───────────┘ └─────────┬─────────┘
52
+ │ │
53
+ │ pred_activity [B, T_s, K] │
54
+ │ pred_class [B, T_s, K, C] │
55
+ │ pred_dir [B, T_s, K, 3] │ (STALE in v10p1)
56
+ │ pred_dist [B, T_s, K] │
57
+ │ │
58
+ ▼ ▼
59
+ ╔══════════════════════╗ ╔═══════════════════╗
60
+ │ HUNGARIAN MATCHING │ │ HUNGARIAN MATCHING│
61
+ │ (per-frame) │ │ (per-frame) │
62
+ ├──────────────────────┤ ├───────────────────┤
63
+ │ For each (b, t): │ │ For each (b, t): │
64
+ │ cost = act_cost + │ │ cost = act_cost +│
65
+ │ cls_nll + │ │ cls_nll + │
66
+ │ dir_cost + │ │ dir_cost +│
67
+ │ dist_cost │ │ dist_cost │
68
+ │ │ │ │
69
+ ⚠️ │ ISSUE #4: Fixed cost │ ────────────────── │ ISSUE #4: But │
70
+ │ weights (1.0 each) │ (v10p1: matches │ loss uses lambda │
71
+ │ but loss uses │ on stale DOA) │ weights (0.0) │
72
+ │ variable lambda │ │ │
73
+ └──────────┬───────────┘ └─────────┬─────────┘
74
+ │ │
75
+ matched_slot matched_slot
76
+ [B, N_gt, T_s] [B, N_gt, T_s]
77
+ │ │
78
+ ▼ ▼
79
+ ╔══════════════════════╗ ╔═══════════════════╗
80
+ │ LOSS COMPUTATION │ │ METRICS COMPUTATION
81
+ ├──────────────────────┤ ├───────────────────┤
82
+ │ loss_activity × │ │ For matched pairs: │
83
+ │ λ_frame_activity │ │ pred_azi_deg │
84
+ │ loss_class × │ │ pred_ele_deg │
85
+ │ λ_frame_class │ │ tgt_azi_deg │
86
+ │ loss_direction × ◄─ v10p1: 0.0 │ tgt_ele_deg │
87
+ │ λ_frame_direction │ │ │
88
+ │ loss_distance × ◄─ v10p1: 0.0 │ azi_mae = MAE │
89
+ │ λ_frame_distance │ │ ele_mae = MAE │
90
+ └──────────┬───────────┘ └─────────┬─────────┘
91
+ │ │
92
+ ⚠️ ISSUE #2 │ v9: Full training │
93
+ ────────────► │ v10 phase-1: FROZEN dir/dist head │
94
+ │ no gradients! │
95
+ │ │
96
+ ▼ ▼
97
+ loss_total azi_mae, ele_mae
98
+ (backprop) (forward only)
99
+
100
+
101
+ KEY ASYMMETRIES CAUSING TRAIN/VALID GAP
102
+ ════════════════════════════════════════════════════════════════════════════════════════════════
103
+
104
+ ┌─────────────────────────────────┬──────────────────────────┬──────────────┐
105
+ │ Aspect │ Training │ Validation │
106
+ ├─────────────────────────────────┼──────────────────────────┼──────────────┤
107
+ │ SpecAugment │ Masking 45% of spectrum │ NO masking │
108
+ │ │ →音频特性不同 │
109
+ ├─────────────────────────────────┼──────────────────────────┼──────────────┤
110
+ │ Spatial Augmentation (Rotation) │ NONE │ NONE │
111
+ │ │ → DOA diversity limited │ → Unseen │
112
+ │ │ to ~1 epoch │ angles │
113
+ ├─────────────────────────────────┼──────────────────────────┼──────────────┤
114
+ │ Direction Head State (v10p1) │ FROZEN (no gradients) │ FROZEN + │
115
+ │ │ λ=0.0 (no loss signal) │ STALE (from │
116
+ │ │ │ v9 epoch 3)│
117
+ ├─────────────────────────────────┼──────────────────────────┼──────────────┤
118
+ │ Matching Cost Weights │ Unweighted sum │ Same unweighted
119
+ │ │ act + cls + dir + dist │ sum, but on │
120
+ │ │ (but loss uses lambda) │ frozen DOA │
121
+ ├─────────────────────────────────┼──────────────────────────┼──────────────┤
122
+ │ Gradient Flow │ Through loss (v9) or │ Validation │
123
+ │ │ None (v10p1) │ mode: no │
124
+ │ │ │ gradients │
125
+ └─────────────────────────────────┴──────────────────────────┴──────────────┘
126
+
127
+
128
+ IMPACT CASCADE (Most Severe Scenario: ov3, v10 Phase-1)
129
+ ════════════════════════════════════════════════════════════════════════════════════════════════
130
+
131
+ 1. Training sees only ~single orientation per azimuth per epoch
132
+ No rotation aug → limited DOA diversity
133
+
134
+ ├─→ Model sees: (azi=0°, ele=0°), (azi=1°, ele=1°), ...
135
+ └─→ Model learns: limited generalization to unseen angles
136
+
137
+ 2. Direction head frozen at v10 phase-1 start
138
+ λ_frame_direction = 0.0 → no training signal
139
+ freeze_frame_track_spatial_heads = True → no gradients
140
+
141
+ ├─→ Head receives: ZERO gradients from loss
142
+ ├─→ Head receives: ZERO gradients from backward pass
143
+ └─→ Result: Output completely stale (from v9 epoch 3)
144
+
145
+ 3. At validation: direction head outputs used to compute metrics
146
+ But outputs are unchanged from 10 epochs ago
147
+ And model hasn't learned ov2/ov3 multi-source DOA at all
148
+
149
+ ├─→ Validation sees: unseen azimuth combos (not in training)
150
+ ├─→ Validation sees: multi-source situations unfamiliar to frozen head
151
+ └─→ Result: 20-30° azimuth errors
152
+
153
+ 4. SpecAugment makes training acoustic features different
154
+ Masking 45% of W channel during training
155
+ Zero masking at validation
156
+
157
+ └─→ Result: +2-5° additional error on top
158
+
159
+
160
+ FIX #1: Add Rotation Augmentation
161
+ ════════════════════════════════════════════════════════════════════════════════════════════════
162
+
163
+ Current (BROKEN):
164
+ FOA [W, X, Y, Z] → (no change) → Encoder → DOA predictor
165
+
166
+ Fixed (CORRECT):
167
+ FOA [W, X, Y, Z] → Rotate(θ, φ, ψ) → [W', X', Y', Z']
168
+ → Encoder → DOA predictor
169
+
170
+ Labels: (azi, ele, dist) → Rotate_targets(θ, φ, ψ) → (azi', ele', dist)
171
+
172
+ Result: Same class, different DOA → model learns generalization
173
+
174
+
175
+ FIX #2: Make SpecAugment Train/Val Consistent
176
+ ═════════════════════════════════���══════════════════════════════════════════════════════════════
177
+
178
+ Current (BROKEN):
179
+ Training: if self.training: apply_masking() ✓
180
+ Validation: if not self.training: skip_masking() ✗ (mismatch!)
181
+
182
+ Fixed (OPTION A):
183
+ Training: No masking (simplified, possibly less robust)
184
+ Validation: No masking (consistent)
185
+
186
+ Fixed (OPTION B):
187
+ Training: apply_masking() ✓
188
+ Validation: apply_same_masking_with_fixed_seed() ✓ (consistent)
189
+
190
+
191
+ FIX #3: Don't Fully Freeze Direction Head (v10)
192
+ ════════════════════════════════════════════════════════════════════════════════════════════════
193
+
194
+ Current v10 (BROKEN):
195
+ λ_frame_direction = 0.0
196
+ freeze_frame_track_spatial_heads = True
197
+ → Head receives zero signal for 10 epochs
198
+
199
+ Fixed (BETTER):
200
+ λ_frame_direction = 0.1 (not zero!)
201
+ freeze_frame_track_spatial_heads = False
202
+ → Head receives weak signal to fine-tune on multi-source
203
+ → Prevents complete staleness at v10 unfreezing
204
+
docs/v13e_spatial_beats_design.md ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # v13_E 设计文档 — 调整后的定位
2
+
3
+ > 日期:2026-05-02 (调整版)
4
+ > 状态:**暂不运行。v13_D 已在跑且表现超预期(ep10 F=0.402),先等 v13_D 跑完 25 epoch**
5
+ > 定位:v13_D 之后的**扩展实验**(可能的 v13_F),而非替代
6
+
7
+ ---
8
+
9
+ ## 重要前提(来自 postmortem)
10
+
11
+ 之前此文档写的是 "v13_D 崩溃,v13_E 抢救"。**这个前提错了。** v13_D 在 ep8 spatial loss ramp 放开后 F 从 0.19 跳到 0.40,ep10 已经超过 v12 best (0.378) 并仍在上涨。v13_D 是目前最强的实验。
12
+
13
+ v13_E 的实际价值是 **在 v13_D 之后再加一层**:启用 num_active head 的训练 + 让 SELD official evaluator 用 top-K̂ gate。这是 v13_D **没有**做的改动,理论上正交。
14
+
15
+ ---
16
+
17
+ ## v13_E 设计(保持不变,但动机重写)
18
+
19
+ ### 改动
20
+
21
+ - **[E-1]** cls warmup 8 epochs + 20 total epochs
22
+ - **[E-2]** 启用 num_active head 训练(`lambda_frame_num_active = 0.5`)+ 推理 top-K̂ gate
23
+
24
+ ### 和 v13_D 的关系
25
+
26
+ | 改动 | v13_D | v13_E |
27
+ |---|---|---|
28
+ | cls warmup 长度 | 8 | 8 |
29
+ | cosine LR | ✓ | ✗ |
30
+ | Top-K rank loss | ✓ | ✗ |
31
+ | EMA | ✓ | ✗ |
32
+ | resume optimizer | ✓ | ✗ |
33
+ | num_active head 训练 | ✗ | **✓** |
34
+ | SELD top-K̂ gate | ✗ | **✓** |
35
+
36
+ v13_E 是一个**剥离了 v13_D 的 loss/schedule 改动、只加 num_active 的对照组**。不能直接比 "哪个 F 高",而是回答:
37
+
38
+ > **在同样 8 ep warmup 下,num_active head + top-K̂ gate 单独带来多少 F?**
39
+
40
+ 如果 v13_E 最终 F > v13_D,说明 num_active 是正贡献,应该在 v13_F 里合并;反之说明它和 v13_D 的 Top-K rank 路径冲突。
41
+
42
+ ---
43
+
44
+ ## 何时运行 v13_E
45
+
46
+ **先看 v13_D 最终结果**:
47
+ 1. 如果 v13_D 稳定在 0.43 以上 → 写 v13_F = v13_D + num_active 做融合
48
+ 2. 如果 v13_D 最终不到 0.40 → 跑 v13_E 看 num_active 单独能达到多少
49
+ 3. 如果算力充足 → 现在就可以跑 v13_E 做独立 ablation(对照组)
50
+
51
+ ---
52
+
53
+ ## 预期(基于 v13_D 实测)
54
+
55
+ | 指标 | v12 best | v13_D ep10 | v13_E 预期 |
56
+ |---|---|---|---|
57
+ | F20 | 0.378 | 0.402 | 0.39~0.42 |
58
+ | aR | 0.126 | 0.069 | 0.13~0.18 |
59
+
60
+ v13_E 的 **aR 预期比 v13_D 高**(num_active 补漏报),但 **F 可能持平或略低**(没有 Top-K rank + cosine + EMA 的增益)。
61
+
62
+ ---
63
+
64
+ ## 代码状态
65
+
66
+ - ✅ `spatial_loss.py` — `_build_frame_track_official_segment_dicts` 加 `use_num_active_gate` 参数
67
+ - ✅ `train_spatial_beats.py` — `make_ov1_unified_v13e_config` 已注册,CLI dispatch OK
68
+ - ✅ `run_ov1_unified_v13e.sh` — 可运行
69
+ - ✅ smoke test 通过
70
+
71
+ ## 不影响其它实验
72
+
73
+ 所有改动通过 config flag,默认 False。v12/v13_B/v13_C/v13_D 不受影响,照旧可跑。
74
+
75
+ ## 参考
76
+
77
+ - 主要错误诊断记录:`docs/v13_honest_postmortem.md`
78
+ - v13_D 设计:`docs/v13d_spatial_beats_design.md`
eval_voxaudio_ood_results/flow2gan/per_sample.json ADDED
@@ -0,0 +1,2161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "s1_rank0_opensource_emilia_zh_rawdata_English_003050": {
3
+ "gt_top_class": "speech",
4
+ "rc_top_class": "speech",
5
+ "gt_frames_preview": [
6
+ [
7
+ {
8
+ "track": 0,
9
+ "activity": 0.573,
10
+ "class_idx": 16,
11
+ "class_name": "speech",
12
+ "class_conf": 0.45,
13
+ "azi_deg": -42.47,
14
+ "ele_deg": 43.03,
15
+ "dist_m": 1.291
16
+ }
17
+ ],
18
+ [
19
+ {
20
+ "track": 0,
21
+ "activity": 0.779,
22
+ "class_idx": 16,
23
+ "class_name": "speech",
24
+ "class_conf": 0.503,
25
+ "azi_deg": -42.38,
26
+ "ele_deg": 42.74,
27
+ "dist_m": 1.258
28
+ }
29
+ ],
30
+ [
31
+ {
32
+ "track": 0,
33
+ "activity": 0.784,
34
+ "class_idx": 16,
35
+ "class_name": "speech",
36
+ "class_conf": 0.523,
37
+ "azi_deg": -42.73,
38
+ "ele_deg": 42.45,
39
+ "dist_m": 1.252
40
+ }
41
+ ],
42
+ [
43
+ {
44
+ "track": 0,
45
+ "activity": 0.771,
46
+ "class_idx": 16,
47
+ "class_name": "speech",
48
+ "class_conf": 0.549,
49
+ "azi_deg": -43.1,
50
+ "ele_deg": 42.33,
51
+ "dist_m": 1.256
52
+ }
53
+ ],
54
+ [
55
+ {
56
+ "track": 0,
57
+ "activity": 0.756,
58
+ "class_idx": 16,
59
+ "class_name": "speech",
60
+ "class_conf": 0.568,
61
+ "azi_deg": -43.15,
62
+ "ele_deg": 42.18,
63
+ "dist_m": 1.258
64
+ }
65
+ ]
66
+ ],
67
+ "rc_frames_preview": [
68
+ [
69
+ {
70
+ "track": 0,
71
+ "activity": 0.78,
72
+ "class_idx": 16,
73
+ "class_name": "speech",
74
+ "class_conf": 0.409,
75
+ "azi_deg": 54.02,
76
+ "ele_deg": 24.06,
77
+ "dist_m": 1.318
78
+ }
79
+ ],
80
+ [
81
+ {
82
+ "track": 0,
83
+ "activity": 0.936,
84
+ "class_idx": 16,
85
+ "class_name": "speech",
86
+ "class_conf": 0.474,
87
+ "azi_deg": 52.83,
88
+ "ele_deg": 26.9,
89
+ "dist_m": 1.257
90
+ }
91
+ ],
92
+ [
93
+ {
94
+ "track": 0,
95
+ "activity": 0.946,
96
+ "class_idx": 16,
97
+ "class_name": "speech",
98
+ "class_conf": 0.488,
99
+ "azi_deg": 52.99,
100
+ "ele_deg": 27.09,
101
+ "dist_m": 1.24
102
+ }
103
+ ],
104
+ [
105
+ {
106
+ "track": 0,
107
+ "activity": 0.945,
108
+ "class_idx": 16,
109
+ "class_name": "speech",
110
+ "class_conf": 0.508,
111
+ "azi_deg": 53.41,
112
+ "ele_deg": 26.99,
113
+ "dist_m": 1.242
114
+ }
115
+ ],
116
+ [
117
+ {
118
+ "track": 0,
119
+ "activity": 0.944,
120
+ "class_idx": 16,
121
+ "class_name": "speech",
122
+ "class_conf": 0.524,
123
+ "azi_deg": 53.71,
124
+ "ele_deg": 26.78,
125
+ "dist_m": 1.243
126
+ }
127
+ ]
128
+ ],
129
+ "comparison": {
130
+ "T_s": 56,
131
+ "activity_gt_frac": 0.14732142857142858,
132
+ "activity_rc_frac": 0.10267857142857142,
133
+ "activity_jaccard": 0.696969696969697,
134
+ "activity_precision_rc_vs_gt": 1.0,
135
+ "activity_recall_rc_vs_gt": 0.696969696969697,
136
+ "activity_f1_rc_vs_gt": 0.8214285714285715,
137
+ "class_match_rate": 1.0,
138
+ "doa_angular_error_deg_mean": 79.1304527762299,
139
+ "doa_angular_error_deg_median": 79.57596040203721,
140
+ "distance_mae_m": 0.015748614445328712,
141
+ "top_class_agreement": 1,
142
+ "gt_top_class": "speech",
143
+ "rc_top_class": "speech"
144
+ },
145
+ "sample": "s1_rank0_opensource_emilia_zh_rawdata_English_003050"
146
+ },
147
+ "s1_rank10_fsd50k_000123": {
148
+ "gt_top_class": "insect",
149
+ "rc_top_class": "insect",
150
+ "gt_frames_preview": [
151
+ [
152
+ {
153
+ "track": 0,
154
+ "activity": 0.919,
155
+ "class_idx": 35,
156
+ "class_name": "insect",
157
+ "class_conf": 0.876,
158
+ "azi_deg": 175.31,
159
+ "ele_deg": -4.01,
160
+ "dist_m": 3.163
161
+ }
162
+ ],
163
+ [
164
+ {
165
+ "track": 0,
166
+ "activity": 0.942,
167
+ "class_idx": 35,
168
+ "class_name": "insect",
169
+ "class_conf": 0.87,
170
+ "azi_deg": 175.91,
171
+ "ele_deg": -3.56,
172
+ "dist_m": 3.156
173
+ }
174
+ ],
175
+ [
176
+ {
177
+ "track": 0,
178
+ "activity": 0.948,
179
+ "class_idx": 35,
180
+ "class_name": "insect",
181
+ "class_conf": 0.869,
182
+ "azi_deg": 176.06,
183
+ "ele_deg": -3.29,
184
+ "dist_m": 3.154
185
+ }
186
+ ],
187
+ [
188
+ {
189
+ "track": 0,
190
+ "activity": 0.949,
191
+ "class_idx": 35,
192
+ "class_name": "insect",
193
+ "class_conf": 0.871,
194
+ "azi_deg": 176.09,
195
+ "ele_deg": -2.86,
196
+ "dist_m": 3.15
197
+ }
198
+ ],
199
+ [
200
+ {
201
+ "track": 0,
202
+ "activity": 0.95,
203
+ "class_idx": 35,
204
+ "class_name": "insect",
205
+ "class_conf": 0.87,
206
+ "azi_deg": 176.3,
207
+ "ele_deg": -2.76,
208
+ "dist_m": 3.163
209
+ }
210
+ ]
211
+ ],
212
+ "rc_frames_preview": [
213
+ [
214
+ {
215
+ "track": 0,
216
+ "activity": 0.952,
217
+ "class_idx": 35,
218
+ "class_name": "insect",
219
+ "class_conf": 0.881,
220
+ "azi_deg": 5.17,
221
+ "ele_deg": 5.71,
222
+ "dist_m": 2.554
223
+ }
224
+ ],
225
+ [
226
+ {
227
+ "track": 0,
228
+ "activity": 0.973,
229
+ "class_idx": 35,
230
+ "class_name": "insect",
231
+ "class_conf": 0.879,
232
+ "azi_deg": 5.35,
233
+ "ele_deg": 5.3,
234
+ "dist_m": 2.565
235
+ }
236
+ ],
237
+ [
238
+ {
239
+ "track": 0,
240
+ "activity": 0.979,
241
+ "class_idx": 35,
242
+ "class_name": "insect",
243
+ "class_conf": 0.878,
244
+ "azi_deg": 5.46,
245
+ "ele_deg": 5.32,
246
+ "dist_m": 2.568
247
+ }
248
+ ],
249
+ [
250
+ {
251
+ "track": 0,
252
+ "activity": 0.981,
253
+ "class_idx": 35,
254
+ "class_name": "insect",
255
+ "class_conf": 0.876,
256
+ "azi_deg": 5.86,
257
+ "ele_deg": 5.54,
258
+ "dist_m": 2.563
259
+ }
260
+ ],
261
+ [
262
+ {
263
+ "track": 0,
264
+ "activity": 0.982,
265
+ "class_idx": 35,
266
+ "class_name": "insect",
267
+ "class_conf": 0.875,
268
+ "azi_deg": 5.85,
269
+ "ele_deg": 5.47,
270
+ "dist_m": 2.56
271
+ }
272
+ ]
273
+ ],
274
+ "comparison": {
275
+ "T_s": 59,
276
+ "activity_gt_frac": 0.25,
277
+ "activity_rc_frac": 0.25,
278
+ "activity_jaccard": 1.0,
279
+ "activity_precision_rc_vs_gt": 1.0,
280
+ "activity_recall_rc_vs_gt": 1.0,
281
+ "activity_f1_rc_vs_gt": 1.0,
282
+ "class_match_rate": 1.0,
283
+ "doa_angular_error_deg_mean": 170.93909933913847,
284
+ "doa_angular_error_deg_median": 171.11864844527656,
285
+ "distance_mae_m": 0.6171925067901611,
286
+ "top_class_agreement": 1,
287
+ "gt_top_class": "insect",
288
+ "rc_top_class": "insect"
289
+ },
290
+ "sample": "s1_rank10_fsd50k_000123"
291
+ },
292
+ "s1_rank10_insectset_000529": {
293
+ "gt_top_class": "metal_clink",
294
+ "rc_top_class": "metal_clink",
295
+ "gt_frames_preview": [
296
+ [
297
+ {
298
+ "track": 0,
299
+ "activity": 0.598,
300
+ "class_idx": 35,
301
+ "class_name": "insect",
302
+ "class_conf": 0.279,
303
+ "azi_deg": -80.4,
304
+ "ele_deg": 23.47,
305
+ "dist_m": 1.418
306
+ }
307
+ ],
308
+ [
309
+ {
310
+ "track": 0,
311
+ "activity": 0.598,
312
+ "class_idx": 35,
313
+ "class_name": "insect",
314
+ "class_conf": 0.28,
315
+ "azi_deg": -80.17,
316
+ "ele_deg": 23.68,
317
+ "dist_m": 1.402
318
+ }
319
+ ],
320
+ [
321
+ {
322
+ "track": 0,
323
+ "activity": 0.6,
324
+ "class_idx": 35,
325
+ "class_name": "insect",
326
+ "class_conf": 0.28,
327
+ "azi_deg": -80.23,
328
+ "ele_deg": 23.71,
329
+ "dist_m": 1.402
330
+ }
331
+ ],
332
+ [
333
+ {
334
+ "track": 0,
335
+ "activity": 0.618,
336
+ "class_idx": 35,
337
+ "class_name": "insect",
338
+ "class_conf": 0.285,
339
+ "azi_deg": -80.66,
340
+ "ele_deg": 23.91,
341
+ "dist_m": 1.4
342
+ }
343
+ ],
344
+ [
345
+ {
346
+ "track": 0,
347
+ "activity": 0.639,
348
+ "class_idx": 12,
349
+ "class_name": "metal_clink",
350
+ "class_conf": 0.283,
351
+ "azi_deg": -81.02,
352
+ "ele_deg": 24.07,
353
+ "dist_m": 1.401
354
+ }
355
+ ]
356
+ ],
357
+ "rc_frames_preview": [
358
+ [
359
+ {
360
+ "track": 0,
361
+ "activity": 0.545,
362
+ "class_idx": 12,
363
+ "class_name": "metal_clink",
364
+ "class_conf": 0.329,
365
+ "azi_deg": 85.21,
366
+ "ele_deg": 9.44,
367
+ "dist_m": 1.368
368
+ }
369
+ ],
370
+ [
371
+ {
372
+ "track": 0,
373
+ "activity": 0.545,
374
+ "class_idx": 12,
375
+ "class_name": "metal_clink",
376
+ "class_conf": 0.33,
377
+ "azi_deg": 84.93,
378
+ "ele_deg": 9.4,
379
+ "dist_m": 1.331
380
+ }
381
+ ],
382
+ [
383
+ {
384
+ "track": 0,
385
+ "activity": 0.545,
386
+ "class_idx": 12,
387
+ "class_name": "metal_clink",
388
+ "class_conf": 0.333,
389
+ "azi_deg": 84.92,
390
+ "ele_deg": 9.39,
391
+ "dist_m": 1.332
392
+ }
393
+ ],
394
+ [
395
+ {
396
+ "track": 0,
397
+ "activity": 0.545,
398
+ "class_idx": 12,
399
+ "class_name": "metal_clink",
400
+ "class_conf": 0.367,
401
+ "azi_deg": 84.93,
402
+ "ele_deg": 9.44,
403
+ "dist_m": 1.345
404
+ }
405
+ ],
406
+ [
407
+ {
408
+ "track": 0,
409
+ "activity": 0.547,
410
+ "class_idx": 12,
411
+ "class_name": "metal_clink",
412
+ "class_conf": 0.387,
413
+ "azi_deg": 84.94,
414
+ "ele_deg": 9.45,
415
+ "dist_m": 1.354
416
+ }
417
+ ]
418
+ ],
419
+ "comparison": {
420
+ "T_s": 303,
421
+ "activity_gt_frac": 0.18976897689768976,
422
+ "activity_rc_frac": 0.04702970297029703,
423
+ "activity_jaccard": 0.24782608695652175,
424
+ "activity_precision_rc_vs_gt": 1.0,
425
+ "activity_recall_rc_vs_gt": 0.24782608695652175,
426
+ "activity_f1_rc_vs_gt": 0.3972125435540069,
427
+ "class_match_rate": 0.8947368421052632,
428
+ "doa_angular_error_deg_mean": 144.66956845489622,
429
+ "doa_angular_error_deg_median": 144.30882197266905,
430
+ "distance_mae_m": 0.045821111649274826,
431
+ "top_class_agreement": 1,
432
+ "gt_top_class": "metal_clink",
433
+ "rc_top_class": "metal_clink"
434
+ },
435
+ "sample": "s1_rank10_insectset_000529"
436
+ },
437
+ "s1_rank10_laion_630k_002480": {
438
+ "gt_top_class": "keyboard_instrument",
439
+ "rc_top_class": "keyboard_instrument",
440
+ "gt_frames_preview": [
441
+ [],
442
+ [],
443
+ [],
444
+ [],
445
+ []
446
+ ],
447
+ "rc_frames_preview": [
448
+ [],
449
+ [],
450
+ [],
451
+ [],
452
+ []
453
+ ],
454
+ "comparison": {
455
+ "T_s": 300,
456
+ "activity_gt_frac": 0.23416666666666666,
457
+ "activity_rc_frac": 0.23333333333333334,
458
+ "activity_jaccard": 0.99644128113879,
459
+ "activity_precision_rc_vs_gt": 1.0,
460
+ "activity_recall_rc_vs_gt": 0.99644128113879,
461
+ "activity_f1_rc_vs_gt": 0.9982174688057042,
462
+ "class_match_rate": 1.0,
463
+ "doa_angular_error_deg_mean": 85.83817700633047,
464
+ "doa_angular_error_deg_median": 84.23720291735404,
465
+ "distance_mae_m": 0.4552581012248993,
466
+ "top_class_agreement": 1,
467
+ "gt_top_class": "keyboard_instrument",
468
+ "rc_top_class": "keyboard_instrument"
469
+ },
470
+ "sample": "s1_rank10_laion_630k_002480"
471
+ },
472
+ "s1_rank10_temporal_order_001229": {
473
+ "gt_top_class": "human_vocalization",
474
+ "rc_top_class": "human_vocalization",
475
+ "gt_frames_preview": [
476
+ [],
477
+ [],
478
+ [],
479
+ [],
480
+ [
481
+ {
482
+ "track": 0,
483
+ "activity": 0.529,
484
+ "class_idx": 6,
485
+ "class_name": "human_vocalization",
486
+ "class_conf": 0.807,
487
+ "azi_deg": 90.28,
488
+ "ele_deg": 20.51,
489
+ "dist_m": 1.839
490
+ }
491
+ ]
492
+ ],
493
+ "rc_frames_preview": [
494
+ [
495
+ {
496
+ "track": 0,
497
+ "activity": 0.682,
498
+ "class_idx": 6,
499
+ "class_name": "human_vocalization",
500
+ "class_conf": 0.815,
501
+ "azi_deg": 92.34,
502
+ "ele_deg": -13.24,
503
+ "dist_m": 1.281
504
+ }
505
+ ],
506
+ [
507
+ {
508
+ "track": 0,
509
+ "activity": 0.816,
510
+ "class_idx": 6,
511
+ "class_name": "human_vocalization",
512
+ "class_conf": 0.805,
513
+ "azi_deg": 92.99,
514
+ "ele_deg": -13.23,
515
+ "dist_m": 1.25
516
+ }
517
+ ],
518
+ [
519
+ {
520
+ "track": 0,
521
+ "activity": 0.84,
522
+ "class_idx": 6,
523
+ "class_name": "human_vocalization",
524
+ "class_conf": 0.807,
525
+ "azi_deg": 93.55,
526
+ "ele_deg": -13.22,
527
+ "dist_m": 1.226
528
+ }
529
+ ],
530
+ [
531
+ {
532
+ "track": 0,
533
+ "activity": 0.859,
534
+ "class_idx": 6,
535
+ "class_name": "human_vocalization",
536
+ "class_conf": 0.804,
537
+ "azi_deg": 93.91,
538
+ "ele_deg": -13.13,
539
+ "dist_m": 1.213
540
+ }
541
+ ],
542
+ [
543
+ {
544
+ "track": 0,
545
+ "activity": 0.883,
546
+ "class_idx": 6,
547
+ "class_name": "human_vocalization",
548
+ "class_conf": 0.8,
549
+ "azi_deg": 94.32,
550
+ "ele_deg": -12.99,
551
+ "dist_m": 1.216
552
+ }
553
+ ]
554
+ ],
555
+ "comparison": {
556
+ "T_s": 29,
557
+ "activity_gt_frac": 0.15517241379310345,
558
+ "activity_rc_frac": 0.2413793103448276,
559
+ "activity_jaccard": 0.6428571428571429,
560
+ "activity_precision_rc_vs_gt": 0.6428571428571429,
561
+ "activity_recall_rc_vs_gt": 1.0,
562
+ "activity_f1_rc_vs_gt": 0.782608695652174,
563
+ "class_match_rate": 1.0,
564
+ "doa_angular_error_deg_mean": 33.530018422250656,
565
+ "doa_angular_error_deg_median": 33.52572140953926,
566
+ "distance_mae_m": 0.6078136563301086,
567
+ "top_class_agreement": 1,
568
+ "gt_top_class": "human_vocalization",
569
+ "rc_top_class": "human_vocalization"
570
+ },
571
+ "sample": "s1_rank10_temporal_order_001229"
572
+ },
573
+ "s1_rank14_audioset_keyword_000653": {
574
+ "gt_top_class": "human_vocalization",
575
+ "rc_top_class": "singing",
576
+ "gt_frames_preview": [
577
+ [],
578
+ [],
579
+ [],
580
+ [],
581
+ []
582
+ ],
583
+ "rc_frames_preview": [
584
+ [],
585
+ [],
586
+ [],
587
+ [],
588
+ []
589
+ ],
590
+ "comparison": {
591
+ "T_s": 103,
592
+ "activity_gt_frac": 0.2087378640776699,
593
+ "activity_rc_frac": 0.220873786407767,
594
+ "activity_jaccard": 0.945054945054945,
595
+ "activity_precision_rc_vs_gt": 0.945054945054945,
596
+ "activity_recall_rc_vs_gt": 1.0,
597
+ "activity_f1_rc_vs_gt": 0.9717514124293786,
598
+ "class_match_rate": 0.05813953488372093,
599
+ "doa_angular_error_deg_mean": 90.86238526212728,
600
+ "doa_angular_error_deg_median": 90.85882765252757,
601
+ "distance_mae_m": 0.11580874025821686,
602
+ "top_class_agreement": 0,
603
+ "gt_top_class": "human_vocalization",
604
+ "rc_top_class": "singing"
605
+ },
606
+ "sample": "s1_rank14_audioset_keyword_000653"
607
+ },
608
+ "s1_rank14_fsd50k_000024": {
609
+ "gt_top_class": "keyboard_instrument",
610
+ "rc_top_class": "keyboard_instrument",
611
+ "gt_frames_preview": [
612
+ [
613
+ {
614
+ "track": 0,
615
+ "activity": 0.741,
616
+ "class_idx": 7,
617
+ "class_name": "keyboard_instrument",
618
+ "class_conf": 0.85,
619
+ "azi_deg": 106.65,
620
+ "ele_deg": 15.69,
621
+ "dist_m": 1.944
622
+ }
623
+ ],
624
+ [
625
+ {
626
+ "track": 0,
627
+ "activity": 0.833,
628
+ "class_idx": 7,
629
+ "class_name": "keyboard_instrument",
630
+ "class_conf": 0.853,
631
+ "azi_deg": 108.18,
632
+ "ele_deg": 16.01,
633
+ "dist_m": 1.926
634
+ }
635
+ ],
636
+ [
637
+ {
638
+ "track": 0,
639
+ "activity": 0.83,
640
+ "class_idx": 7,
641
+ "class_name": "keyboard_instrument",
642
+ "class_conf": 0.849,
643
+ "azi_deg": 108.66,
644
+ "ele_deg": 15.48,
645
+ "dist_m": 1.927
646
+ }
647
+ ],
648
+ [
649
+ {
650
+ "track": 0,
651
+ "activity": 0.837,
652
+ "class_idx": 7,
653
+ "class_name": "keyboard_instrument",
654
+ "class_conf": 0.851,
655
+ "azi_deg": 108.29,
656
+ "ele_deg": 15.24,
657
+ "dist_m": 1.927
658
+ }
659
+ ],
660
+ [
661
+ {
662
+ "track": 0,
663
+ "activity": 0.844,
664
+ "class_idx": 7,
665
+ "class_name": "keyboard_instrument",
666
+ "class_conf": 0.849,
667
+ "azi_deg": 108.11,
668
+ "ele_deg": 14.97,
669
+ "dist_m": 1.931
670
+ }
671
+ ]
672
+ ],
673
+ "rc_frames_preview": [
674
+ [
675
+ {
676
+ "track": 0,
677
+ "activity": 0.865,
678
+ "class_idx": 7,
679
+ "class_name": "keyboard_instrument",
680
+ "class_conf": 0.857,
681
+ "azi_deg": 22.99,
682
+ "ele_deg": 15.29,
683
+ "dist_m": 1.731
684
+ }
685
+ ],
686
+ [
687
+ {
688
+ "track": 0,
689
+ "activity": 0.911,
690
+ "class_idx": 7,
691
+ "class_name": "keyboard_instrument",
692
+ "class_conf": 0.859,
693
+ "azi_deg": 25.39,
694
+ "ele_deg": 16.98,
695
+ "dist_m": 1.733
696
+ }
697
+ ],
698
+ [
699
+ {
700
+ "track": 0,
701
+ "activity": 0.904,
702
+ "class_idx": 7,
703
+ "class_name": "keyboard_instrument",
704
+ "class_conf": 0.857,
705
+ "azi_deg": 28.19,
706
+ "ele_deg": 16.5,
707
+ "dist_m": 1.743
708
+ }
709
+ ],
710
+ [
711
+ {
712
+ "track": 0,
713
+ "activity": 0.907,
714
+ "class_idx": 7,
715
+ "class_name": "keyboard_instrument",
716
+ "class_conf": 0.858,
717
+ "azi_deg": 29.87,
718
+ "ele_deg": 15.51,
719
+ "dist_m": 1.744
720
+ }
721
+ ],
722
+ [
723
+ {
724
+ "track": 0,
725
+ "activity": 0.909,
726
+ "class_idx": 7,
727
+ "class_name": "keyboard_instrument",
728
+ "class_conf": 0.856,
729
+ "azi_deg": 30.97,
730
+ "ele_deg": 14.81,
731
+ "dist_m": 1.752
732
+ }
733
+ ]
734
+ ],
735
+ "comparison": {
736
+ "T_s": 24,
737
+ "activity_gt_frac": 0.25,
738
+ "activity_rc_frac": 0.25,
739
+ "activity_jaccard": 1.0,
740
+ "activity_precision_rc_vs_gt": 1.0,
741
+ "activity_recall_rc_vs_gt": 1.0,
742
+ "activity_f1_rc_vs_gt": 1.0,
743
+ "class_match_rate": 1.0,
744
+ "doa_angular_error_deg_mean": 70.49814175629024,
745
+ "doa_angular_error_deg_median": 69.16372116736812,
746
+ "distance_mae_m": 0.16031360626220703,
747
+ "top_class_agreement": 1,
748
+ "gt_top_class": "keyboard_instrument",
749
+ "rc_top_class": "keyboard_instrument"
750
+ },
751
+ "sample": "s1_rank14_fsd50k_000024"
752
+ },
753
+ "s1_rank14_temporal_order_003256": {
754
+ "gt_top_class": "keyboard_instrument",
755
+ "rc_top_class": "keyboard_instrument",
756
+ "gt_frames_preview": [
757
+ [],
758
+ [],
759
+ [],
760
+ [],
761
+ []
762
+ ],
763
+ "rc_frames_preview": [
764
+ [
765
+ {
766
+ "track": 0,
767
+ "activity": 0.558,
768
+ "class_idx": 7,
769
+ "class_name": "keyboard_instrument",
770
+ "class_conf": 0.509,
771
+ "azi_deg": 60.61,
772
+ "ele_deg": -4.78,
773
+ "dist_m": 2.385
774
+ }
775
+ ],
776
+ [
777
+ {
778
+ "track": 0,
779
+ "activity": 0.552,
780
+ "class_idx": 7,
781
+ "class_name": "keyboard_instrument",
782
+ "class_conf": 0.517,
783
+ "azi_deg": 66.46,
784
+ "ele_deg": -2.69,
785
+ "dist_m": 2.404
786
+ }
787
+ ],
788
+ [
789
+ {
790
+ "track": 0,
791
+ "activity": 0.545,
792
+ "class_idx": 7,
793
+ "class_name": "keyboard_instrument",
794
+ "class_conf": 0.518,
795
+ "azi_deg": 68.43,
796
+ "ele_deg": -3.15,
797
+ "dist_m": 2.393
798
+ }
799
+ ],
800
+ [
801
+ {
802
+ "track": 0,
803
+ "activity": 0.561,
804
+ "class_idx": 7,
805
+ "class_name": "keyboard_instrument",
806
+ "class_conf": 0.524,
807
+ "azi_deg": 68.83,
808
+ "ele_deg": -3.83,
809
+ "dist_m": 2.395
810
+ }
811
+ ],
812
+ [
813
+ {
814
+ "track": 0,
815
+ "activity": 0.582,
816
+ "class_idx": 7,
817
+ "class_name": "keyboard_instrument",
818
+ "class_conf": 0.521,
819
+ "azi_deg": 68.87,
820
+ "ele_deg": -4.17,
821
+ "dist_m": 2.393
822
+ }
823
+ ]
824
+ ],
825
+ "comparison": {
826
+ "T_s": 13,
827
+ "activity_gt_frac": 0.0,
828
+ "activity_rc_frac": 0.25,
829
+ "activity_jaccard": 0.0,
830
+ "activity_precision_rc_vs_gt": 0.0,
831
+ "activity_recall_rc_vs_gt": 0.0,
832
+ "activity_f1_rc_vs_gt": 0.0,
833
+ "class_match_rate": NaN,
834
+ "doa_angular_error_deg_mean": NaN,
835
+ "doa_angular_error_deg_median": NaN,
836
+ "distance_mae_m": NaN,
837
+ "top_class_agreement": 1,
838
+ "gt_top_class": "keyboard_instrument",
839
+ "rc_top_class": "keyboard_instrument"
840
+ },
841
+ "sample": "s1_rank14_temporal_order_003256"
842
+ },
843
+ "s1_rank15_fsd50k_000489": {
844
+ "gt_top_class": "telephone_alarm",
845
+ "rc_top_class": "telephone_alarm",
846
+ "gt_frames_preview": [
847
+ [],
848
+ [
849
+ {
850
+ "track": 0,
851
+ "activity": 0.512,
852
+ "class_idx": 23,
853
+ "class_name": "telephone_alarm",
854
+ "class_conf": 0.862,
855
+ "azi_deg": 48.36,
856
+ "ele_deg": 28.03,
857
+ "dist_m": 1.952
858
+ }
859
+ ],
860
+ [
861
+ {
862
+ "track": 0,
863
+ "activity": 0.702,
864
+ "class_idx": 23,
865
+ "class_name": "telephone_alarm",
866
+ "class_conf": 0.868,
867
+ "azi_deg": 48.63,
868
+ "ele_deg": 28.02,
869
+ "dist_m": 1.937
870
+ }
871
+ ],
872
+ [
873
+ {
874
+ "track": 0,
875
+ "activity": 0.846,
876
+ "class_idx": 23,
877
+ "class_name": "telephone_alarm",
878
+ "class_conf": 0.871,
879
+ "azi_deg": 49.05,
880
+ "ele_deg": 28.02,
881
+ "dist_m": 1.925
882
+ }
883
+ ],
884
+ [
885
+ {
886
+ "track": 0,
887
+ "activity": 0.875,
888
+ "class_idx": 23,
889
+ "class_name": "telephone_alarm",
890
+ "class_conf": 0.874,
891
+ "azi_deg": 49.06,
892
+ "ele_deg": 27.97,
893
+ "dist_m": 1.918
894
+ }
895
+ ]
896
+ ],
897
+ "rc_frames_preview": [
898
+ [
899
+ {
900
+ "track": 2,
901
+ "activity": 0.677,
902
+ "class_idx": 21,
903
+ "class_name": "footsteps",
904
+ "class_conf": 0.663,
905
+ "azi_deg": 23.13,
906
+ "ele_deg": -7.29,
907
+ "dist_m": 1.936
908
+ }
909
+ ],
910
+ [
911
+ {
912
+ "track": 2,
913
+ "activity": 0.677,
914
+ "class_idx": 21,
915
+ "class_name": "footsteps",
916
+ "class_conf": 0.673,
917
+ "azi_deg": 22.95,
918
+ "ele_deg": -7.84,
919
+ "dist_m": 1.905
920
+ }
921
+ ],
922
+ [
923
+ {
924
+ "track": 2,
925
+ "activity": 0.683,
926
+ "class_idx": 21,
927
+ "class_name": "footsteps",
928
+ "class_conf": 0.674,
929
+ "azi_deg": 22.1,
930
+ "ele_deg": -9.11,
931
+ "dist_m": 1.875
932
+ }
933
+ ],
934
+ [
935
+ {
936
+ "track": 2,
937
+ "activity": 0.698,
938
+ "class_idx": 21,
939
+ "class_name": "footsteps",
940
+ "class_conf": 0.651,
941
+ "azi_deg": 21.47,
942
+ "ele_deg": -10.36,
943
+ "dist_m": 1.854
944
+ }
945
+ ],
946
+ [
947
+ {
948
+ "track": 2,
949
+ "activity": 0.679,
950
+ "class_idx": 21,
951
+ "class_name": "footsteps",
952
+ "class_conf": 0.648,
953
+ "azi_deg": 21.9,
954
+ "ele_deg": -10.96,
955
+ "dist_m": 1.854
956
+ }
957
+ ]
958
+ ],
959
+ "comparison": {
960
+ "T_s": 155,
961
+ "activity_gt_frac": 0.22258064516129034,
962
+ "activity_rc_frac": 0.4258064516129032,
963
+ "activity_jaccard": 0.4725274725274725,
964
+ "activity_precision_rc_vs_gt": 0.48863636363636365,
965
+ "activity_recall_rc_vs_gt": 0.9347826086956522,
966
+ "activity_f1_rc_vs_gt": 0.6417910447761195,
967
+ "class_match_rate": 1.0,
968
+ "doa_angular_error_deg_mean": 59.21240288527094,
969
+ "doa_angular_error_deg_median": 57.07808634075911,
970
+ "distance_mae_m": 0.47250187397003174,
971
+ "top_class_agreement": 1,
972
+ "gt_top_class": "telephone_alarm",
973
+ "rc_top_class": "telephone_alarm"
974
+ },
975
+ "sample": "s1_rank15_fsd50k_000489"
976
+ },
977
+ "s1_rank17_audioset_keyword_001725": {
978
+ "gt_top_class": "guitar",
979
+ "rc_top_class": "guitar",
980
+ "gt_frames_preview": [
981
+ [
982
+ {
983
+ "track": 0,
984
+ "activity": 0.978,
985
+ "class_idx": 2,
986
+ "class_name": "guitar",
987
+ "class_conf": 0.871,
988
+ "azi_deg": -6.63,
989
+ "ele_deg": 1.21,
990
+ "dist_m": 2.677
991
+ }
992
+ ],
993
+ [
994
+ {
995
+ "track": 0,
996
+ "activity": 0.979,
997
+ "class_idx": 2,
998
+ "class_name": "guitar",
999
+ "class_conf": 0.871,
1000
+ "azi_deg": -6.29,
1001
+ "ele_deg": 0.95,
1002
+ "dist_m": 2.66
1003
+ }
1004
+ ],
1005
+ [
1006
+ {
1007
+ "track": 0,
1008
+ "activity": 0.982,
1009
+ "class_idx": 2,
1010
+ "class_name": "guitar",
1011
+ "class_conf": 0.876,
1012
+ "azi_deg": -6.28,
1013
+ "ele_deg": 0.53,
1014
+ "dist_m": 2.64
1015
+ }
1016
+ ],
1017
+ [
1018
+ {
1019
+ "track": 0,
1020
+ "activity": 0.982,
1021
+ "class_idx": 2,
1022
+ "class_name": "guitar",
1023
+ "class_conf": 0.874,
1024
+ "azi_deg": -7.43,
1025
+ "ele_deg": 0.61,
1026
+ "dist_m": 2.621
1027
+ }
1028
+ ],
1029
+ [
1030
+ {
1031
+ "track": 0,
1032
+ "activity": 0.983,
1033
+ "class_idx": 2,
1034
+ "class_name": "guitar",
1035
+ "class_conf": 0.879,
1036
+ "azi_deg": -8.48,
1037
+ "ele_deg": 0.63,
1038
+ "dist_m": 2.606
1039
+ }
1040
+ ]
1041
+ ],
1042
+ "rc_frames_preview": [
1043
+ [
1044
+ {
1045
+ "track": 0,
1046
+ "activity": 0.974,
1047
+ "class_idx": 2,
1048
+ "class_name": "guitar",
1049
+ "class_conf": 0.852,
1050
+ "azi_deg": 27.19,
1051
+ "ele_deg": 7.38,
1052
+ "dist_m": 2.048
1053
+ }
1054
+ ],
1055
+ [
1056
+ {
1057
+ "track": 0,
1058
+ "activity": 0.974,
1059
+ "class_idx": 2,
1060
+ "class_name": "guitar",
1061
+ "class_conf": 0.856,
1062
+ "azi_deg": 27.5,
1063
+ "ele_deg": 7.3,
1064
+ "dist_m": 2.039
1065
+ }
1066
+ ],
1067
+ [
1068
+ {
1069
+ "track": 0,
1070
+ "activity": 0.975,
1071
+ "class_idx": 2,
1072
+ "class_name": "guitar",
1073
+ "class_conf": 0.867,
1074
+ "azi_deg": 27.55,
1075
+ "ele_deg": 7.15,
1076
+ "dist_m": 2.022
1077
+ }
1078
+ ],
1079
+ [
1080
+ {
1081
+ "track": 0,
1082
+ "activity": 0.975,
1083
+ "class_idx": 2,
1084
+ "class_name": "guitar",
1085
+ "class_conf": 0.871,
1086
+ "azi_deg": 27.6,
1087
+ "ele_deg": 7.34,
1088
+ "dist_m": 1.997
1089
+ }
1090
+ ],
1091
+ [
1092
+ {
1093
+ "track": 0,
1094
+ "activity": 0.975,
1095
+ "class_idx": 2,
1096
+ "class_name": "guitar",
1097
+ "class_conf": 0.879,
1098
+ "azi_deg": 27.57,
1099
+ "ele_deg": 7.43,
1100
+ "dist_m": 1.984
1101
+ }
1102
+ ]
1103
+ ],
1104
+ "comparison": {
1105
+ "T_s": 104,
1106
+ "activity_gt_frac": 0.24759615384615385,
1107
+ "activity_rc_frac": 0.25,
1108
+ "activity_jaccard": 0.9903846153846154,
1109
+ "activity_precision_rc_vs_gt": 0.9903846153846154,
1110
+ "activity_recall_rc_vs_gt": 1.0,
1111
+ "activity_f1_rc_vs_gt": 0.9951690821256038,
1112
+ "class_match_rate": 1.0,
1113
+ "doa_angular_error_deg_mean": 40.05919563604413,
1114
+ "doa_angular_error_deg_median": 40.015548939435035,
1115
+ "distance_mae_m": 0.5851235389709473,
1116
+ "top_class_agreement": 1,
1117
+ "gt_top_class": "guitar",
1118
+ "rc_top_class": "guitar"
1119
+ },
1120
+ "sample": "s1_rank17_audioset_keyword_001725"
1121
+ },
1122
+ "s1_rank18_opensource_emilia_zh_rawdata_English_000545": {
1123
+ "gt_top_class": "speech",
1124
+ "rc_top_class": "speech",
1125
+ "gt_frames_preview": [
1126
+ [
1127
+ {
1128
+ "track": 0,
1129
+ "activity": 0.616,
1130
+ "class_idx": 16,
1131
+ "class_name": "speech",
1132
+ "class_conf": 0.859,
1133
+ "azi_deg": 54.28,
1134
+ "ele_deg": 32.99,
1135
+ "dist_m": 1.793
1136
+ }
1137
+ ],
1138
+ [
1139
+ {
1140
+ "track": 0,
1141
+ "activity": 0.644,
1142
+ "class_idx": 16,
1143
+ "class_name": "speech",
1144
+ "class_conf": 0.864,
1145
+ "azi_deg": 54.3,
1146
+ "ele_deg": 32.91,
1147
+ "dist_m": 1.788
1148
+ }
1149
+ ],
1150
+ [
1151
+ {
1152
+ "track": 0,
1153
+ "activity": 0.729,
1154
+ "class_idx": 16,
1155
+ "class_name": "speech",
1156
+ "class_conf": 0.87,
1157
+ "azi_deg": 54.21,
1158
+ "ele_deg": 32.44,
1159
+ "dist_m": 1.792
1160
+ }
1161
+ ],
1162
+ [
1163
+ {
1164
+ "track": 0,
1165
+ "activity": 0.809,
1166
+ "class_idx": 16,
1167
+ "class_name": "speech",
1168
+ "class_conf": 0.876,
1169
+ "azi_deg": 54.15,
1170
+ "ele_deg": 32.04,
1171
+ "dist_m": 1.808
1172
+ }
1173
+ ],
1174
+ [
1175
+ {
1176
+ "track": 0,
1177
+ "activity": 0.824,
1178
+ "class_idx": 16,
1179
+ "class_name": "speech",
1180
+ "class_conf": 0.877,
1181
+ "azi_deg": 53.89,
1182
+ "ele_deg": 31.84,
1183
+ "dist_m": 1.814
1184
+ }
1185
+ ]
1186
+ ],
1187
+ "rc_frames_preview": [
1188
+ [],
1189
+ [],
1190
+ [],
1191
+ [],
1192
+ []
1193
+ ],
1194
+ "comparison": {
1195
+ "T_s": 147,
1196
+ "activity_gt_frac": 0.12585034013605442,
1197
+ "activity_rc_frac": 0.0,
1198
+ "activity_jaccard": 0.0,
1199
+ "activity_precision_rc_vs_gt": 0.0,
1200
+ "activity_recall_rc_vs_gt": 0.0,
1201
+ "activity_f1_rc_vs_gt": 0.0,
1202
+ "class_match_rate": NaN,
1203
+ "doa_angular_error_deg_mean": NaN,
1204
+ "doa_angular_error_deg_median": NaN,
1205
+ "distance_mae_m": NaN,
1206
+ "top_class_agreement": 1,
1207
+ "gt_top_class": "speech",
1208
+ "rc_top_class": "speech"
1209
+ },
1210
+ "sample": "s1_rank18_opensource_emilia_zh_rawdata_English_000545"
1211
+ },
1212
+ "s1_rank20_laion_630k_002181": {
1213
+ "gt_top_class": "drum",
1214
+ "rc_top_class": "paper",
1215
+ "gt_frames_preview": [
1216
+ [
1217
+ {
1218
+ "track": 0,
1219
+ "activity": 0.895,
1220
+ "class_idx": 4,
1221
+ "class_name": "drum",
1222
+ "class_conf": 0.562,
1223
+ "azi_deg": -117.81,
1224
+ "ele_deg": 27.03,
1225
+ "dist_m": 1.466
1226
+ }
1227
+ ],
1228
+ [
1229
+ {
1230
+ "track": 0,
1231
+ "activity": 0.896,
1232
+ "class_idx": 4,
1233
+ "class_name": "drum",
1234
+ "class_conf": 0.568,
1235
+ "azi_deg": -118.0,
1236
+ "ele_deg": 26.75,
1237
+ "dist_m": 1.479
1238
+ }
1239
+ ],
1240
+ [
1241
+ {
1242
+ "track": 0,
1243
+ "activity": 0.908,
1244
+ "class_idx": 4,
1245
+ "class_name": "drum",
1246
+ "class_conf": 0.578,
1247
+ "azi_deg": -118.28,
1248
+ "ele_deg": 26.29,
1249
+ "dist_m": 1.475
1250
+ }
1251
+ ],
1252
+ [
1253
+ {
1254
+ "track": 0,
1255
+ "activity": 0.922,
1256
+ "class_idx": 4,
1257
+ "class_name": "drum",
1258
+ "class_conf": 0.576,
1259
+ "azi_deg": -118.63,
1260
+ "ele_deg": 25.93,
1261
+ "dist_m": 1.462
1262
+ }
1263
+ ],
1264
+ [
1265
+ {
1266
+ "track": 0,
1267
+ "activity": 0.933,
1268
+ "class_idx": 4,
1269
+ "class_name": "drum",
1270
+ "class_conf": 0.572,
1271
+ "azi_deg": -118.84,
1272
+ "ele_deg": 25.37,
1273
+ "dist_m": 1.459
1274
+ }
1275
+ ]
1276
+ ],
1277
+ "rc_frames_preview": [
1278
+ [
1279
+ {
1280
+ "track": 1,
1281
+ "activity": 0.873,
1282
+ "class_idx": 49,
1283
+ "class_name": "paper",
1284
+ "class_conf": 0.509,
1285
+ "azi_deg": -66.87,
1286
+ "ele_deg": 18.59,
1287
+ "dist_m": 1.061
1288
+ }
1289
+ ],
1290
+ [
1291
+ {
1292
+ "track": 1,
1293
+ "activity": 0.882,
1294
+ "class_idx": 49,
1295
+ "class_name": "paper",
1296
+ "class_conf": 0.498,
1297
+ "azi_deg": -67.45,
1298
+ "ele_deg": 18.45,
1299
+ "dist_m": 1.101
1300
+ }
1301
+ ],
1302
+ [
1303
+ {
1304
+ "track": 0,
1305
+ "activity": 0.532,
1306
+ "class_idx": 49,
1307
+ "class_name": "paper",
1308
+ "class_conf": 0.214,
1309
+ "azi_deg": 73.84,
1310
+ "ele_deg": -6.8,
1311
+ "dist_m": 1.224
1312
+ },
1313
+ {
1314
+ "track": 1,
1315
+ "activity": 0.923,
1316
+ "class_idx": 49,
1317
+ "class_name": "paper",
1318
+ "class_conf": 0.497,
1319
+ "azi_deg": -68.01,
1320
+ "ele_deg": 20.55,
1321
+ "dist_m": 1.1
1322
+ }
1323
+ ],
1324
+ [
1325
+ {
1326
+ "track": 0,
1327
+ "activity": 0.562,
1328
+ "class_idx": 49,
1329
+ "class_name": "paper",
1330
+ "class_conf": 0.218,
1331
+ "azi_deg": 74.44,
1332
+ "ele_deg": -6.5,
1333
+ "dist_m": 1.213
1334
+ },
1335
+ {
1336
+ "track": 1,
1337
+ "activity": 0.941,
1338
+ "class_idx": 49,
1339
+ "class_name": "paper",
1340
+ "class_conf": 0.502,
1341
+ "azi_deg": -68.83,
1342
+ "ele_deg": 23.11,
1343
+ "dist_m": 1.08
1344
+ }
1345
+ ],
1346
+ [
1347
+ {
1348
+ "track": 0,
1349
+ "activity": 0.559,
1350
+ "class_idx": 49,
1351
+ "class_name": "paper",
1352
+ "class_conf": 0.216,
1353
+ "azi_deg": 74.76,
1354
+ "ele_deg": -6.29,
1355
+ "dist_m": 1.223
1356
+ },
1357
+ {
1358
+ "track": 1,
1359
+ "activity": 0.946,
1360
+ "class_idx": 49,
1361
+ "class_name": "paper",
1362
+ "class_conf": 0.498,
1363
+ "azi_deg": -69.96,
1364
+ "ele_deg": 24.2,
1365
+ "dist_m": 1.084
1366
+ }
1367
+ ]
1368
+ ],
1369
+ "comparison": {
1370
+ "T_s": 174,
1371
+ "activity_gt_frac": 0.27873563218390807,
1372
+ "activity_rc_frac": 0.22988505747126436,
1373
+ "activity_jaccard": 0.3258426966292135,
1374
+ "activity_precision_rc_vs_gt": 0.54375,
1375
+ "activity_recall_rc_vs_gt": 0.4484536082474227,
1376
+ "activity_f1_rc_vs_gt": 0.4915254237288136,
1377
+ "class_match_rate": 0.9195402298850575,
1378
+ "doa_angular_error_deg_mean": 85.07512953508508,
1379
+ "doa_angular_error_deg_median": 80.06309352146265,
1380
+ "distance_mae_m": 0.1350812166929245,
1381
+ "top_class_agreement": 0,
1382
+ "gt_top_class": "drum",
1383
+ "rc_top_class": "paper"
1384
+ },
1385
+ "sample": "s1_rank20_laion_630k_002181"
1386
+ },
1387
+ "s1_rank20_vocalsound_000310": {
1388
+ "gt_top_class": "laughter",
1389
+ "rc_top_class": "laughter",
1390
+ "gt_frames_preview": [
1391
+ [],
1392
+ [
1393
+ {
1394
+ "track": 0,
1395
+ "activity": 0.514,
1396
+ "class_idx": 14,
1397
+ "class_name": "laughter",
1398
+ "class_conf": 0.735,
1399
+ "azi_deg": -54.88,
1400
+ "ele_deg": 6.41,
1401
+ "dist_m": 2.726
1402
+ }
1403
+ ],
1404
+ [
1405
+ {
1406
+ "track": 0,
1407
+ "activity": 0.505,
1408
+ "class_idx": 14,
1409
+ "class_name": "laughter",
1410
+ "class_conf": 0.779,
1411
+ "azi_deg": -59.56,
1412
+ "ele_deg": 7.71,
1413
+ "dist_m": 2.714
1414
+ }
1415
+ ],
1416
+ [
1417
+ {
1418
+ "track": 0,
1419
+ "activity": 0.532,
1420
+ "class_idx": 14,
1421
+ "class_name": "laughter",
1422
+ "class_conf": 0.788,
1423
+ "azi_deg": -63.85,
1424
+ "ele_deg": 9.31,
1425
+ "dist_m": 2.698
1426
+ }
1427
+ ],
1428
+ [
1429
+ {
1430
+ "track": 0,
1431
+ "activity": 0.535,
1432
+ "class_idx": 14,
1433
+ "class_name": "laughter",
1434
+ "class_conf": 0.81,
1435
+ "azi_deg": -66.25,
1436
+ "ele_deg": 10.24,
1437
+ "dist_m": 2.685
1438
+ }
1439
+ ]
1440
+ ],
1441
+ "rc_frames_preview": [
1442
+ [
1443
+ {
1444
+ "track": 1,
1445
+ "activity": 0.741,
1446
+ "class_idx": 9,
1447
+ "class_name": "tool",
1448
+ "class_conf": 0.461,
1449
+ "azi_deg": 45.47,
1450
+ "ele_deg": 5.18,
1451
+ "dist_m": 1.193
1452
+ }
1453
+ ],
1454
+ [
1455
+ {
1456
+ "track": 0,
1457
+ "activity": 0.531,
1458
+ "class_idx": 14,
1459
+ "class_name": "laughter",
1460
+ "class_conf": 0.463,
1461
+ "azi_deg": 39.45,
1462
+ "ele_deg": -3.27,
1463
+ "dist_m": 3.978
1464
+ },
1465
+ {
1466
+ "track": 1,
1467
+ "activity": 0.501,
1468
+ "class_idx": 9,
1469
+ "class_name": "tool",
1470
+ "class_conf": 0.512,
1471
+ "azi_deg": 45.99,
1472
+ "ele_deg": 5.23,
1473
+ "dist_m": 1.195
1474
+ }
1475
+ ],
1476
+ [
1477
+ {
1478
+ "track": 0,
1479
+ "activity": 0.599,
1480
+ "class_idx": 14,
1481
+ "class_name": "laughter",
1482
+ "class_conf": 0.531,
1483
+ "azi_deg": 38.56,
1484
+ "ele_deg": -3.77,
1485
+ "dist_m": 3.918
1486
+ }
1487
+ ],
1488
+ [
1489
+ {
1490
+ "track": 0,
1491
+ "activity": 0.635,
1492
+ "class_idx": 14,
1493
+ "class_name": "laughter",
1494
+ "class_conf": 0.534,
1495
+ "azi_deg": 36.6,
1496
+ "ele_deg": -4.42,
1497
+ "dist_m": 3.932
1498
+ }
1499
+ ],
1500
+ [
1501
+ {
1502
+ "track": 0,
1503
+ "activity": 0.658,
1504
+ "class_idx": 14,
1505
+ "class_name": "laughter",
1506
+ "class_conf": 0.546,
1507
+ "azi_deg": 34.18,
1508
+ "ele_deg": -5.08,
1509
+ "dist_m": 3.895
1510
+ }
1511
+ ]
1512
+ ],
1513
+ "comparison": {
1514
+ "T_s": 53,
1515
+ "activity_gt_frac": 0.37735849056603776,
1516
+ "activity_rc_frac": 0.3915094339622642,
1517
+ "activity_jaccard": 0.25384615384615383,
1518
+ "activity_precision_rc_vs_gt": 0.39759036144578314,
1519
+ "activity_recall_rc_vs_gt": 0.4125,
1520
+ "activity_f1_rc_vs_gt": 0.40490797546012275,
1521
+ "class_match_rate": 1.0,
1522
+ "doa_angular_error_deg_mean": 144.76748800532044,
1523
+ "doa_angular_error_deg_median": 155.86862105053123,
1524
+ "distance_mae_m": 0.43952104449272156,
1525
+ "top_class_agreement": 1,
1526
+ "gt_top_class": "laughter",
1527
+ "rc_top_class": "laughter"
1528
+ },
1529
+ "sample": "s1_rank20_vocalsound_000310"
1530
+ },
1531
+ "s1_rank21_insectset_000328": {
1532
+ "gt_top_class": "water",
1533
+ "rc_top_class": "water",
1534
+ "gt_frames_preview": [
1535
+ [
1536
+ {
1537
+ "track": 0,
1538
+ "activity": 0.578,
1539
+ "class_idx": 5,
1540
+ "class_name": "water",
1541
+ "class_conf": 0.565,
1542
+ "azi_deg": 143.21,
1543
+ "ele_deg": 5.5,
1544
+ "dist_m": 2.413
1545
+ }
1546
+ ],
1547
+ [
1548
+ {
1549
+ "track": 0,
1550
+ "activity": 0.578,
1551
+ "class_idx": 5,
1552
+ "class_name": "water",
1553
+ "class_conf": 0.542,
1554
+ "azi_deg": 142.53,
1555
+ "ele_deg": 5.38,
1556
+ "dist_m": 2.417
1557
+ }
1558
+ ],
1559
+ [
1560
+ {
1561
+ "track": 0,
1562
+ "activity": 0.581,
1563
+ "class_idx": 5,
1564
+ "class_name": "water",
1565
+ "class_conf": 0.542,
1566
+ "azi_deg": 142.57,
1567
+ "ele_deg": 5.38,
1568
+ "dist_m": 2.416
1569
+ }
1570
+ ],
1571
+ [
1572
+ {
1573
+ "track": 0,
1574
+ "activity": 0.604,
1575
+ "class_idx": 5,
1576
+ "class_name": "water",
1577
+ "class_conf": 0.534,
1578
+ "azi_deg": 143.0,
1579
+ "ele_deg": 5.6,
1580
+ "dist_m": 2.409
1581
+ }
1582
+ ],
1583
+ [
1584
+ {
1585
+ "track": 0,
1586
+ "activity": 0.629,
1587
+ "class_idx": 5,
1588
+ "class_name": "water",
1589
+ "class_conf": 0.533,
1590
+ "azi_deg": 143.29,
1591
+ "ele_deg": 5.62,
1592
+ "dist_m": 2.402
1593
+ }
1594
+ ]
1595
+ ],
1596
+ "rc_frames_preview": [
1597
+ [],
1598
+ [],
1599
+ [],
1600
+ [],
1601
+ []
1602
+ ],
1603
+ "comparison": {
1604
+ "T_s": 303,
1605
+ "activity_gt_frac": 0.2433993399339934,
1606
+ "activity_rc_frac": 0.0,
1607
+ "activity_jaccard": 0.0,
1608
+ "activity_precision_rc_vs_gt": 0.0,
1609
+ "activity_recall_rc_vs_gt": 0.0,
1610
+ "activity_f1_rc_vs_gt": 0.0,
1611
+ "class_match_rate": NaN,
1612
+ "doa_angular_error_deg_mean": NaN,
1613
+ "doa_angular_error_deg_median": NaN,
1614
+ "distance_mae_m": NaN,
1615
+ "top_class_agreement": 1,
1616
+ "gt_top_class": "water",
1617
+ "rc_top_class": "water"
1618
+ },
1619
+ "sample": "s1_rank21_insectset_000328"
1620
+ },
1621
+ "s1_rank21_opensource_emilia_zh_rawdata_English_002504": {
1622
+ "gt_top_class": "speech",
1623
+ "rc_top_class": "speech",
1624
+ "gt_frames_preview": [
1625
+ [],
1626
+ [],
1627
+ [],
1628
+ [
1629
+ {
1630
+ "track": 0,
1631
+ "activity": 0.754,
1632
+ "class_idx": 16,
1633
+ "class_name": "speech",
1634
+ "class_conf": 0.898,
1635
+ "azi_deg": -177.3,
1636
+ "ele_deg": 2.6,
1637
+ "dist_m": 3.281
1638
+ }
1639
+ ],
1640
+ [
1641
+ {
1642
+ "track": 0,
1643
+ "activity": 0.846,
1644
+ "class_idx": 16,
1645
+ "class_name": "speech",
1646
+ "class_conf": 0.899,
1647
+ "azi_deg": -176.82,
1648
+ "ele_deg": 2.84,
1649
+ "dist_m": 3.314
1650
+ }
1651
+ ]
1652
+ ],
1653
+ "rc_frames_preview": [
1654
+ [],
1655
+ [],
1656
+ [
1657
+ {
1658
+ "track": 0,
1659
+ "activity": 0.592,
1660
+ "class_idx": 16,
1661
+ "class_name": "speech",
1662
+ "class_conf": 0.901,
1663
+ "azi_deg": -48.08,
1664
+ "ele_deg": 27.04,
1665
+ "dist_m": 3.432
1666
+ }
1667
+ ],
1668
+ [
1669
+ {
1670
+ "track": 0,
1671
+ "activity": 0.787,
1672
+ "class_idx": 16,
1673
+ "class_name": "speech",
1674
+ "class_conf": 0.902,
1675
+ "azi_deg": -45.94,
1676
+ "ele_deg": 24.45,
1677
+ "dist_m": 3.466
1678
+ }
1679
+ ],
1680
+ [
1681
+ {
1682
+ "track": 0,
1683
+ "activity": 0.85,
1684
+ "class_idx": 16,
1685
+ "class_name": "speech",
1686
+ "class_conf": 0.903,
1687
+ "azi_deg": -45.85,
1688
+ "ele_deg": 24.34,
1689
+ "dist_m": 3.462
1690
+ }
1691
+ ]
1692
+ ],
1693
+ "comparison": {
1694
+ "T_s": 166,
1695
+ "activity_gt_frac": 0.23795180722891565,
1696
+ "activity_rc_frac": 0.24096385542168675,
1697
+ "activity_jaccard": 0.9875,
1698
+ "activity_precision_rc_vs_gt": 0.9875,
1699
+ "activity_recall_rc_vs_gt": 1.0,
1700
+ "activity_f1_rc_vs_gt": 0.9937106918238994,
1701
+ "class_match_rate": 1.0,
1702
+ "doa_angular_error_deg_mean": 118.74276500598859,
1703
+ "doa_angular_error_deg_median": 118.9755259772574,
1704
+ "distance_mae_m": 0.10789383202791214,
1705
+ "top_class_agreement": 1,
1706
+ "gt_top_class": "speech",
1707
+ "rc_top_class": "speech"
1708
+ },
1709
+ "sample": "s1_rank21_opensource_emilia_zh_rawdata_English_002504"
1710
+ },
1711
+ "s1_rank22_audiosl_000699": {
1712
+ "gt_top_class": "water",
1713
+ "rc_top_class": "water",
1714
+ "gt_frames_preview": [
1715
+ [
1716
+ {
1717
+ "track": 0,
1718
+ "activity": 0.552,
1719
+ "class_idx": 5,
1720
+ "class_name": "water",
1721
+ "class_conf": 0.635,
1722
+ "azi_deg": -136.67,
1723
+ "ele_deg": 6.51,
1724
+ "dist_m": 1.162
1725
+ }
1726
+ ],
1727
+ [
1728
+ {
1729
+ "track": 0,
1730
+ "activity": 0.667,
1731
+ "class_idx": 5,
1732
+ "class_name": "water",
1733
+ "class_conf": 0.481,
1734
+ "azi_deg": -137.17,
1735
+ "ele_deg": 6.84,
1736
+ "dist_m": 1.152
1737
+ }
1738
+ ],
1739
+ [
1740
+ {
1741
+ "track": 0,
1742
+ "activity": 0.676,
1743
+ "class_idx": 5,
1744
+ "class_name": "water",
1745
+ "class_conf": 0.503,
1746
+ "azi_deg": -137.0,
1747
+ "ele_deg": 7.0,
1748
+ "dist_m": 1.152
1749
+ }
1750
+ ],
1751
+ [
1752
+ {
1753
+ "track": 0,
1754
+ "activity": 0.686,
1755
+ "class_idx": 5,
1756
+ "class_name": "water",
1757
+ "class_conf": 0.485,
1758
+ "azi_deg": -136.81,
1759
+ "ele_deg": 7.29,
1760
+ "dist_m": 1.165
1761
+ }
1762
+ ],
1763
+ [
1764
+ {
1765
+ "track": 0,
1766
+ "activity": 0.701,
1767
+ "class_idx": 5,
1768
+ "class_name": "water",
1769
+ "class_conf": 0.477,
1770
+ "azi_deg": -136.63,
1771
+ "ele_deg": 7.32,
1772
+ "dist_m": 1.168
1773
+ }
1774
+ ]
1775
+ ],
1776
+ "rc_frames_preview": [
1777
+ [
1778
+ {
1779
+ "track": 0,
1780
+ "activity": 0.663,
1781
+ "class_idx": 5,
1782
+ "class_name": "water",
1783
+ "class_conf": 0.724,
1784
+ "azi_deg": 162.33,
1785
+ "ele_deg": -1.57,
1786
+ "dist_m": 1.961
1787
+ }
1788
+ ],
1789
+ [
1790
+ {
1791
+ "track": 0,
1792
+ "activity": 0.645,
1793
+ "class_idx": 5,
1794
+ "class_name": "water",
1795
+ "class_conf": 0.695,
1796
+ "azi_deg": 159.21,
1797
+ "ele_deg": 0.18,
1798
+ "dist_m": 1.927
1799
+ }
1800
+ ],
1801
+ [
1802
+ {
1803
+ "track": 0,
1804
+ "activity": 0.647,
1805
+ "class_idx": 5,
1806
+ "class_name": "water",
1807
+ "class_conf": 0.689,
1808
+ "azi_deg": 156.9,
1809
+ "ele_deg": -0.06,
1810
+ "dist_m": 1.934
1811
+ }
1812
+ ],
1813
+ [
1814
+ {
1815
+ "track": 0,
1816
+ "activity": 0.657,
1817
+ "class_idx": 5,
1818
+ "class_name": "water",
1819
+ "class_conf": 0.674,
1820
+ "azi_deg": 153.65,
1821
+ "ele_deg": 1.39,
1822
+ "dist_m": 1.969
1823
+ }
1824
+ ],
1825
+ [
1826
+ {
1827
+ "track": 0,
1828
+ "activity": 0.677,
1829
+ "class_idx": 5,
1830
+ "class_name": "water",
1831
+ "class_conf": 0.662,
1832
+ "azi_deg": 152.03,
1833
+ "ele_deg": 0.78,
1834
+ "dist_m": 1.963
1835
+ }
1836
+ ]
1837
+ ],
1838
+ "comparison": {
1839
+ "T_s": 18,
1840
+ "activity_gt_frac": 0.25,
1841
+ "activity_rc_frac": 0.25,
1842
+ "activity_jaccard": 1.0,
1843
+ "activity_precision_rc_vs_gt": 1.0,
1844
+ "activity_recall_rc_vs_gt": 1.0,
1845
+ "activity_f1_rc_vs_gt": 1.0,
1846
+ "class_match_rate": 1.0,
1847
+ "doa_angular_error_deg_mean": 76.84438644086536,
1848
+ "doa_angular_error_deg_median": 79.00909168012444,
1849
+ "distance_mae_m": 0.7773998379707336,
1850
+ "top_class_agreement": 1,
1851
+ "gt_top_class": "water",
1852
+ "rc_top_class": "water"
1853
+ },
1854
+ "sample": "s1_rank22_audiosl_000699"
1855
+ },
1856
+ "s1_rank23_vocalsound_000569": {
1857
+ "gt_top_class": "breathing",
1858
+ "rc_top_class": "breathing",
1859
+ "gt_frames_preview": [
1860
+ [],
1861
+ [],
1862
+ [],
1863
+ [],
1864
+ []
1865
+ ],
1866
+ "rc_frames_preview": [
1867
+ [],
1868
+ [],
1869
+ [],
1870
+ [],
1871
+ []
1872
+ ],
1873
+ "comparison": {
1874
+ "T_s": 43,
1875
+ "activity_gt_frac": 0.18023255813953487,
1876
+ "activity_rc_frac": 0.16279069767441862,
1877
+ "activity_jaccard": 0.9032258064516129,
1878
+ "activity_precision_rc_vs_gt": 1.0,
1879
+ "activity_recall_rc_vs_gt": 0.9032258064516129,
1880
+ "activity_f1_rc_vs_gt": 0.9491525423728813,
1881
+ "class_match_rate": 1.0,
1882
+ "doa_angular_error_deg_mean": 56.3986508535398,
1883
+ "doa_angular_error_deg_median": 56.36460045021309,
1884
+ "distance_mae_m": 0.7125850319862366,
1885
+ "top_class_agreement": 1,
1886
+ "gt_top_class": "breathing",
1887
+ "rc_top_class": "breathing"
1888
+ },
1889
+ "sample": "s1_rank23_vocalsound_000569"
1890
+ },
1891
+ "s1_rank4_opensource_emilia_zh_rawdata_Chinese_003160": {
1892
+ "gt_top_class": "speech",
1893
+ "rc_top_class": "speech",
1894
+ "gt_frames_preview": [
1895
+ [],
1896
+ [],
1897
+ [],
1898
+ [],
1899
+ []
1900
+ ],
1901
+ "rc_frames_preview": [
1902
+ [],
1903
+ [],
1904
+ [],
1905
+ [],
1906
+ []
1907
+ ],
1908
+ "comparison": {
1909
+ "T_s": 134,
1910
+ "activity_gt_frac": 0.055970149253731345,
1911
+ "activity_rc_frac": 0.16791044776119404,
1912
+ "activity_jaccard": 0.3333333333333333,
1913
+ "activity_precision_rc_vs_gt": 0.3333333333333333,
1914
+ "activity_recall_rc_vs_gt": 1.0,
1915
+ "activity_f1_rc_vs_gt": 0.5,
1916
+ "class_match_rate": 1.0,
1917
+ "doa_angular_error_deg_mean": 8.119132344958041,
1918
+ "doa_angular_error_deg_median": 8.141335319703707,
1919
+ "distance_mae_m": 0.119281105697155,
1920
+ "top_class_agreement": 1,
1921
+ "gt_top_class": "speech",
1922
+ "rc_top_class": "speech"
1923
+ },
1924
+ "sample": "s1_rank4_opensource_emilia_zh_rawdata_Chinese_003160"
1925
+ },
1926
+ "s1_rank5_opensource_emilia_zh_rawdata_English_001914": {
1927
+ "gt_top_class": "speech",
1928
+ "rc_top_class": "speech",
1929
+ "gt_frames_preview": [
1930
+ [
1931
+ {
1932
+ "track": 0,
1933
+ "activity": 0.556,
1934
+ "class_idx": 16,
1935
+ "class_name": "speech",
1936
+ "class_conf": 0.867,
1937
+ "azi_deg": 14.18,
1938
+ "ele_deg": 9.65,
1939
+ "dist_m": 1.336
1940
+ }
1941
+ ],
1942
+ [
1943
+ {
1944
+ "track": 0,
1945
+ "activity": 0.748,
1946
+ "class_idx": 16,
1947
+ "class_name": "speech",
1948
+ "class_conf": 0.879,
1949
+ "azi_deg": 14.36,
1950
+ "ele_deg": 9.37,
1951
+ "dist_m": 1.305
1952
+ }
1953
+ ],
1954
+ [
1955
+ {
1956
+ "track": 0,
1957
+ "activity": 0.792,
1958
+ "class_idx": 16,
1959
+ "class_name": "speech",
1960
+ "class_conf": 0.881,
1961
+ "azi_deg": 15.14,
1962
+ "ele_deg": 9.25,
1963
+ "dist_m": 1.301
1964
+ }
1965
+ ],
1966
+ [
1967
+ {
1968
+ "track": 0,
1969
+ "activity": 0.816,
1970
+ "class_idx": 16,
1971
+ "class_name": "speech",
1972
+ "class_conf": 0.882,
1973
+ "azi_deg": 15.62,
1974
+ "ele_deg": 9.07,
1975
+ "dist_m": 1.298
1976
+ }
1977
+ ],
1978
+ [
1979
+ {
1980
+ "track": 0,
1981
+ "activity": 0.831,
1982
+ "class_idx": 16,
1983
+ "class_name": "speech",
1984
+ "class_conf": 0.883,
1985
+ "azi_deg": 15.93,
1986
+ "ele_deg": 8.96,
1987
+ "dist_m": 1.303
1988
+ }
1989
+ ]
1990
+ ],
1991
+ "rc_frames_preview": [
1992
+ [],
1993
+ [
1994
+ {
1995
+ "track": 0,
1996
+ "activity": 0.628,
1997
+ "class_idx": 16,
1998
+ "class_name": "speech",
1999
+ "class_conf": 0.839,
2000
+ "azi_deg": -1.0,
2001
+ "ele_deg": 5.48,
2002
+ "dist_m": 1.254
2003
+ }
2004
+ ],
2005
+ [
2006
+ {
2007
+ "track": 0,
2008
+ "activity": 0.667,
2009
+ "class_idx": 16,
2010
+ "class_name": "speech",
2011
+ "class_conf": 0.845,
2012
+ "azi_deg": -0.62,
2013
+ "ele_deg": 5.8,
2014
+ "dist_m": 1.256
2015
+ }
2016
+ ],
2017
+ [
2018
+ {
2019
+ "track": 0,
2020
+ "activity": 0.689,
2021
+ "class_idx": 16,
2022
+ "class_name": "speech",
2023
+ "class_conf": 0.85,
2024
+ "azi_deg": -0.22,
2025
+ "ele_deg": 5.91,
2026
+ "dist_m": 1.257
2027
+ }
2028
+ ],
2029
+ [
2030
+ {
2031
+ "track": 0,
2032
+ "activity": 0.711,
2033
+ "class_idx": 16,
2034
+ "class_name": "speech",
2035
+ "class_conf": 0.855,
2036
+ "azi_deg": 0.06,
2037
+ "ele_deg": 6.02,
2038
+ "dist_m": 1.263
2039
+ }
2040
+ ]
2041
+ ],
2042
+ "comparison": {
2043
+ "T_s": 46,
2044
+ "activity_gt_frac": 0.25,
2045
+ "activity_rc_frac": 0.17391304347826086,
2046
+ "activity_jaccard": 0.6956521739130435,
2047
+ "activity_precision_rc_vs_gt": 1.0,
2048
+ "activity_recall_rc_vs_gt": 0.6956521739130435,
2049
+ "activity_f1_rc_vs_gt": 0.8205128205128205,
2050
+ "class_match_rate": 1.0,
2051
+ "doa_angular_error_deg_mean": 15.894430784590146,
2052
+ "doa_angular_error_deg_median": 15.89357490949132,
2053
+ "distance_mae_m": 0.045476723462343216,
2054
+ "top_class_agreement": 1,
2055
+ "gt_top_class": "speech",
2056
+ "rc_top_class": "speech"
2057
+ },
2058
+ "sample": "s1_rank5_opensource_emilia_zh_rawdata_English_001914"
2059
+ },
2060
+ "s1_rank9_opensource_emilia_zh_rawdata_English_001143": {
2061
+ "gt_top_class": "speech",
2062
+ "rc_top_class": "speech",
2063
+ "gt_frames_preview": [
2064
+ [],
2065
+ [],
2066
+ [
2067
+ {
2068
+ "track": 0,
2069
+ "activity": 0.638,
2070
+ "class_idx": 16,
2071
+ "class_name": "speech",
2072
+ "class_conf": 0.889,
2073
+ "azi_deg": 28.91,
2074
+ "ele_deg": -18.71,
2075
+ "dist_m": 1.646
2076
+ }
2077
+ ],
2078
+ [
2079
+ {
2080
+ "track": 0,
2081
+ "activity": 0.754,
2082
+ "class_idx": 16,
2083
+ "class_name": "speech",
2084
+ "class_conf": 0.894,
2085
+ "azi_deg": 29.12,
2086
+ "ele_deg": -18.26,
2087
+ "dist_m": 1.635
2088
+ }
2089
+ ],
2090
+ [
2091
+ {
2092
+ "track": 0,
2093
+ "activity": 0.8,
2094
+ "class_idx": 16,
2095
+ "class_name": "speech",
2096
+ "class_conf": 0.895,
2097
+ "azi_deg": 29.48,
2098
+ "ele_deg": -18.06,
2099
+ "dist_m": 1.616
2100
+ }
2101
+ ]
2102
+ ],
2103
+ "rc_frames_preview": [
2104
+ [],
2105
+ [],
2106
+ [
2107
+ {
2108
+ "track": 0,
2109
+ "activity": 0.501,
2110
+ "class_idx": 16,
2111
+ "class_name": "speech",
2112
+ "class_conf": 0.832,
2113
+ "azi_deg": 11.65,
2114
+ "ele_deg": 2.68,
2115
+ "dist_m": 2.734
2116
+ }
2117
+ ],
2118
+ [
2119
+ {
2120
+ "track": 0,
2121
+ "activity": 0.549,
2122
+ "class_idx": 16,
2123
+ "class_name": "speech",
2124
+ "class_conf": 0.839,
2125
+ "azi_deg": 12.3,
2126
+ "ele_deg": 2.21,
2127
+ "dist_m": 2.7
2128
+ }
2129
+ ],
2130
+ [
2131
+ {
2132
+ "track": 0,
2133
+ "activity": 0.546,
2134
+ "class_idx": 16,
2135
+ "class_name": "speech",
2136
+ "class_conf": 0.842,
2137
+ "azi_deg": 12.53,
2138
+ "ele_deg": 2.17,
2139
+ "dist_m": 2.667
2140
+ }
2141
+ ]
2142
+ ],
2143
+ "comparison": {
2144
+ "T_s": 121,
2145
+ "activity_gt_frac": 0.2190082644628099,
2146
+ "activity_rc_frac": 0.010330578512396695,
2147
+ "activity_jaccard": 0.04716981132075472,
2148
+ "activity_precision_rc_vs_gt": 1.0,
2149
+ "activity_recall_rc_vs_gt": 0.04716981132075472,
2150
+ "activity_f1_rc_vs_gt": 0.09009009009009009,
2151
+ "class_match_rate": 1.0,
2152
+ "doa_angular_error_deg_mean": 26.39029939407406,
2153
+ "doa_angular_error_deg_median": 26.224675606221066,
2154
+ "distance_mae_m": 1.0609242916107178,
2155
+ "top_class_agreement": 1,
2156
+ "gt_top_class": "speech",
2157
+ "rc_top_class": "speech"
2158
+ },
2159
+ "sample": "s1_rank9_opensource_emilia_zh_rawdata_English_001143"
2160
+ }
2161
+ }
eval_voxaudio_ood_results/flow2gan/summary.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "mean_activity_jaccard": 0.5769315608191647,
3
+ "n_valid_activity_jaccard": 20,
4
+ "mean_activity_precision_rc_vs_gt": 0.7164553380856092,
5
+ "n_valid_activity_precision_rc_vs_gt": 20,
6
+ "mean_activity_recall_rc_vs_gt": 0.6691510536846746,
7
+ "n_valid_activity_recall_rc_vs_gt": 20,
8
+ "mean_activity_f1_rc_vs_gt": 0.6429039181380094,
9
+ "n_valid_activity_f1_rc_vs_gt": 20,
10
+ "mean_class_match_rate": 0.9336715651102376,
11
+ "n_valid_class_match_rate": 17,
12
+ "mean_doa_angular_error_deg_mean": 76.88068964135293,
13
+ "n_valid_doa_angular_error_deg_mean": 17,
14
+ "mean_doa_angular_error_deg_median": 77.08370928011593,
15
+ "n_valid_doa_angular_error_deg_median": 17,
16
+ "mean_distance_mae_m": 0.38080851963775997,
17
+ "n_valid_distance_mae_m": 17,
18
+ "mean_top_class_agreement": 0.9,
19
+ "n_valid_top_class_agreement": 20,
20
+ "mean_activity_gt_frac": 0.20619253654594938,
21
+ "n_valid_activity_gt_frac": 20,
22
+ "mean_activity_rc_frac": 0.19492021351895925,
23
+ "n_valid_activity_rc_frac": 20,
24
+ "n_samples": 20
25
+ }
eval_voxaudio_ood_results/stable_audio_vae/per_sample.json ADDED
@@ -0,0 +1,2342 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "s1_rank0_opensource_emilia_zh_rawdata_English_003050": {
3
+ "gt_top_class": "speech",
4
+ "rc_top_class": "speech",
5
+ "gt_frames_preview": [
6
+ [
7
+ {
8
+ "track": 0,
9
+ "activity": 0.573,
10
+ "class_idx": 16,
11
+ "class_name": "speech",
12
+ "class_conf": 0.45,
13
+ "azi_deg": -42.47,
14
+ "ele_deg": 43.03,
15
+ "dist_m": 1.291
16
+ }
17
+ ],
18
+ [
19
+ {
20
+ "track": 0,
21
+ "activity": 0.779,
22
+ "class_idx": 16,
23
+ "class_name": "speech",
24
+ "class_conf": 0.503,
25
+ "azi_deg": -42.38,
26
+ "ele_deg": 42.74,
27
+ "dist_m": 1.258
28
+ }
29
+ ],
30
+ [
31
+ {
32
+ "track": 0,
33
+ "activity": 0.784,
34
+ "class_idx": 16,
35
+ "class_name": "speech",
36
+ "class_conf": 0.523,
37
+ "azi_deg": -42.73,
38
+ "ele_deg": 42.45,
39
+ "dist_m": 1.252
40
+ }
41
+ ],
42
+ [
43
+ {
44
+ "track": 0,
45
+ "activity": 0.771,
46
+ "class_idx": 16,
47
+ "class_name": "speech",
48
+ "class_conf": 0.549,
49
+ "azi_deg": -43.1,
50
+ "ele_deg": 42.33,
51
+ "dist_m": 1.256
52
+ }
53
+ ],
54
+ [
55
+ {
56
+ "track": 0,
57
+ "activity": 0.756,
58
+ "class_idx": 16,
59
+ "class_name": "speech",
60
+ "class_conf": 0.568,
61
+ "azi_deg": -43.15,
62
+ "ele_deg": 42.18,
63
+ "dist_m": 1.258
64
+ }
65
+ ]
66
+ ],
67
+ "rc_frames_preview": [
68
+ [],
69
+ [
70
+ {
71
+ "track": 0,
72
+ "activity": 0.76,
73
+ "class_idx": 16,
74
+ "class_name": "speech",
75
+ "class_conf": 0.638,
76
+ "azi_deg": -69.75,
77
+ "ele_deg": 44.4,
78
+ "dist_m": 1.274
79
+ }
80
+ ],
81
+ [
82
+ {
83
+ "track": 0,
84
+ "activity": 0.815,
85
+ "class_idx": 16,
86
+ "class_name": "speech",
87
+ "class_conf": 0.663,
88
+ "azi_deg": -68.66,
89
+ "ele_deg": 45.92,
90
+ "dist_m": 1.263
91
+ }
92
+ ],
93
+ [
94
+ {
95
+ "track": 0,
96
+ "activity": 0.82,
97
+ "class_idx": 16,
98
+ "class_name": "speech",
99
+ "class_conf": 0.689,
100
+ "azi_deg": -67.29,
101
+ "ele_deg": 48.27,
102
+ "dist_m": 1.278
103
+ }
104
+ ],
105
+ [
106
+ {
107
+ "track": 0,
108
+ "activity": 0.823,
109
+ "class_idx": 16,
110
+ "class_name": "speech",
111
+ "class_conf": 0.709,
112
+ "azi_deg": -65.96,
113
+ "ele_deg": 50.76,
114
+ "dist_m": 1.283
115
+ }
116
+ ]
117
+ ],
118
+ "comparison": {
119
+ "T_s": 56,
120
+ "activity_gt_frac": 0.14732142857142858,
121
+ "activity_rc_frac": 0.33482142857142855,
122
+ "activity_jaccard": 0.42105263157894735,
123
+ "activity_precision_rc_vs_gt": 0.4266666666666667,
124
+ "activity_recall_rc_vs_gt": 0.9696969696969697,
125
+ "activity_f1_rc_vs_gt": 0.5925925925925927,
126
+ "class_match_rate": 1.0,
127
+ "doa_angular_error_deg_mean": 25.783820754450428,
128
+ "doa_angular_error_deg_median": 27.798307736241252,
129
+ "distance_mae_m": 0.04339398071169853,
130
+ "top_class_agreement": 1,
131
+ "gt_top_class": "speech",
132
+ "rc_top_class": "speech"
133
+ },
134
+ "sample": "s1_rank0_opensource_emilia_zh_rawdata_English_003050"
135
+ },
136
+ "s1_rank10_fsd50k_000123": {
137
+ "gt_top_class": "insect",
138
+ "rc_top_class": "insect",
139
+ "gt_frames_preview": [
140
+ [
141
+ {
142
+ "track": 0,
143
+ "activity": 0.919,
144
+ "class_idx": 35,
145
+ "class_name": "insect",
146
+ "class_conf": 0.876,
147
+ "azi_deg": 175.31,
148
+ "ele_deg": -4.01,
149
+ "dist_m": 3.163
150
+ }
151
+ ],
152
+ [
153
+ {
154
+ "track": 0,
155
+ "activity": 0.942,
156
+ "class_idx": 35,
157
+ "class_name": "insect",
158
+ "class_conf": 0.87,
159
+ "azi_deg": 175.91,
160
+ "ele_deg": -3.56,
161
+ "dist_m": 3.156
162
+ }
163
+ ],
164
+ [
165
+ {
166
+ "track": 0,
167
+ "activity": 0.948,
168
+ "class_idx": 35,
169
+ "class_name": "insect",
170
+ "class_conf": 0.869,
171
+ "azi_deg": 176.06,
172
+ "ele_deg": -3.29,
173
+ "dist_m": 3.154
174
+ }
175
+ ],
176
+ [
177
+ {
178
+ "track": 0,
179
+ "activity": 0.949,
180
+ "class_idx": 35,
181
+ "class_name": "insect",
182
+ "class_conf": 0.871,
183
+ "azi_deg": 176.09,
184
+ "ele_deg": -2.86,
185
+ "dist_m": 3.15
186
+ }
187
+ ],
188
+ [
189
+ {
190
+ "track": 0,
191
+ "activity": 0.95,
192
+ "class_idx": 35,
193
+ "class_name": "insect",
194
+ "class_conf": 0.87,
195
+ "azi_deg": 176.3,
196
+ "ele_deg": -2.76,
197
+ "dist_m": 3.163
198
+ }
199
+ ]
200
+ ],
201
+ "rc_frames_preview": [
202
+ [
203
+ {
204
+ "track": 0,
205
+ "activity": 0.855,
206
+ "class_idx": 35,
207
+ "class_name": "insect",
208
+ "class_conf": 0.873,
209
+ "azi_deg": 27.42,
210
+ "ele_deg": 46.91,
211
+ "dist_m": 2.989
212
+ }
213
+ ],
214
+ [
215
+ {
216
+ "track": 0,
217
+ "activity": 0.883,
218
+ "class_idx": 35,
219
+ "class_name": "insect",
220
+ "class_conf": 0.869,
221
+ "azi_deg": 26.97,
222
+ "ele_deg": 57.93,
223
+ "dist_m": 3.032
224
+ }
225
+ ],
226
+ [
227
+ {
228
+ "track": 0,
229
+ "activity": 0.892,
230
+ "class_idx": 35,
231
+ "class_name": "insect",
232
+ "class_conf": 0.87,
233
+ "azi_deg": 21.78,
234
+ "ele_deg": 57.42,
235
+ "dist_m": 3.06
236
+ }
237
+ ],
238
+ [
239
+ {
240
+ "track": 0,
241
+ "activity": 0.895,
242
+ "class_idx": 35,
243
+ "class_name": "insect",
244
+ "class_conf": 0.871,
245
+ "azi_deg": 19.13,
246
+ "ele_deg": 55.22,
247
+ "dist_m": 3.071
248
+ }
249
+ ],
250
+ [
251
+ {
252
+ "track": 0,
253
+ "activity": 0.897,
254
+ "class_idx": 35,
255
+ "class_name": "insect",
256
+ "class_conf": 0.871,
257
+ "azi_deg": 14.64,
258
+ "ele_deg": 54.94,
259
+ "dist_m": 3.083
260
+ }
261
+ ]
262
+ ],
263
+ "comparison": {
264
+ "T_s": 59,
265
+ "activity_gt_frac": 0.25,
266
+ "activity_rc_frac": 0.25,
267
+ "activity_jaccard": 1.0,
268
+ "activity_precision_rc_vs_gt": 1.0,
269
+ "activity_recall_rc_vs_gt": 1.0,
270
+ "activity_f1_rc_vs_gt": 1.0,
271
+ "class_match_rate": 1.0,
272
+ "doa_angular_error_deg_mean": 127.00635625778561,
273
+ "doa_angular_error_deg_median": 127.1003823514237,
274
+ "distance_mae_m": 0.06403368711471558,
275
+ "top_class_agreement": 1,
276
+ "gt_top_class": "insect",
277
+ "rc_top_class": "insect"
278
+ },
279
+ "sample": "s1_rank10_fsd50k_000123"
280
+ },
281
+ "s1_rank10_insectset_000529": {
282
+ "gt_top_class": "metal_clink",
283
+ "rc_top_class": "water",
284
+ "gt_frames_preview": [
285
+ [
286
+ {
287
+ "track": 0,
288
+ "activity": 0.598,
289
+ "class_idx": 35,
290
+ "class_name": "insect",
291
+ "class_conf": 0.279,
292
+ "azi_deg": -80.4,
293
+ "ele_deg": 23.47,
294
+ "dist_m": 1.418
295
+ }
296
+ ],
297
+ [
298
+ {
299
+ "track": 0,
300
+ "activity": 0.598,
301
+ "class_idx": 35,
302
+ "class_name": "insect",
303
+ "class_conf": 0.28,
304
+ "azi_deg": -80.17,
305
+ "ele_deg": 23.68,
306
+ "dist_m": 1.402
307
+ }
308
+ ],
309
+ [
310
+ {
311
+ "track": 0,
312
+ "activity": 0.6,
313
+ "class_idx": 35,
314
+ "class_name": "insect",
315
+ "class_conf": 0.28,
316
+ "azi_deg": -80.23,
317
+ "ele_deg": 23.71,
318
+ "dist_m": 1.402
319
+ }
320
+ ],
321
+ [
322
+ {
323
+ "track": 0,
324
+ "activity": 0.618,
325
+ "class_idx": 35,
326
+ "class_name": "insect",
327
+ "class_conf": 0.285,
328
+ "azi_deg": -80.66,
329
+ "ele_deg": 23.91,
330
+ "dist_m": 1.4
331
+ }
332
+ ],
333
+ [
334
+ {
335
+ "track": 0,
336
+ "activity": 0.639,
337
+ "class_idx": 12,
338
+ "class_name": "metal_clink",
339
+ "class_conf": 0.283,
340
+ "azi_deg": -81.02,
341
+ "ele_deg": 24.07,
342
+ "dist_m": 1.401
343
+ }
344
+ ]
345
+ ],
346
+ "rc_frames_preview": [
347
+ [
348
+ {
349
+ "track": 0,
350
+ "activity": 0.779,
351
+ "class_idx": 5,
352
+ "class_name": "water",
353
+ "class_conf": 0.883,
354
+ "azi_deg": -92.92,
355
+ "ele_deg": 0.4,
356
+ "dist_m": 2.091
357
+ }
358
+ ],
359
+ [
360
+ {
361
+ "track": 0,
362
+ "activity": 0.779,
363
+ "class_idx": 5,
364
+ "class_name": "water",
365
+ "class_conf": 0.883,
366
+ "azi_deg": -92.83,
367
+ "ele_deg": 0.23,
368
+ "dist_m": 2.068
369
+ }
370
+ ],
371
+ [
372
+ {
373
+ "track": 0,
374
+ "activity": 0.78,
375
+ "class_idx": 5,
376
+ "class_name": "water",
377
+ "class_conf": 0.883,
378
+ "azi_deg": -92.83,
379
+ "ele_deg": 0.22,
380
+ "dist_m": 2.067
381
+ }
382
+ ],
383
+ [
384
+ {
385
+ "track": 0,
386
+ "activity": 0.782,
387
+ "class_idx": 5,
388
+ "class_name": "water",
389
+ "class_conf": 0.884,
390
+ "azi_deg": -92.83,
391
+ "ele_deg": 0.24,
392
+ "dist_m": 2.077
393
+ }
394
+ ],
395
+ [
396
+ {
397
+ "track": 0,
398
+ "activity": 0.784,
399
+ "class_idx": 5,
400
+ "class_name": "water",
401
+ "class_conf": 0.885,
402
+ "azi_deg": -92.89,
403
+ "ele_deg": 0.21,
404
+ "dist_m": 2.069
405
+ }
406
+ ]
407
+ ],
408
+ "comparison": {
409
+ "T_s": 303,
410
+ "activity_gt_frac": 0.18976897689768976,
411
+ "activity_rc_frac": 0.24422442244224424,
412
+ "activity_jaccard": 0.777027027027027,
413
+ "activity_precision_rc_vs_gt": 0.777027027027027,
414
+ "activity_recall_rc_vs_gt": 1.0,
415
+ "activity_f1_rc_vs_gt": 0.8745247148288973,
416
+ "class_match_rate": 0.0,
417
+ "doa_angular_error_deg_mean": 27.849191621326984,
418
+ "doa_angular_error_deg_median": 27.659807577232716,
419
+ "distance_mae_m": 0.5773266553878784,
420
+ "top_class_agreement": 0,
421
+ "gt_top_class": "metal_clink",
422
+ "rc_top_class": "water"
423
+ },
424
+ "sample": "s1_rank10_insectset_000529"
425
+ },
426
+ "s1_rank10_laion_630k_002480": {
427
+ "gt_top_class": "keyboard_instrument",
428
+ "rc_top_class": "keyboard_instrument",
429
+ "gt_frames_preview": [
430
+ [],
431
+ [],
432
+ [],
433
+ [],
434
+ []
435
+ ],
436
+ "rc_frames_preview": [
437
+ [],
438
+ [],
439
+ [],
440
+ [],
441
+ []
442
+ ],
443
+ "comparison": {
444
+ "T_s": 300,
445
+ "activity_gt_frac": 0.23416666666666666,
446
+ "activity_rc_frac": 0.18916666666666668,
447
+ "activity_jaccard": 0.8078291814946619,
448
+ "activity_precision_rc_vs_gt": 1.0,
449
+ "activity_recall_rc_vs_gt": 0.8078291814946619,
450
+ "activity_f1_rc_vs_gt": 0.8937007874015748,
451
+ "class_match_rate": 1.0,
452
+ "doa_angular_error_deg_mean": 91.05960096447677,
453
+ "doa_angular_error_deg_median": 93.2186996184102,
454
+ "distance_mae_m": 0.6004083156585693,
455
+ "top_class_agreement": 1,
456
+ "gt_top_class": "keyboard_instrument",
457
+ "rc_top_class": "keyboard_instrument"
458
+ },
459
+ "sample": "s1_rank10_laion_630k_002480"
460
+ },
461
+ "s1_rank10_temporal_order_001229": {
462
+ "gt_top_class": "human_vocalization",
463
+ "rc_top_class": "aircraft",
464
+ "gt_frames_preview": [
465
+ [],
466
+ [],
467
+ [],
468
+ [],
469
+ [
470
+ {
471
+ "track": 0,
472
+ "activity": 0.529,
473
+ "class_idx": 6,
474
+ "class_name": "human_vocalization",
475
+ "class_conf": 0.807,
476
+ "azi_deg": 90.28,
477
+ "ele_deg": 20.51,
478
+ "dist_m": 1.839
479
+ }
480
+ ]
481
+ ],
482
+ "rc_frames_preview": [
483
+ [
484
+ {
485
+ "track": 0,
486
+ "activity": 0.752,
487
+ "class_idx": 55,
488
+ "class_name": "aircraft",
489
+ "class_conf": 0.72,
490
+ "azi_deg": 83.54,
491
+ "ele_deg": 12.55,
492
+ "dist_m": 1.735
493
+ }
494
+ ],
495
+ [
496
+ {
497
+ "track": 0,
498
+ "activity": 0.847,
499
+ "class_idx": 55,
500
+ "class_name": "aircraft",
501
+ "class_conf": 0.674,
502
+ "azi_deg": 84.23,
503
+ "ele_deg": 12.05,
504
+ "dist_m": 1.718
505
+ }
506
+ ],
507
+ [
508
+ {
509
+ "track": 0,
510
+ "activity": 0.871,
511
+ "class_idx": 55,
512
+ "class_name": "aircraft",
513
+ "class_conf": 0.67,
514
+ "azi_deg": 84.23,
515
+ "ele_deg": 12.17,
516
+ "dist_m": 1.711
517
+ }
518
+ ],
519
+ [
520
+ {
521
+ "track": 0,
522
+ "activity": 0.884,
523
+ "class_idx": 55,
524
+ "class_name": "aircraft",
525
+ "class_conf": 0.647,
526
+ "azi_deg": 84.51,
527
+ "ele_deg": 12.17,
528
+ "dist_m": 1.715
529
+ }
530
+ ],
531
+ [
532
+ {
533
+ "track": 0,
534
+ "activity": 0.905,
535
+ "class_idx": 55,
536
+ "class_name": "aircraft",
537
+ "class_conf": 0.63,
538
+ "azi_deg": 84.8,
539
+ "ele_deg": 12.14,
540
+ "dist_m": 1.723
541
+ }
542
+ ]
543
+ ],
544
+ "comparison": {
545
+ "T_s": 29,
546
+ "activity_gt_frac": 0.15517241379310345,
547
+ "activity_rc_frac": 0.25,
548
+ "activity_jaccard": 0.6206896551724138,
549
+ "activity_precision_rc_vs_gt": 0.6206896551724138,
550
+ "activity_recall_rc_vs_gt": 1.0,
551
+ "activity_f1_rc_vs_gt": 0.7659574468085107,
552
+ "class_match_rate": 0.0,
553
+ "doa_angular_error_deg_mean": 9.779323255239277,
554
+ "doa_angular_error_deg_median": 9.76161397793466,
555
+ "distance_mae_m": 0.07901198416948318,
556
+ "top_class_agreement": 0,
557
+ "gt_top_class": "human_vocalization",
558
+ "rc_top_class": "aircraft"
559
+ },
560
+ "sample": "s1_rank10_temporal_order_001229"
561
+ },
562
+ "s1_rank14_audioset_keyword_000653": {
563
+ "gt_top_class": "human_vocalization",
564
+ "rc_top_class": "singing",
565
+ "gt_frames_preview": [
566
+ [],
567
+ [],
568
+ [],
569
+ [],
570
+ []
571
+ ],
572
+ "rc_frames_preview": [
573
+ [
574
+ {
575
+ "track": 2,
576
+ "activity": 0.68,
577
+ "class_idx": 41,
578
+ "class_name": "singing",
579
+ "class_conf": 0.656,
580
+ "azi_deg": 87.79,
581
+ "ele_deg": 5.37,
582
+ "dist_m": 1.912
583
+ }
584
+ ],
585
+ [
586
+ {
587
+ "track": 2,
588
+ "activity": 0.741,
589
+ "class_idx": 41,
590
+ "class_name": "singing",
591
+ "class_conf": 0.706,
592
+ "azi_deg": 94.37,
593
+ "ele_deg": 6.85,
594
+ "dist_m": 1.895
595
+ }
596
+ ],
597
+ [
598
+ {
599
+ "track": 2,
600
+ "activity": 0.791,
601
+ "class_idx": 41,
602
+ "class_name": "singing",
603
+ "class_conf": 0.716,
604
+ "azi_deg": 101.93,
605
+ "ele_deg": 7.44,
606
+ "dist_m": 1.914
607
+ }
608
+ ],
609
+ [
610
+ {
611
+ "track": 2,
612
+ "activity": 0.799,
613
+ "class_idx": 41,
614
+ "class_name": "singing",
615
+ "class_conf": 0.74,
616
+ "azi_deg": 102.74,
617
+ "ele_deg": 8.14,
618
+ "dist_m": 1.925
619
+ }
620
+ ],
621
+ [
622
+ {
623
+ "track": 2,
624
+ "activity": 0.815,
625
+ "class_idx": 41,
626
+ "class_name": "singing",
627
+ "class_conf": 0.752,
628
+ "azi_deg": 102.41,
629
+ "ele_deg": 7.91,
630
+ "dist_m": 1.932
631
+ }
632
+ ]
633
+ ],
634
+ "comparison": {
635
+ "T_s": 103,
636
+ "activity_gt_frac": 0.2087378640776699,
637
+ "activity_rc_frac": 0.23058252427184467,
638
+ "activity_jaccard": 0.865979381443299,
639
+ "activity_precision_rc_vs_gt": 0.8842105263157894,
640
+ "activity_recall_rc_vs_gt": 0.9767441860465116,
641
+ "activity_f1_rc_vs_gt": 0.9281767955801105,
642
+ "class_match_rate": 0.0,
643
+ "doa_angular_error_deg_mean": 42.55166666428091,
644
+ "doa_angular_error_deg_median": 43.37530110194772,
645
+ "distance_mae_m": 0.04641605168581009,
646
+ "top_class_agreement": 0,
647
+ "gt_top_class": "human_vocalization",
648
+ "rc_top_class": "singing"
649
+ },
650
+ "sample": "s1_rank14_audioset_keyword_000653"
651
+ },
652
+ "s1_rank14_fsd50k_000024": {
653
+ "gt_top_class": "keyboard_instrument",
654
+ "rc_top_class": "keyboard_instrument",
655
+ "gt_frames_preview": [
656
+ [
657
+ {
658
+ "track": 0,
659
+ "activity": 0.741,
660
+ "class_idx": 7,
661
+ "class_name": "keyboard_instrument",
662
+ "class_conf": 0.85,
663
+ "azi_deg": 106.65,
664
+ "ele_deg": 15.69,
665
+ "dist_m": 1.944
666
+ }
667
+ ],
668
+ [
669
+ {
670
+ "track": 0,
671
+ "activity": 0.833,
672
+ "class_idx": 7,
673
+ "class_name": "keyboard_instrument",
674
+ "class_conf": 0.853,
675
+ "azi_deg": 108.18,
676
+ "ele_deg": 16.01,
677
+ "dist_m": 1.926
678
+ }
679
+ ],
680
+ [
681
+ {
682
+ "track": 0,
683
+ "activity": 0.83,
684
+ "class_idx": 7,
685
+ "class_name": "keyboard_instrument",
686
+ "class_conf": 0.849,
687
+ "azi_deg": 108.66,
688
+ "ele_deg": 15.48,
689
+ "dist_m": 1.927
690
+ }
691
+ ],
692
+ [
693
+ {
694
+ "track": 0,
695
+ "activity": 0.837,
696
+ "class_idx": 7,
697
+ "class_name": "keyboard_instrument",
698
+ "class_conf": 0.851,
699
+ "azi_deg": 108.29,
700
+ "ele_deg": 15.24,
701
+ "dist_m": 1.927
702
+ }
703
+ ],
704
+ [
705
+ {
706
+ "track": 0,
707
+ "activity": 0.844,
708
+ "class_idx": 7,
709
+ "class_name": "keyboard_instrument",
710
+ "class_conf": 0.849,
711
+ "azi_deg": 108.11,
712
+ "ele_deg": 14.97,
713
+ "dist_m": 1.931
714
+ }
715
+ ]
716
+ ],
717
+ "rc_frames_preview": [
718
+ [
719
+ {
720
+ "track": 0,
721
+ "activity": 0.737,
722
+ "class_idx": 7,
723
+ "class_name": "keyboard_instrument",
724
+ "class_conf": 0.704,
725
+ "azi_deg": 23.81,
726
+ "ele_deg": -7.93,
727
+ "dist_m": 2.056
728
+ }
729
+ ],
730
+ [
731
+ {
732
+ "track": 0,
733
+ "activity": 0.726,
734
+ "class_idx": 7,
735
+ "class_name": "keyboard_instrument",
736
+ "class_conf": 0.724,
737
+ "azi_deg": 27.52,
738
+ "ele_deg": -7.4,
739
+ "dist_m": 2.089
740
+ }
741
+ ],
742
+ [
743
+ {
744
+ "track": 0,
745
+ "activity": 0.679,
746
+ "class_idx": 7,
747
+ "class_name": "keyboard_instrument",
748
+ "class_conf": 0.728,
749
+ "azi_deg": 31.46,
750
+ "ele_deg": -7.61,
751
+ "dist_m": 2.161
752
+ }
753
+ ],
754
+ [
755
+ {
756
+ "track": 0,
757
+ "activity": 0.69,
758
+ "class_idx": 7,
759
+ "class_name": "keyboard_instrument",
760
+ "class_conf": 0.725,
761
+ "azi_deg": 34.22,
762
+ "ele_deg": -7.9,
763
+ "dist_m": 2.199
764
+ }
765
+ ],
766
+ [
767
+ {
768
+ "track": 0,
769
+ "activity": 0.681,
770
+ "class_idx": 7,
771
+ "class_name": "keyboard_instrument",
772
+ "class_conf": 0.721,
773
+ "azi_deg": 35.02,
774
+ "ele_deg": -8.41,
775
+ "dist_m": 2.214
776
+ }
777
+ ]
778
+ ],
779
+ "comparison": {
780
+ "T_s": 24,
781
+ "activity_gt_frac": 0.25,
782
+ "activity_rc_frac": 0.14583333333333334,
783
+ "activity_jaccard": 0.5833333333333334,
784
+ "activity_precision_rc_vs_gt": 1.0,
785
+ "activity_recall_rc_vs_gt": 0.5833333333333334,
786
+ "activity_f1_rc_vs_gt": 0.7368421052631579,
787
+ "class_match_rate": 1.0,
788
+ "doa_angular_error_deg_mean": 76.69808704781899,
789
+ "doa_angular_error_deg_median": 75.03630868300726,
790
+ "distance_mae_m": 0.27358776330947876,
791
+ "top_class_agreement": 1,
792
+ "gt_top_class": "keyboard_instrument",
793
+ "rc_top_class": "keyboard_instrument"
794
+ },
795
+ "sample": "s1_rank14_fsd50k_000024"
796
+ },
797
+ "s1_rank14_temporal_order_003256": {
798
+ "gt_top_class": "keyboard_instrument",
799
+ "rc_top_class": "musical_instrument",
800
+ "gt_frames_preview": [
801
+ [],
802
+ [],
803
+ [],
804
+ [],
805
+ []
806
+ ],
807
+ "rc_frames_preview": [
808
+ [
809
+ {
810
+ "track": 0,
811
+ "activity": 0.572,
812
+ "class_idx": 28,
813
+ "class_name": "musical_instrument",
814
+ "class_conf": 0.51,
815
+ "azi_deg": -90.21,
816
+ "ele_deg": -0.92,
817
+ "dist_m": 2.995
818
+ }
819
+ ],
820
+ [
821
+ {
822
+ "track": 0,
823
+ "activity": 0.597,
824
+ "class_idx": 28,
825
+ "class_name": "musical_instrument",
826
+ "class_conf": 0.502,
827
+ "azi_deg": -92.91,
828
+ "ele_deg": 0.17,
829
+ "dist_m": 3.018
830
+ }
831
+ ],
832
+ [
833
+ {
834
+ "track": 0,
835
+ "activity": 0.576,
836
+ "class_idx": 28,
837
+ "class_name": "musical_instrument",
838
+ "class_conf": 0.509,
839
+ "azi_deg": -94.17,
840
+ "ele_deg": -0.5,
841
+ "dist_m": 3.008
842
+ }
843
+ ],
844
+ [
845
+ {
846
+ "track": 0,
847
+ "activity": 0.577,
848
+ "class_idx": 28,
849
+ "class_name": "musical_instrument",
850
+ "class_conf": 0.511,
851
+ "azi_deg": -94.64,
852
+ "ele_deg": -1.03,
853
+ "dist_m": 3.01
854
+ }
855
+ ],
856
+ [
857
+ {
858
+ "track": 0,
859
+ "activity": 0.584,
860
+ "class_idx": 28,
861
+ "class_name": "musical_instrument",
862
+ "class_conf": 0.51,
863
+ "azi_deg": -94.84,
864
+ "ele_deg": -1.36,
865
+ "dist_m": 3.007
866
+ }
867
+ ]
868
+ ],
869
+ "comparison": {
870
+ "T_s": 13,
871
+ "activity_gt_frac": 0.0,
872
+ "activity_rc_frac": 0.25,
873
+ "activity_jaccard": 0.0,
874
+ "activity_precision_rc_vs_gt": 0.0,
875
+ "activity_recall_rc_vs_gt": 0.0,
876
+ "activity_f1_rc_vs_gt": 0.0,
877
+ "class_match_rate": NaN,
878
+ "doa_angular_error_deg_mean": NaN,
879
+ "doa_angular_error_deg_median": NaN,
880
+ "distance_mae_m": NaN,
881
+ "top_class_agreement": 0,
882
+ "gt_top_class": "keyboard_instrument",
883
+ "rc_top_class": "musical_instrument"
884
+ },
885
+ "sample": "s1_rank14_temporal_order_003256"
886
+ },
887
+ "s1_rank15_fsd50k_000489": {
888
+ "gt_top_class": "telephone_alarm",
889
+ "rc_top_class": "insect",
890
+ "gt_frames_preview": [
891
+ [],
892
+ [
893
+ {
894
+ "track": 0,
895
+ "activity": 0.512,
896
+ "class_idx": 23,
897
+ "class_name": "telephone_alarm",
898
+ "class_conf": 0.862,
899
+ "azi_deg": 48.36,
900
+ "ele_deg": 28.03,
901
+ "dist_m": 1.952
902
+ }
903
+ ],
904
+ [
905
+ {
906
+ "track": 0,
907
+ "activity": 0.702,
908
+ "class_idx": 23,
909
+ "class_name": "telephone_alarm",
910
+ "class_conf": 0.868,
911
+ "azi_deg": 48.63,
912
+ "ele_deg": 28.02,
913
+ "dist_m": 1.937
914
+ }
915
+ ],
916
+ [
917
+ {
918
+ "track": 0,
919
+ "activity": 0.846,
920
+ "class_idx": 23,
921
+ "class_name": "telephone_alarm",
922
+ "class_conf": 0.871,
923
+ "azi_deg": 49.05,
924
+ "ele_deg": 28.02,
925
+ "dist_m": 1.925
926
+ }
927
+ ],
928
+ [
929
+ {
930
+ "track": 0,
931
+ "activity": 0.875,
932
+ "class_idx": 23,
933
+ "class_name": "telephone_alarm",
934
+ "class_conf": 0.874,
935
+ "azi_deg": 49.06,
936
+ "ele_deg": 27.97,
937
+ "dist_m": 1.918
938
+ }
939
+ ]
940
+ ],
941
+ "rc_frames_preview": [
942
+ [
943
+ {
944
+ "track": 1,
945
+ "activity": 0.65,
946
+ "class_idx": 35,
947
+ "class_name": "insect",
948
+ "class_conf": 0.901,
949
+ "azi_deg": -93.81,
950
+ "ele_deg": -6.54,
951
+ "dist_m": 1.97
952
+ }
953
+ ],
954
+ [
955
+ {
956
+ "track": 1,
957
+ "activity": 0.66,
958
+ "class_idx": 35,
959
+ "class_name": "insect",
960
+ "class_conf": 0.901,
961
+ "azi_deg": -94.94,
962
+ "ele_deg": -9.26,
963
+ "dist_m": 1.977
964
+ }
965
+ ],
966
+ [
967
+ {
968
+ "track": 1,
969
+ "activity": 0.679,
970
+ "class_idx": 35,
971
+ "class_name": "insect",
972
+ "class_conf": 0.907,
973
+ "azi_deg": -103.52,
974
+ "ele_deg": -12.06,
975
+ "dist_m": 1.975
976
+ }
977
+ ],
978
+ [
979
+ {
980
+ "track": 1,
981
+ "activity": 0.685,
982
+ "class_idx": 35,
983
+ "class_name": "insect",
984
+ "class_conf": 0.916,
985
+ "azi_deg": -114.9,
986
+ "ele_deg": -11.86,
987
+ "dist_m": 1.953
988
+ }
989
+ ],
990
+ [
991
+ {
992
+ "track": 1,
993
+ "activity": 0.662,
994
+ "class_idx": 35,
995
+ "class_name": "insect",
996
+ "class_conf": 0.923,
997
+ "azi_deg": -115.02,
998
+ "ele_deg": -13.05,
999
+ "dist_m": 1.963
1000
+ }
1001
+ ]
1002
+ ],
1003
+ "comparison": {
1004
+ "T_s": 155,
1005
+ "activity_gt_frac": 0.22258064516129034,
1006
+ "activity_rc_frac": 0.4338709677419355,
1007
+ "activity_jaccard": 0.0,
1008
+ "activity_precision_rc_vs_gt": 0.0,
1009
+ "activity_recall_rc_vs_gt": 0.0,
1010
+ "activity_f1_rc_vs_gt": 0.0,
1011
+ "class_match_rate": NaN,
1012
+ "doa_angular_error_deg_mean": NaN,
1013
+ "doa_angular_error_deg_median": NaN,
1014
+ "distance_mae_m": NaN,
1015
+ "top_class_agreement": 0,
1016
+ "gt_top_class": "telephone_alarm",
1017
+ "rc_top_class": "insect"
1018
+ },
1019
+ "sample": "s1_rank15_fsd50k_000489"
1020
+ },
1021
+ "s1_rank17_audioset_keyword_001725": {
1022
+ "gt_top_class": "guitar",
1023
+ "rc_top_class": "guitar",
1024
+ "gt_frames_preview": [
1025
+ [
1026
+ {
1027
+ "track": 0,
1028
+ "activity": 0.978,
1029
+ "class_idx": 2,
1030
+ "class_name": "guitar",
1031
+ "class_conf": 0.871,
1032
+ "azi_deg": -6.63,
1033
+ "ele_deg": 1.21,
1034
+ "dist_m": 2.677
1035
+ }
1036
+ ],
1037
+ [
1038
+ {
1039
+ "track": 0,
1040
+ "activity": 0.979,
1041
+ "class_idx": 2,
1042
+ "class_name": "guitar",
1043
+ "class_conf": 0.871,
1044
+ "azi_deg": -6.29,
1045
+ "ele_deg": 0.95,
1046
+ "dist_m": 2.66
1047
+ }
1048
+ ],
1049
+ [
1050
+ {
1051
+ "track": 0,
1052
+ "activity": 0.982,
1053
+ "class_idx": 2,
1054
+ "class_name": "guitar",
1055
+ "class_conf": 0.876,
1056
+ "azi_deg": -6.28,
1057
+ "ele_deg": 0.53,
1058
+ "dist_m": 2.64
1059
+ }
1060
+ ],
1061
+ [
1062
+ {
1063
+ "track": 0,
1064
+ "activity": 0.982,
1065
+ "class_idx": 2,
1066
+ "class_name": "guitar",
1067
+ "class_conf": 0.874,
1068
+ "azi_deg": -7.43,
1069
+ "ele_deg": 0.61,
1070
+ "dist_m": 2.621
1071
+ }
1072
+ ],
1073
+ [
1074
+ {
1075
+ "track": 0,
1076
+ "activity": 0.983,
1077
+ "class_idx": 2,
1078
+ "class_name": "guitar",
1079
+ "class_conf": 0.879,
1080
+ "azi_deg": -8.48,
1081
+ "ele_deg": 0.63,
1082
+ "dist_m": 2.606
1083
+ }
1084
+ ]
1085
+ ],
1086
+ "rc_frames_preview": [
1087
+ [
1088
+ {
1089
+ "track": 0,
1090
+ "activity": 0.904,
1091
+ "class_idx": 41,
1092
+ "class_name": "singing",
1093
+ "class_conf": 0.444,
1094
+ "azi_deg": 62.71,
1095
+ "ele_deg": 12.7,
1096
+ "dist_m": 2.582
1097
+ }
1098
+ ],
1099
+ [
1100
+ {
1101
+ "track": 0,
1102
+ "activity": 0.876,
1103
+ "class_idx": 41,
1104
+ "class_name": "singing",
1105
+ "class_conf": 0.436,
1106
+ "azi_deg": 63.33,
1107
+ "ele_deg": 12.37,
1108
+ "dist_m": 2.572
1109
+ }
1110
+ ],
1111
+ [
1112
+ {
1113
+ "track": 0,
1114
+ "activity": 0.822,
1115
+ "class_idx": 2,
1116
+ "class_name": "guitar",
1117
+ "class_conf": 0.406,
1118
+ "azi_deg": 64.24,
1119
+ "ele_deg": 12.0,
1120
+ "dist_m": 2.561
1121
+ }
1122
+ ],
1123
+ [
1124
+ {
1125
+ "track": 0,
1126
+ "activity": 0.799,
1127
+ "class_idx": 2,
1128
+ "class_name": "guitar",
1129
+ "class_conf": 0.484,
1130
+ "azi_deg": 64.22,
1131
+ "ele_deg": 12.22,
1132
+ "dist_m": 2.543
1133
+ }
1134
+ ],
1135
+ [
1136
+ {
1137
+ "track": 0,
1138
+ "activity": 0.797,
1139
+ "class_idx": 2,
1140
+ "class_name": "guitar",
1141
+ "class_conf": 0.541,
1142
+ "azi_deg": 64.26,
1143
+ "ele_deg": 12.28,
1144
+ "dist_m": 2.532
1145
+ }
1146
+ ]
1147
+ ],
1148
+ "comparison": {
1149
+ "T_s": 104,
1150
+ "activity_gt_frac": 0.24759615384615385,
1151
+ "activity_rc_frac": 0.34375,
1152
+ "activity_jaccard": 0.24873096446700507,
1153
+ "activity_precision_rc_vs_gt": 0.34265734265734266,
1154
+ "activity_recall_rc_vs_gt": 0.47572815533980584,
1155
+ "activity_f1_rc_vs_gt": 0.39837398373983735,
1156
+ "class_match_rate": 0.9591836734693877,
1157
+ "doa_angular_error_deg_mean": 76.3891275171291,
1158
+ "doa_angular_error_deg_median": 76.95587494145566,
1159
+ "distance_mae_m": 0.07019432634115219,
1160
+ "top_class_agreement": 1,
1161
+ "gt_top_class": "guitar",
1162
+ "rc_top_class": "guitar"
1163
+ },
1164
+ "sample": "s1_rank17_audioset_keyword_001725"
1165
+ },
1166
+ "s1_rank18_opensource_emilia_zh_rawdata_English_000545": {
1167
+ "gt_top_class": "speech",
1168
+ "rc_top_class": "speech",
1169
+ "gt_frames_preview": [
1170
+ [
1171
+ {
1172
+ "track": 0,
1173
+ "activity": 0.616,
1174
+ "class_idx": 16,
1175
+ "class_name": "speech",
1176
+ "class_conf": 0.859,
1177
+ "azi_deg": 54.28,
1178
+ "ele_deg": 32.99,
1179
+ "dist_m": 1.793
1180
+ }
1181
+ ],
1182
+ [
1183
+ {
1184
+ "track": 0,
1185
+ "activity": 0.644,
1186
+ "class_idx": 16,
1187
+ "class_name": "speech",
1188
+ "class_conf": 0.864,
1189
+ "azi_deg": 54.3,
1190
+ "ele_deg": 32.91,
1191
+ "dist_m": 1.788
1192
+ }
1193
+ ],
1194
+ [
1195
+ {
1196
+ "track": 0,
1197
+ "activity": 0.729,
1198
+ "class_idx": 16,
1199
+ "class_name": "speech",
1200
+ "class_conf": 0.87,
1201
+ "azi_deg": 54.21,
1202
+ "ele_deg": 32.44,
1203
+ "dist_m": 1.792
1204
+ }
1205
+ ],
1206
+ [
1207
+ {
1208
+ "track": 0,
1209
+ "activity": 0.809,
1210
+ "class_idx": 16,
1211
+ "class_name": "speech",
1212
+ "class_conf": 0.876,
1213
+ "azi_deg": 54.15,
1214
+ "ele_deg": 32.04,
1215
+ "dist_m": 1.808
1216
+ }
1217
+ ],
1218
+ [
1219
+ {
1220
+ "track": 0,
1221
+ "activity": 0.824,
1222
+ "class_idx": 16,
1223
+ "class_name": "speech",
1224
+ "class_conf": 0.877,
1225
+ "azi_deg": 53.89,
1226
+ "ele_deg": 31.84,
1227
+ "dist_m": 1.814
1228
+ }
1229
+ ]
1230
+ ],
1231
+ "rc_frames_preview": [
1232
+ [
1233
+ {
1234
+ "track": 0,
1235
+ "activity": 0.875,
1236
+ "class_idx": 16,
1237
+ "class_name": "speech",
1238
+ "class_conf": 0.88,
1239
+ "azi_deg": 62.47,
1240
+ "ele_deg": 26.38,
1241
+ "dist_m": 2.243
1242
+ }
1243
+ ],
1244
+ [
1245
+ {
1246
+ "track": 0,
1247
+ "activity": 0.886,
1248
+ "class_idx": 16,
1249
+ "class_name": "speech",
1250
+ "class_conf": 0.883,
1251
+ "azi_deg": 62.83,
1252
+ "ele_deg": 26.17,
1253
+ "dist_m": 2.242
1254
+ }
1255
+ ],
1256
+ [
1257
+ {
1258
+ "track": 0,
1259
+ "activity": 0.916,
1260
+ "class_idx": 16,
1261
+ "class_name": "speech",
1262
+ "class_conf": 0.887,
1263
+ "azi_deg": 62.98,
1264
+ "ele_deg": 25.38,
1265
+ "dist_m": 2.255
1266
+ }
1267
+ ],
1268
+ [
1269
+ {
1270
+ "track": 0,
1271
+ "activity": 0.941,
1272
+ "class_idx": 16,
1273
+ "class_name": "speech",
1274
+ "class_conf": 0.891,
1275
+ "azi_deg": 62.94,
1276
+ "ele_deg": 24.93,
1277
+ "dist_m": 2.265
1278
+ }
1279
+ ],
1280
+ [
1281
+ {
1282
+ "track": 0,
1283
+ "activity": 0.945,
1284
+ "class_idx": 16,
1285
+ "class_name": "speech",
1286
+ "class_conf": 0.891,
1287
+ "azi_deg": 62.55,
1288
+ "ele_deg": 25.05,
1289
+ "dist_m": 2.249
1290
+ }
1291
+ ]
1292
+ ],
1293
+ "comparison": {
1294
+ "T_s": 147,
1295
+ "activity_gt_frac": 0.12585034013605442,
1296
+ "activity_rc_frac": 0.12414965986394558,
1297
+ "activity_jaccard": 0.9864864864864865,
1298
+ "activity_precision_rc_vs_gt": 1.0,
1299
+ "activity_recall_rc_vs_gt": 0.9864864864864865,
1300
+ "activity_f1_rc_vs_gt": 0.9931972789115647,
1301
+ "class_match_rate": 1.0,
1302
+ "doa_angular_error_deg_mean": 8.14774736211984,
1303
+ "doa_angular_error_deg_median": 7.7812438074287185,
1304
+ "distance_mae_m": 0.3401370048522949,
1305
+ "top_class_agreement": 1,
1306
+ "gt_top_class": "speech",
1307
+ "rc_top_class": "speech"
1308
+ },
1309
+ "sample": "s1_rank18_opensource_emilia_zh_rawdata_English_000545"
1310
+ },
1311
+ "s1_rank20_laion_630k_002181": {
1312
+ "gt_top_class": "drum",
1313
+ "rc_top_class": "drum",
1314
+ "gt_frames_preview": [
1315
+ [
1316
+ {
1317
+ "track": 0,
1318
+ "activity": 0.895,
1319
+ "class_idx": 4,
1320
+ "class_name": "drum",
1321
+ "class_conf": 0.562,
1322
+ "azi_deg": -117.81,
1323
+ "ele_deg": 27.03,
1324
+ "dist_m": 1.466
1325
+ }
1326
+ ],
1327
+ [
1328
+ {
1329
+ "track": 0,
1330
+ "activity": 0.896,
1331
+ "class_idx": 4,
1332
+ "class_name": "drum",
1333
+ "class_conf": 0.568,
1334
+ "azi_deg": -118.0,
1335
+ "ele_deg": 26.75,
1336
+ "dist_m": 1.479
1337
+ }
1338
+ ],
1339
+ [
1340
+ {
1341
+ "track": 0,
1342
+ "activity": 0.908,
1343
+ "class_idx": 4,
1344
+ "class_name": "drum",
1345
+ "class_conf": 0.578,
1346
+ "azi_deg": -118.28,
1347
+ "ele_deg": 26.29,
1348
+ "dist_m": 1.475
1349
+ }
1350
+ ],
1351
+ [
1352
+ {
1353
+ "track": 0,
1354
+ "activity": 0.922,
1355
+ "class_idx": 4,
1356
+ "class_name": "drum",
1357
+ "class_conf": 0.576,
1358
+ "azi_deg": -118.63,
1359
+ "ele_deg": 25.93,
1360
+ "dist_m": 1.462
1361
+ }
1362
+ ],
1363
+ [
1364
+ {
1365
+ "track": 0,
1366
+ "activity": 0.933,
1367
+ "class_idx": 4,
1368
+ "class_name": "drum",
1369
+ "class_conf": 0.572,
1370
+ "azi_deg": -118.84,
1371
+ "ele_deg": 25.37,
1372
+ "dist_m": 1.459
1373
+ }
1374
+ ]
1375
+ ],
1376
+ "rc_frames_preview": [
1377
+ [
1378
+ {
1379
+ "track": 0,
1380
+ "activity": 0.879,
1381
+ "class_idx": 4,
1382
+ "class_name": "drum",
1383
+ "class_conf": 0.897,
1384
+ "azi_deg": -107.94,
1385
+ "ele_deg": 30.18,
1386
+ "dist_m": 1.241
1387
+ }
1388
+ ],
1389
+ [
1390
+ {
1391
+ "track": 0,
1392
+ "activity": 0.884,
1393
+ "class_idx": 4,
1394
+ "class_name": "drum",
1395
+ "class_conf": 0.894,
1396
+ "azi_deg": -108.13,
1397
+ "ele_deg": 30.04,
1398
+ "dist_m": 1.255
1399
+ }
1400
+ ],
1401
+ [
1402
+ {
1403
+ "track": 0,
1404
+ "activity": 0.909,
1405
+ "class_idx": 4,
1406
+ "class_name": "drum",
1407
+ "class_conf": 0.9,
1408
+ "azi_deg": -108.26,
1409
+ "ele_deg": 29.87,
1410
+ "dist_m": 1.251
1411
+ }
1412
+ ],
1413
+ [
1414
+ {
1415
+ "track": 0,
1416
+ "activity": 0.926,
1417
+ "class_idx": 4,
1418
+ "class_name": "drum",
1419
+ "class_conf": 0.914,
1420
+ "azi_deg": -108.49,
1421
+ "ele_deg": 29.94,
1422
+ "dist_m": 1.247
1423
+ }
1424
+ ],
1425
+ [
1426
+ {
1427
+ "track": 0,
1428
+ "activity": 0.935,
1429
+ "class_idx": 4,
1430
+ "class_name": "drum",
1431
+ "class_conf": 0.917,
1432
+ "azi_deg": -108.61,
1433
+ "ele_deg": 29.76,
1434
+ "dist_m": 1.246
1435
+ }
1436
+ ]
1437
+ ],
1438
+ "comparison": {
1439
+ "T_s": 174,
1440
+ "activity_gt_frac": 0.27873563218390807,
1441
+ "activity_rc_frac": 0.1867816091954023,
1442
+ "activity_jaccard": 0.5,
1443
+ "activity_precision_rc_vs_gt": 0.8307692307692308,
1444
+ "activity_recall_rc_vs_gt": 0.5567010309278351,
1445
+ "activity_f1_rc_vs_gt": 0.6666666666666667,
1446
+ "class_match_rate": 1.0,
1447
+ "doa_angular_error_deg_mean": 7.890753480227976,
1448
+ "doa_angular_error_deg_median": 7.5300053632301385,
1449
+ "distance_mae_m": 0.17731787264347076,
1450
+ "top_class_agreement": 1,
1451
+ "gt_top_class": "drum",
1452
+ "rc_top_class": "drum"
1453
+ },
1454
+ "sample": "s1_rank20_laion_630k_002181"
1455
+ },
1456
+ "s1_rank20_vocalsound_000310": {
1457
+ "gt_top_class": "laughter",
1458
+ "rc_top_class": "laughter",
1459
+ "gt_frames_preview": [
1460
+ [],
1461
+ [
1462
+ {
1463
+ "track": 0,
1464
+ "activity": 0.514,
1465
+ "class_idx": 14,
1466
+ "class_name": "laughter",
1467
+ "class_conf": 0.735,
1468
+ "azi_deg": -54.88,
1469
+ "ele_deg": 6.41,
1470
+ "dist_m": 2.726
1471
+ }
1472
+ ],
1473
+ [
1474
+ {
1475
+ "track": 0,
1476
+ "activity": 0.505,
1477
+ "class_idx": 14,
1478
+ "class_name": "laughter",
1479
+ "class_conf": 0.779,
1480
+ "azi_deg": -59.56,
1481
+ "ele_deg": 7.71,
1482
+ "dist_m": 2.714
1483
+ }
1484
+ ],
1485
+ [
1486
+ {
1487
+ "track": 0,
1488
+ "activity": 0.532,
1489
+ "class_idx": 14,
1490
+ "class_name": "laughter",
1491
+ "class_conf": 0.788,
1492
+ "azi_deg": -63.85,
1493
+ "ele_deg": 9.31,
1494
+ "dist_m": 2.698
1495
+ }
1496
+ ],
1497
+ [
1498
+ {
1499
+ "track": 0,
1500
+ "activity": 0.535,
1501
+ "class_idx": 14,
1502
+ "class_name": "laughter",
1503
+ "class_conf": 0.81,
1504
+ "azi_deg": -66.25,
1505
+ "ele_deg": 10.24,
1506
+ "dist_m": 2.685
1507
+ }
1508
+ ]
1509
+ ],
1510
+ "rc_frames_preview": [
1511
+ [
1512
+ {
1513
+ "track": 1,
1514
+ "activity": 0.719,
1515
+ "class_idx": 9,
1516
+ "class_name": "tool",
1517
+ "class_conf": 0.425,
1518
+ "azi_deg": -70.36,
1519
+ "ele_deg": 36.59,
1520
+ "dist_m": 1.33
1521
+ }
1522
+ ],
1523
+ [
1524
+ {
1525
+ "track": 0,
1526
+ "activity": 0.536,
1527
+ "class_idx": 14,
1528
+ "class_name": "laughter",
1529
+ "class_conf": 0.841,
1530
+ "azi_deg": -96.61,
1531
+ "ele_deg": 6.14,
1532
+ "dist_m": 3.096
1533
+ },
1534
+ {
1535
+ "track": 1,
1536
+ "activity": 0.517,
1537
+ "class_idx": 9,
1538
+ "class_name": "tool",
1539
+ "class_conf": 0.496,
1540
+ "azi_deg": -75.77,
1541
+ "ele_deg": 36.41,
1542
+ "dist_m": 1.321
1543
+ }
1544
+ ],
1545
+ [
1546
+ {
1547
+ "track": 0,
1548
+ "activity": 0.594,
1549
+ "class_idx": 14,
1550
+ "class_name": "laughter",
1551
+ "class_conf": 0.846,
1552
+ "azi_deg": -96.2,
1553
+ "ele_deg": 6.22,
1554
+ "dist_m": 3.115
1555
+ },
1556
+ {
1557
+ "track": 3,
1558
+ "activity": 0.54,
1559
+ "class_idx": 14,
1560
+ "class_name": "laughter",
1561
+ "class_conf": 0.707,
1562
+ "azi_deg": 11.53,
1563
+ "ele_deg": 7.56,
1564
+ "dist_m": 4.027
1565
+ }
1566
+ ],
1567
+ [
1568
+ {
1569
+ "track": 0,
1570
+ "activity": 0.632,
1571
+ "class_idx": 14,
1572
+ "class_name": "laughter",
1573
+ "class_conf": 0.839,
1574
+ "azi_deg": -96.32,
1575
+ "ele_deg": 6.57,
1576
+ "dist_m": 3.147
1577
+ },
1578
+ {
1579
+ "track": 3,
1580
+ "activity": 0.575,
1581
+ "class_idx": 14,
1582
+ "class_name": "laughter",
1583
+ "class_conf": 0.721,
1584
+ "azi_deg": 31.66,
1585
+ "ele_deg": 3.11,
1586
+ "dist_m": 4.065
1587
+ }
1588
+ ],
1589
+ [
1590
+ {
1591
+ "track": 0,
1592
+ "activity": 0.64,
1593
+ "class_idx": 14,
1594
+ "class_name": "laughter",
1595
+ "class_conf": 0.847,
1596
+ "azi_deg": -96.3,
1597
+ "ele_deg": 6.96,
1598
+ "dist_m": 3.136
1599
+ },
1600
+ {
1601
+ "track": 3,
1602
+ "activity": 0.598,
1603
+ "class_idx": 14,
1604
+ "class_name": "laughter",
1605
+ "class_conf": 0.744,
1606
+ "azi_deg": 40.4,
1607
+ "ele_deg": 0.67,
1608
+ "dist_m": 4.029
1609
+ }
1610
+ ]
1611
+ ],
1612
+ "comparison": {
1613
+ "T_s": 53,
1614
+ "activity_gt_frac": 0.37735849056603776,
1615
+ "activity_rc_frac": 0.32547169811320753,
1616
+ "activity_jaccard": 0.3925233644859813,
1617
+ "activity_precision_rc_vs_gt": 0.6086956521739131,
1618
+ "activity_recall_rc_vs_gt": 0.525,
1619
+ "activity_f1_rc_vs_gt": 0.563758389261745,
1620
+ "class_match_rate": 1.0,
1621
+ "doa_angular_error_deg_mean": 67.5833671782668,
1622
+ "doa_angular_error_deg_median": 70.11200394426966,
1623
+ "distance_mae_m": 0.394989550113678,
1624
+ "top_class_agreement": 1,
1625
+ "gt_top_class": "laughter",
1626
+ "rc_top_class": "laughter"
1627
+ },
1628
+ "sample": "s1_rank20_vocalsound_000310"
1629
+ },
1630
+ "s1_rank21_insectset_000328": {
1631
+ "gt_top_class": "water",
1632
+ "rc_top_class": "water",
1633
+ "gt_frames_preview": [
1634
+ [
1635
+ {
1636
+ "track": 0,
1637
+ "activity": 0.578,
1638
+ "class_idx": 5,
1639
+ "class_name": "water",
1640
+ "class_conf": 0.565,
1641
+ "azi_deg": 143.21,
1642
+ "ele_deg": 5.5,
1643
+ "dist_m": 2.413
1644
+ }
1645
+ ],
1646
+ [
1647
+ {
1648
+ "track": 0,
1649
+ "activity": 0.578,
1650
+ "class_idx": 5,
1651
+ "class_name": "water",
1652
+ "class_conf": 0.542,
1653
+ "azi_deg": 142.53,
1654
+ "ele_deg": 5.38,
1655
+ "dist_m": 2.417
1656
+ }
1657
+ ],
1658
+ [
1659
+ {
1660
+ "track": 0,
1661
+ "activity": 0.581,
1662
+ "class_idx": 5,
1663
+ "class_name": "water",
1664
+ "class_conf": 0.542,
1665
+ "azi_deg": 142.57,
1666
+ "ele_deg": 5.38,
1667
+ "dist_m": 2.416
1668
+ }
1669
+ ],
1670
+ [
1671
+ {
1672
+ "track": 0,
1673
+ "activity": 0.604,
1674
+ "class_idx": 5,
1675
+ "class_name": "water",
1676
+ "class_conf": 0.534,
1677
+ "azi_deg": 143.0,
1678
+ "ele_deg": 5.6,
1679
+ "dist_m": 2.409
1680
+ }
1681
+ ],
1682
+ [
1683
+ {
1684
+ "track": 0,
1685
+ "activity": 0.629,
1686
+ "class_idx": 5,
1687
+ "class_name": "water",
1688
+ "class_conf": 0.533,
1689
+ "azi_deg": 143.29,
1690
+ "ele_deg": 5.62,
1691
+ "dist_m": 2.402
1692
+ }
1693
+ ]
1694
+ ],
1695
+ "rc_frames_preview": [
1696
+ [
1697
+ {
1698
+ "track": 0,
1699
+ "activity": 0.843,
1700
+ "class_idx": 5,
1701
+ "class_name": "water",
1702
+ "class_conf": 0.908,
1703
+ "azi_deg": 27.7,
1704
+ "ele_deg": -3.23,
1705
+ "dist_m": 2.458
1706
+ }
1707
+ ],
1708
+ [
1709
+ {
1710
+ "track": 0,
1711
+ "activity": 0.843,
1712
+ "class_idx": 5,
1713
+ "class_name": "water",
1714
+ "class_conf": 0.909,
1715
+ "azi_deg": 29.56,
1716
+ "ele_deg": -2.94,
1717
+ "dist_m": 2.43
1718
+ }
1719
+ ],
1720
+ [
1721
+ {
1722
+ "track": 0,
1723
+ "activity": 0.844,
1724
+ "class_idx": 5,
1725
+ "class_name": "water",
1726
+ "class_conf": 0.909,
1727
+ "azi_deg": 29.3,
1728
+ "ele_deg": -2.97,
1729
+ "dist_m": 2.429
1730
+ }
1731
+ ],
1732
+ [
1733
+ {
1734
+ "track": 0,
1735
+ "activity": 0.855,
1736
+ "class_idx": 5,
1737
+ "class_name": "water",
1738
+ "class_conf": 0.907,
1739
+ "azi_deg": 27.87,
1740
+ "ele_deg": -2.89,
1741
+ "dist_m": 2.413
1742
+ }
1743
+ ],
1744
+ [
1745
+ {
1746
+ "track": 0,
1747
+ "activity": 0.865,
1748
+ "class_idx": 5,
1749
+ "class_name": "water",
1750
+ "class_conf": 0.906,
1751
+ "azi_deg": 25.92,
1752
+ "ele_deg": -3.05,
1753
+ "dist_m": 2.407
1754
+ }
1755
+ ]
1756
+ ],
1757
+ "comparison": {
1758
+ "T_s": 303,
1759
+ "activity_gt_frac": 0.2433993399339934,
1760
+ "activity_rc_frac": 0.25,
1761
+ "activity_jaccard": 0.9735973597359736,
1762
+ "activity_precision_rc_vs_gt": 0.9735973597359736,
1763
+ "activity_recall_rc_vs_gt": 1.0,
1764
+ "activity_f1_rc_vs_gt": 0.9866220735785952,
1765
+ "class_match_rate": 0.8033898305084746,
1766
+ "doa_angular_error_deg_mean": 119.9124931557582,
1767
+ "doa_angular_error_deg_median": 122.31229223115486,
1768
+ "distance_mae_m": 0.025827614590525627,
1769
+ "top_class_agreement": 1,
1770
+ "gt_top_class": "water",
1771
+ "rc_top_class": "water"
1772
+ },
1773
+ "sample": "s1_rank21_insectset_000328"
1774
+ },
1775
+ "s1_rank21_opensource_emilia_zh_rawdata_English_002504": {
1776
+ "gt_top_class": "speech",
1777
+ "rc_top_class": "singing",
1778
+ "gt_frames_preview": [
1779
+ [],
1780
+ [],
1781
+ [],
1782
+ [
1783
+ {
1784
+ "track": 0,
1785
+ "activity": 0.754,
1786
+ "class_idx": 16,
1787
+ "class_name": "speech",
1788
+ "class_conf": 0.898,
1789
+ "azi_deg": -177.3,
1790
+ "ele_deg": 2.6,
1791
+ "dist_m": 3.281
1792
+ }
1793
+ ],
1794
+ [
1795
+ {
1796
+ "track": 0,
1797
+ "activity": 0.846,
1798
+ "class_idx": 16,
1799
+ "class_name": "speech",
1800
+ "class_conf": 0.899,
1801
+ "azi_deg": -176.82,
1802
+ "ele_deg": 2.84,
1803
+ "dist_m": 3.314
1804
+ }
1805
+ ]
1806
+ ],
1807
+ "rc_frames_preview": [
1808
+ [],
1809
+ [],
1810
+ [
1811
+ {
1812
+ "track": 0,
1813
+ "activity": 0.589,
1814
+ "class_idx": 41,
1815
+ "class_name": "singing",
1816
+ "class_conf": 0.577,
1817
+ "azi_deg": 5.85,
1818
+ "ele_deg": 11.35,
1819
+ "dist_m": 3.307
1820
+ }
1821
+ ],
1822
+ [
1823
+ {
1824
+ "track": 0,
1825
+ "activity": 0.811,
1826
+ "class_idx": 41,
1827
+ "class_name": "singing",
1828
+ "class_conf": 0.574,
1829
+ "azi_deg": 5.28,
1830
+ "ele_deg": 10.56,
1831
+ "dist_m": 3.34
1832
+ }
1833
+ ],
1834
+ [
1835
+ {
1836
+ "track": 0,
1837
+ "activity": 0.879,
1838
+ "class_idx": 41,
1839
+ "class_name": "singing",
1840
+ "class_conf": 0.586,
1841
+ "azi_deg": 4.92,
1842
+ "ele_deg": 10.47,
1843
+ "dist_m": 3.32
1844
+ }
1845
+ ]
1846
+ ],
1847
+ "comparison": {
1848
+ "T_s": 166,
1849
+ "activity_gt_frac": 0.23795180722891565,
1850
+ "activity_rc_frac": 0.23644578313253012,
1851
+ "activity_jaccard": 0.9811320754716981,
1852
+ "activity_precision_rc_vs_gt": 0.9936305732484076,
1853
+ "activity_recall_rc_vs_gt": 0.9873417721518988,
1854
+ "activity_f1_rc_vs_gt": 0.9904761904761905,
1855
+ "class_match_rate": 0.0,
1856
+ "doa_angular_error_deg_mean": 165.7686791980546,
1857
+ "doa_angular_error_deg_median": 166.00337916434938,
1858
+ "distance_mae_m": 0.04731345176696777,
1859
+ "top_class_agreement": 0,
1860
+ "gt_top_class": "speech",
1861
+ "rc_top_class": "singing"
1862
+ },
1863
+ "sample": "s1_rank21_opensource_emilia_zh_rawdata_English_002504"
1864
+ },
1865
+ "s1_rank22_audiosl_000699": {
1866
+ "gt_top_class": "water",
1867
+ "rc_top_class": "insect",
1868
+ "gt_frames_preview": [
1869
+ [
1870
+ {
1871
+ "track": 0,
1872
+ "activity": 0.552,
1873
+ "class_idx": 5,
1874
+ "class_name": "water",
1875
+ "class_conf": 0.635,
1876
+ "azi_deg": -136.67,
1877
+ "ele_deg": 6.51,
1878
+ "dist_m": 1.162
1879
+ }
1880
+ ],
1881
+ [
1882
+ {
1883
+ "track": 0,
1884
+ "activity": 0.667,
1885
+ "class_idx": 5,
1886
+ "class_name": "water",
1887
+ "class_conf": 0.481,
1888
+ "azi_deg": -137.17,
1889
+ "ele_deg": 6.84,
1890
+ "dist_m": 1.152
1891
+ }
1892
+ ],
1893
+ [
1894
+ {
1895
+ "track": 0,
1896
+ "activity": 0.676,
1897
+ "class_idx": 5,
1898
+ "class_name": "water",
1899
+ "class_conf": 0.503,
1900
+ "azi_deg": -137.0,
1901
+ "ele_deg": 7.0,
1902
+ "dist_m": 1.152
1903
+ }
1904
+ ],
1905
+ [
1906
+ {
1907
+ "track": 0,
1908
+ "activity": 0.686,
1909
+ "class_idx": 5,
1910
+ "class_name": "water",
1911
+ "class_conf": 0.485,
1912
+ "azi_deg": -136.81,
1913
+ "ele_deg": 7.29,
1914
+ "dist_m": 1.165
1915
+ }
1916
+ ],
1917
+ [
1918
+ {
1919
+ "track": 0,
1920
+ "activity": 0.701,
1921
+ "class_idx": 5,
1922
+ "class_name": "water",
1923
+ "class_conf": 0.477,
1924
+ "azi_deg": -136.63,
1925
+ "ele_deg": 7.32,
1926
+ "dist_m": 1.168
1927
+ }
1928
+ ]
1929
+ ],
1930
+ "rc_frames_preview": [
1931
+ [
1932
+ {
1933
+ "track": 0,
1934
+ "activity": 0.51,
1935
+ "class_idx": 35,
1936
+ "class_name": "insect",
1937
+ "class_conf": 0.142,
1938
+ "azi_deg": -105.79,
1939
+ "ele_deg": 11.21,
1940
+ "dist_m": 2.523
1941
+ }
1942
+ ],
1943
+ [],
1944
+ [],
1945
+ [],
1946
+ []
1947
+ ],
1948
+ "comparison": {
1949
+ "T_s": 18,
1950
+ "activity_gt_frac": 0.25,
1951
+ "activity_rc_frac": 0.013888888888888888,
1952
+ "activity_jaccard": 0.05555555555555555,
1953
+ "activity_precision_rc_vs_gt": 1.0,
1954
+ "activity_recall_rc_vs_gt": 0.05555555555555555,
1955
+ "activity_f1_rc_vs_gt": 0.10526315789473684,
1956
+ "class_match_rate": 0.0,
1957
+ "doa_angular_error_deg_mean": 30.84871783215931,
1958
+ "doa_angular_error_deg_median": 30.84871783215931,
1959
+ "distance_mae_m": 1.3608733415603638,
1960
+ "top_class_agreement": 0,
1961
+ "gt_top_class": "water",
1962
+ "rc_top_class": "insect"
1963
+ },
1964
+ "sample": "s1_rank22_audiosl_000699"
1965
+ },
1966
+ "s1_rank23_vocalsound_000569": {
1967
+ "gt_top_class": "breathing",
1968
+ "rc_top_class": "breathing",
1969
+ "gt_frames_preview": [
1970
+ [],
1971
+ [],
1972
+ [],
1973
+ [],
1974
+ []
1975
+ ],
1976
+ "rc_frames_preview": [
1977
+ [],
1978
+ [],
1979
+ [],
1980
+ [],
1981
+ []
1982
+ ],
1983
+ "comparison": {
1984
+ "T_s": 43,
1985
+ "activity_gt_frac": 0.18023255813953487,
1986
+ "activity_rc_frac": 0.09883720930232558,
1987
+ "activity_jaccard": 0.5483870967741935,
1988
+ "activity_precision_rc_vs_gt": 1.0,
1989
+ "activity_recall_rc_vs_gt": 0.5483870967741935,
1990
+ "activity_f1_rc_vs_gt": 0.7083333333333333,
1991
+ "class_match_rate": 1.0,
1992
+ "doa_angular_error_deg_mean": 15.514509384296726,
1993
+ "doa_angular_error_deg_median": 15.521100533230307,
1994
+ "distance_mae_m": 0.40135037899017334,
1995
+ "top_class_agreement": 1,
1996
+ "gt_top_class": "breathing",
1997
+ "rc_top_class": "breathing"
1998
+ },
1999
+ "sample": "s1_rank23_vocalsound_000569"
2000
+ },
2001
+ "s1_rank4_opensource_emilia_zh_rawdata_Chinese_003160": {
2002
+ "gt_top_class": "speech",
2003
+ "rc_top_class": "speech",
2004
+ "gt_frames_preview": [
2005
+ [],
2006
+ [],
2007
+ [],
2008
+ [],
2009
+ []
2010
+ ],
2011
+ "rc_frames_preview": [
2012
+ [],
2013
+ [],
2014
+ [],
2015
+ [],
2016
+ []
2017
+ ],
2018
+ "comparison": {
2019
+ "T_s": 134,
2020
+ "activity_gt_frac": 0.055970149253731345,
2021
+ "activity_rc_frac": 0.0,
2022
+ "activity_jaccard": 0.0,
2023
+ "activity_precision_rc_vs_gt": 0.0,
2024
+ "activity_recall_rc_vs_gt": 0.0,
2025
+ "activity_f1_rc_vs_gt": 0.0,
2026
+ "class_match_rate": NaN,
2027
+ "doa_angular_error_deg_mean": NaN,
2028
+ "doa_angular_error_deg_median": NaN,
2029
+ "distance_mae_m": NaN,
2030
+ "top_class_agreement": 1,
2031
+ "gt_top_class": "speech",
2032
+ "rc_top_class": "speech"
2033
+ },
2034
+ "sample": "s1_rank4_opensource_emilia_zh_rawdata_Chinese_003160"
2035
+ },
2036
+ "s1_rank5_opensource_emilia_zh_rawdata_English_001914": {
2037
+ "gt_top_class": "speech",
2038
+ "rc_top_class": "speech",
2039
+ "gt_frames_preview": [
2040
+ [
2041
+ {
2042
+ "track": 0,
2043
+ "activity": 0.556,
2044
+ "class_idx": 16,
2045
+ "class_name": "speech",
2046
+ "class_conf": 0.867,
2047
+ "azi_deg": 14.18,
2048
+ "ele_deg": 9.65,
2049
+ "dist_m": 1.336
2050
+ }
2051
+ ],
2052
+ [
2053
+ {
2054
+ "track": 0,
2055
+ "activity": 0.748,
2056
+ "class_idx": 16,
2057
+ "class_name": "speech",
2058
+ "class_conf": 0.879,
2059
+ "azi_deg": 14.36,
2060
+ "ele_deg": 9.37,
2061
+ "dist_m": 1.305
2062
+ }
2063
+ ],
2064
+ [
2065
+ {
2066
+ "track": 0,
2067
+ "activity": 0.792,
2068
+ "class_idx": 16,
2069
+ "class_name": "speech",
2070
+ "class_conf": 0.881,
2071
+ "azi_deg": 15.14,
2072
+ "ele_deg": 9.25,
2073
+ "dist_m": 1.301
2074
+ }
2075
+ ],
2076
+ [
2077
+ {
2078
+ "track": 0,
2079
+ "activity": 0.816,
2080
+ "class_idx": 16,
2081
+ "class_name": "speech",
2082
+ "class_conf": 0.882,
2083
+ "azi_deg": 15.62,
2084
+ "ele_deg": 9.07,
2085
+ "dist_m": 1.298
2086
+ }
2087
+ ],
2088
+ [
2089
+ {
2090
+ "track": 0,
2091
+ "activity": 0.831,
2092
+ "class_idx": 16,
2093
+ "class_name": "speech",
2094
+ "class_conf": 0.883,
2095
+ "azi_deg": 15.93,
2096
+ "ele_deg": 8.96,
2097
+ "dist_m": 1.303
2098
+ }
2099
+ ]
2100
+ ],
2101
+ "rc_frames_preview": [
2102
+ [],
2103
+ [
2104
+ {
2105
+ "track": 0,
2106
+ "activity": 0.699,
2107
+ "class_idx": 16,
2108
+ "class_name": "speech",
2109
+ "class_conf": 0.867,
2110
+ "azi_deg": 71.33,
2111
+ "ele_deg": 77.79,
2112
+ "dist_m": 1.375
2113
+ },
2114
+ {
2115
+ "track": 2,
2116
+ "activity": 0.554,
2117
+ "class_idx": 16,
2118
+ "class_name": "speech",
2119
+ "class_conf": 0.897,
2120
+ "azi_deg": 26.81,
2121
+ "ele_deg": 13.54,
2122
+ "dist_m": 4.611
2123
+ }
2124
+ ],
2125
+ [
2126
+ {
2127
+ "track": 0,
2128
+ "activity": 0.797,
2129
+ "class_idx": 16,
2130
+ "class_name": "speech",
2131
+ "class_conf": 0.871,
2132
+ "azi_deg": 115.38,
2133
+ "ele_deg": 75.36,
2134
+ "dist_m": 1.352
2135
+ },
2136
+ {
2137
+ "track": 2,
2138
+ "activity": 0.615,
2139
+ "class_idx": 16,
2140
+ "class_name": "speech",
2141
+ "class_conf": 0.897,
2142
+ "azi_deg": 25.01,
2143
+ "ele_deg": 11.4,
2144
+ "dist_m": 4.588
2145
+ }
2146
+ ],
2147
+ [
2148
+ {
2149
+ "track": 0,
2150
+ "activity": 0.831,
2151
+ "class_idx": 16,
2152
+ "class_name": "speech",
2153
+ "class_conf": 0.875,
2154
+ "azi_deg": 125.98,
2155
+ "ele_deg": 70.05,
2156
+ "dist_m": 1.34
2157
+ },
2158
+ {
2159
+ "track": 2,
2160
+ "activity": 0.632,
2161
+ "class_idx": 16,
2162
+ "class_name": "speech",
2163
+ "class_conf": 0.898,
2164
+ "azi_deg": 27.78,
2165
+ "ele_deg": 14.0,
2166
+ "dist_m": 4.539
2167
+ }
2168
+ ],
2169
+ [
2170
+ {
2171
+ "track": 0,
2172
+ "activity": 0.855,
2173
+ "class_idx": 16,
2174
+ "class_name": "speech",
2175
+ "class_conf": 0.879,
2176
+ "azi_deg": 130.07,
2177
+ "ele_deg": 67.12,
2178
+ "dist_m": 1.339
2179
+ },
2180
+ {
2181
+ "track": 2,
2182
+ "activity": 0.64,
2183
+ "class_idx": 16,
2184
+ "class_name": "speech",
2185
+ "class_conf": 0.899,
2186
+ "azi_deg": 27.53,
2187
+ "ele_deg": 13.78,
2188
+ "dist_m": 4.528
2189
+ }
2190
+ ]
2191
+ ],
2192
+ "comparison": {
2193
+ "T_s": 46,
2194
+ "activity_gt_frac": 0.25,
2195
+ "activity_rc_frac": 0.4673913043478261,
2196
+ "activity_jaccard": 0.5172413793103449,
2197
+ "activity_precision_rc_vs_gt": 0.5232558139534884,
2198
+ "activity_recall_rc_vs_gt": 0.9782608695652174,
2199
+ "activity_f1_rc_vs_gt": 0.6818181818181819,
2200
+ "class_match_rate": 1.0,
2201
+ "doa_angular_error_deg_mean": 101.46253253626486,
2202
+ "doa_angular_error_deg_median": 103.82595102357277,
2203
+ "distance_mae_m": 0.01654328778386116,
2204
+ "top_class_agreement": 1,
2205
+ "gt_top_class": "speech",
2206
+ "rc_top_class": "speech"
2207
+ },
2208
+ "sample": "s1_rank5_opensource_emilia_zh_rawdata_English_001914"
2209
+ },
2210
+ "s1_rank9_opensource_emilia_zh_rawdata_English_001143": {
2211
+ "gt_top_class": "speech",
2212
+ "rc_top_class": "speech",
2213
+ "gt_frames_preview": [
2214
+ [],
2215
+ [],
2216
+ [
2217
+ {
2218
+ "track": 0,
2219
+ "activity": 0.638,
2220
+ "class_idx": 16,
2221
+ "class_name": "speech",
2222
+ "class_conf": 0.889,
2223
+ "azi_deg": 28.91,
2224
+ "ele_deg": -18.71,
2225
+ "dist_m": 1.646
2226
+ }
2227
+ ],
2228
+ [
2229
+ {
2230
+ "track": 0,
2231
+ "activity": 0.754,
2232
+ "class_idx": 16,
2233
+ "class_name": "speech",
2234
+ "class_conf": 0.894,
2235
+ "azi_deg": 29.12,
2236
+ "ele_deg": -18.26,
2237
+ "dist_m": 1.635
2238
+ }
2239
+ ],
2240
+ [
2241
+ {
2242
+ "track": 0,
2243
+ "activity": 0.8,
2244
+ "class_idx": 16,
2245
+ "class_name": "speech",
2246
+ "class_conf": 0.895,
2247
+ "azi_deg": 29.48,
2248
+ "ele_deg": -18.06,
2249
+ "dist_m": 1.616
2250
+ }
2251
+ ]
2252
+ ],
2253
+ "rc_frames_preview": [
2254
+ [],
2255
+ [
2256
+ {
2257
+ "track": 3,
2258
+ "activity": 0.546,
2259
+ "class_idx": 16,
2260
+ "class_name": "speech",
2261
+ "class_conf": 0.698,
2262
+ "azi_deg": 19.6,
2263
+ "ele_deg": 8.15,
2264
+ "dist_m": 3.044
2265
+ }
2266
+ ],
2267
+ [
2268
+ {
2269
+ "track": 3,
2270
+ "activity": 0.665,
2271
+ "class_idx": 16,
2272
+ "class_name": "speech",
2273
+ "class_conf": 0.733,
2274
+ "azi_deg": 21.22,
2275
+ "ele_deg": 7.23,
2276
+ "dist_m": 3.096
2277
+ }
2278
+ ],
2279
+ [
2280
+ {
2281
+ "track": 0,
2282
+ "activity": 0.612,
2283
+ "class_idx": 16,
2284
+ "class_name": "speech",
2285
+ "class_conf": 0.89,
2286
+ "azi_deg": 39.92,
2287
+ "ele_deg": 44.31,
2288
+ "dist_m": 1.815
2289
+ },
2290
+ {
2291
+ "track": 3,
2292
+ "activity": 0.703,
2293
+ "class_idx": 16,
2294
+ "class_name": "speech",
2295
+ "class_conf": 0.767,
2296
+ "azi_deg": 23.01,
2297
+ "ele_deg": 5.55,
2298
+ "dist_m": 3.156
2299
+ }
2300
+ ],
2301
+ [
2302
+ {
2303
+ "track": 0,
2304
+ "activity": 0.667,
2305
+ "class_idx": 16,
2306
+ "class_name": "speech",
2307
+ "class_conf": 0.893,
2308
+ "azi_deg": 46.21,
2309
+ "ele_deg": 46.59,
2310
+ "dist_m": 1.756
2311
+ },
2312
+ {
2313
+ "track": 3,
2314
+ "activity": 0.713,
2315
+ "class_idx": 16,
2316
+ "class_name": "speech",
2317
+ "class_conf": 0.79,
2318
+ "azi_deg": 24.48,
2319
+ "ele_deg": 5.5,
2320
+ "dist_m": 3.188
2321
+ }
2322
+ ]
2323
+ ],
2324
+ "comparison": {
2325
+ "T_s": 121,
2326
+ "activity_gt_frac": 0.2190082644628099,
2327
+ "activity_rc_frac": 0.3119834710743802,
2328
+ "activity_jaccard": 0.18433179723502305,
2329
+ "activity_precision_rc_vs_gt": 0.26490066225165565,
2330
+ "activity_recall_rc_vs_gt": 0.37735849056603776,
2331
+ "activity_f1_rc_vs_gt": 0.31128404669260706,
2332
+ "class_match_rate": 1.0,
2333
+ "doa_angular_error_deg_mean": 73.54726624326348,
2334
+ "doa_angular_error_deg_median": 74.05986765172213,
2335
+ "distance_mae_m": 0.0948357805609703,
2336
+ "top_class_agreement": 1,
2337
+ "gt_top_class": "speech",
2338
+ "rc_top_class": "speech"
2339
+ },
2340
+ "sample": "s1_rank9_opensource_emilia_zh_rawdata_English_001143"
2341
+ }
2342
+ }
eval_voxaudio_ood_results/stable_audio_vae/summary.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "mean_activity_jaccard": 0.5231948644785973,
3
+ "n_valid_activity_jaccard": 20,
4
+ "mean_activity_precision_rc_vs_gt": 0.6623050254985954,
5
+ "n_valid_activity_precision_rc_vs_gt": 20,
6
+ "mean_activity_recall_rc_vs_gt": 0.6414211563969253,
7
+ "n_valid_activity_recall_rc_vs_gt": 20,
8
+ "mean_activity_f1_rc_vs_gt": 0.6098793872424151,
9
+ "n_valid_activity_f1_rc_vs_gt": 20,
10
+ "mean_class_match_rate": 0.691916088469286,
11
+ "n_valid_class_match_rate": 17,
12
+ "mean_doa_angular_error_deg_mean": 62.811367085465875,
13
+ "n_valid_doa_angular_error_deg_mean": 17,
14
+ "mean_doa_angular_error_deg_median": 63.464756325810015,
15
+ "n_valid_doa_angular_error_deg_median": 17,
16
+ "mean_distance_mae_m": 0.2713859439553583,
17
+ "n_valid_distance_mae_m": 17,
18
+ "mean_top_class_agreement": 0.65,
19
+ "n_valid_top_class_agreement": 20,
20
+ "mean_activity_gt_frac": 0.20619253654594938,
21
+ "n_valid_activity_gt_frac": 20,
22
+ "mean_activity_rc_frac": 0.23435994834729795,
23
+ "n_valid_activity_rc_frac": 20,
24
+ "n_samples": 20
25
+ }
eval_voxaudio_ood_results/summary_all.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "dacvae": {
3
+ "mean_activity_jaccard": 0.5918084170858557,
4
+ "n_valid_activity_jaccard": 20,
5
+ "mean_activity_precision_rc_vs_gt": 0.6852889043752468,
6
+ "n_valid_activity_precision_rc_vs_gt": 20,
7
+ "mean_activity_recall_rc_vs_gt": 0.6344722344150824,
8
+ "n_valid_activity_recall_rc_vs_gt": 20,
9
+ "mean_activity_f1_rc_vs_gt": 0.6424736331628711,
10
+ "n_valid_activity_f1_rc_vs_gt": 20,
11
+ "mean_class_match_rate": 0.6962217958284476,
12
+ "n_valid_class_match_rate": 15,
13
+ "mean_doa_angular_error_deg_mean": 52.59217336237226,
14
+ "n_valid_doa_angular_error_deg_mean": 15,
15
+ "mean_doa_angular_error_deg_median": 52.82468845575132,
16
+ "n_valid_doa_angular_error_deg_median": 15,
17
+ "mean_distance_mae_m": 0.3990471541881561,
18
+ "n_valid_distance_mae_m": 15,
19
+ "mean_top_class_agreement": 0.8,
20
+ "n_valid_top_class_agreement": 20,
21
+ "mean_activity_gt_frac": 0.20619253654594938,
22
+ "n_valid_activity_gt_frac": 20,
23
+ "mean_activity_rc_frac": 0.16576634449721478,
24
+ "n_valid_activity_rc_frac": 20,
25
+ "n_samples": 20
26
+ },
27
+ "flow2gan": {
28
+ "mean_activity_jaccard": 0.5769315608191647,
29
+ "n_valid_activity_jaccard": 20,
30
+ "mean_activity_precision_rc_vs_gt": 0.7164553380856092,
31
+ "n_valid_activity_precision_rc_vs_gt": 20,
32
+ "mean_activity_recall_rc_vs_gt": 0.6691510536846746,
33
+ "n_valid_activity_recall_rc_vs_gt": 20,
34
+ "mean_activity_f1_rc_vs_gt": 0.6429039181380094,
35
+ "n_valid_activity_f1_rc_vs_gt": 20,
36
+ "mean_class_match_rate": 0.9336715651102376,
37
+ "n_valid_class_match_rate": 17,
38
+ "mean_doa_angular_error_deg_mean": 76.88068964135293,
39
+ "n_valid_doa_angular_error_deg_mean": 17,
40
+ "mean_doa_angular_error_deg_median": 77.08370928011593,
41
+ "n_valid_doa_angular_error_deg_median": 17,
42
+ "mean_distance_mae_m": 0.38080851963775997,
43
+ "n_valid_distance_mae_m": 17,
44
+ "mean_top_class_agreement": 0.9,
45
+ "n_valid_top_class_agreement": 20,
46
+ "mean_activity_gt_frac": 0.20619253654594938,
47
+ "n_valid_activity_gt_frac": 20,
48
+ "mean_activity_rc_frac": 0.19492021351895925,
49
+ "n_valid_activity_rc_frac": 20,
50
+ "n_samples": 20
51
+ },
52
+ "mono_vae": {
53
+ "mean_activity_jaccard": 0.31722869095023537,
54
+ "n_valid_activity_jaccard": 20,
55
+ "mean_activity_precision_rc_vs_gt": 0.4940412106193858,
56
+ "n_valid_activity_precision_rc_vs_gt": 20,
57
+ "mean_activity_recall_rc_vs_gt": 0.3829143289520885,
58
+ "n_valid_activity_recall_rc_vs_gt": 20,
59
+ "mean_activity_f1_rc_vs_gt": 0.39739221883924747,
60
+ "n_valid_activity_f1_rc_vs_gt": 20,
61
+ "mean_class_match_rate": 0.6986156079044322,
62
+ "n_valid_class_match_rate": 13,
63
+ "mean_doa_angular_error_deg_mean": 79.70906496987138,
64
+ "n_valid_doa_angular_error_deg_mean": 13,
65
+ "mean_doa_angular_error_deg_median": 79.14171551956005,
66
+ "n_valid_doa_angular_error_deg_median": 13,
67
+ "mean_distance_mae_m": 0.6082516071888117,
68
+ "n_valid_distance_mae_m": 13,
69
+ "mean_top_class_agreement": 0.65,
70
+ "n_valid_top_class_agreement": 20,
71
+ "mean_activity_gt_frac": 0.2116135538807752,
72
+ "n_valid_activity_gt_frac": 20,
73
+ "mean_activity_rc_frac": 0.1629927504276915,
74
+ "n_valid_activity_rc_frac": 20,
75
+ "n_samples": 20
76
+ },
77
+ "stable_audio_vae": {
78
+ "mean_activity_jaccard": 0.5231948644785973,
79
+ "n_valid_activity_jaccard": 20,
80
+ "mean_activity_precision_rc_vs_gt": 0.6623050254985954,
81
+ "n_valid_activity_precision_rc_vs_gt": 20,
82
+ "mean_activity_recall_rc_vs_gt": 0.6414211563969253,
83
+ "n_valid_activity_recall_rc_vs_gt": 20,
84
+ "mean_activity_f1_rc_vs_gt": 0.6098793872424151,
85
+ "n_valid_activity_f1_rc_vs_gt": 20,
86
+ "mean_class_match_rate": 0.691916088469286,
87
+ "n_valid_class_match_rate": 17,
88
+ "mean_doa_angular_error_deg_mean": 62.811367085465875,
89
+ "n_valid_doa_angular_error_deg_mean": 17,
90
+ "mean_doa_angular_error_deg_median": 63.464756325810015,
91
+ "n_valid_doa_angular_error_deg_median": 17,
92
+ "mean_distance_mae_m": 0.2713859439553583,
93
+ "n_valid_distance_mae_m": 17,
94
+ "mean_top_class_agreement": 0.65,
95
+ "n_valid_top_class_agreement": 20,
96
+ "mean_activity_gt_frac": 0.20619253654594938,
97
+ "n_valid_activity_gt_frac": 20,
98
+ "mean_activity_rc_frac": 0.23435994834729795,
99
+ "n_valid_activity_rc_frac": 20,
100
+ "n_samples": 20
101
+ },
102
+ "foa_vae": {
103
+ "mean_activity_jaccard": 0.2869139591647025,
104
+ "n_valid_activity_jaccard": 70,
105
+ "mean_activity_precision_rc_vs_gt": 0.3842437781867924,
106
+ "n_valid_activity_precision_rc_vs_gt": 70,
107
+ "mean_activity_recall_rc_vs_gt": 0.3355311172735717,
108
+ "n_valid_activity_recall_rc_vs_gt": 70,
109
+ "mean_activity_f1_rc_vs_gt": 0.33179939006604536,
110
+ "n_valid_activity_f1_rc_vs_gt": 70,
111
+ "mean_class_match_rate": 0.7050047775153738,
112
+ "n_valid_class_match_rate": 39,
113
+ "mean_doa_angular_error_deg_mean": 47.81001474877972,
114
+ "n_valid_doa_angular_error_deg_mean": 39,
115
+ "mean_doa_angular_error_deg_median": 48.616234655597715,
116
+ "n_valid_doa_angular_error_deg_median": 39,
117
+ "mean_distance_mae_m": 0.4801142644137144,
118
+ "n_valid_distance_mae_m": 39,
119
+ "mean_top_class_agreement": 0.6857142857142857,
120
+ "n_valid_top_class_agreement": 70,
121
+ "mean_activity_gt_frac": 0.21816713600161056,
122
+ "n_valid_activity_gt_frac": 70,
123
+ "mean_activity_rc_frac": 0.15041837313585474,
124
+ "n_valid_activity_rc_frac": 70,
125
+ "n_samples": 70
126
+ }
127
+ }