The Program stucks on

[qiaoyu1002]

Dear authors，

Could please check the log attached, where the problem is？ My program is stucked on.

arguments: 1 0x7ffede2239ec 0.010 0.10 0 58 128 3 1234 1 1 3 0x555972b71bb0 0x5559787dcfe0 0x555953d49250 (nil)
reassignments threshold: 0
yinyang groups: 0
[0] *dest: 0x7f751b200000 - 0x7f751b207400 (29696)
[0] device_centroids: 0x7f751b207400 - 0x7f751b207a00 (1536)
[0] device_assignments: 0x7f751b207a00 - 0x7f751b207ae8 (232)
[0] device_assignments_prev: 0x7f751b207c00 - 0x7f751b207ce8 (232)
[0] device_ccounts: 0x7f751b207e00 - 0x7f751b207e0c (12)
GPU #0 memory: used 1662779392 bytes (3.3%), free 49387536384 bytes, total 51050315776 bytes
GPU #0 has 49152 bytes of shared memory per block
transposing the samples...
transpose <<<(4, 2), (32, 8)>>> 58, 128
performing kmeans++...
kmeans++: dump 58 128 0x555973d8eab0
kmeans++: dev #0: 0x7f751b200000 0x7f751b207400 0x7f751b207a00
step 1[0] dev_dists: 0x7f751b208000 - 0x7f751b208040 (64)
step 2[0] dev_dists: 0x7f751b208000 - 0x7f751b208040 (64)
done
too few clusters for this yinyang_t => Lloyd
plans: [(0, 58)]
planc: [(0, 3)]
iteration 1: 58 reassignments
iteration 2: 5 reassignments
iteration 3: 3 reassignments
iteration 4: 3 reassignments
iteration 5: 3 reassignments
iteration 6: 1 reassignments
iteration 7: 4 reassignments
iteration 8: 8 reassignments
iteration 9: 3 reassignments
iteration 10: 4 reassignments
iteration 11: 2 reassignments
iteration 12: 1 reassignments
iteration 13: 1 reassignments
iteration 14: 0 reassignments
return kmcudaSuccess
arguments: 1 0x7ffede2239ec 0.010 0.10 0 87 128 3 1234 1 1 3 0x555977d4d8c0 0x5559778afe80 0x555953dbc9d0 (nil)
reassignments threshold: 0
yinyang groups: 0
[0] *dest: 0x7f751b200000 - 0x7f751b20ae00 (44544)
[0] device_centroids: 0x7f751b20ae00 - 0x7f751b20b400 (1536)
[0] device_assignments: 0x7f751b20b400 - 0x7f751b20b55c (348)
[0] device_assignments_prev: 0x7f751b20b600 - 0x7f751b20b75c (348)
[0] device_ccounts: 0x7f751b20b800 - 0x7f751b20b80c (12)
GPU #0 memory: used 1660682240 bytes (3.3%), free 49389633536 bytes, total 51050315776 bytes
GPU #0 has 49152 bytes of shared memory per block
transposing the samples...
transpose <<<(4, 3), (32, 8)>>> 87, 128
performing kmeans++...
kmeans++: dump 87 128 0x5559540f44c0
kmeans++: dev #0: 0x7f751b200000 0x7f751b20ae00 0x7f751b20b400
step 1[0] dev_dists: 0x7f751b20ba00 - 0x7f751b20ba40 (64)
step 2[0] dev_dists: 0x7f751b20ba00 - 0x7f751b20ba40 (64)
done
too few clusters for this yinyang_t => Lloyd
plans: [(0, 87)]
planc: [(0, 3)]
iteration 1: 87 reassignments
iteration 2: 9 reassignments
iteration 3: 5 reassignments
iteration 4: 7 reassignments
iteration 5: 3 reassignments
iteration 6: 3 reassignments
iteration 7: 1 reassignments
iteration 8: 3 reassignments
iteration 9: 1 reassignments
iteration 10: 4 reassignments
iteration 11: 3 reassignments
iteration 12: 2 reassignments
iteration 13: 0 reassignments
return kmcudaSuccess
arguments: 1 0x7ffede2239ec 0.010 0.10 0 25 128 3 1234 1 1 3 0x555978938ac0 0x555978947750 0x5559540c4d30 (nil)
reassignments threshold: 0
yinyang groups: 0
[0] *dest: 0x7f751b200000 - 0x7f751b203200 (12800)
[0] device_centroids: 0x7f751b203200 - 0x7f751b203800 (1536)
[0] device_assignments: 0x7f751b203800 - 0x7f751b203864 (100)
[0] device_assignments_prev: 0x7f751b203a00 - 0x7f751b203a64 (100)
[0] device_ccounts: 0x7f751b203c00 - 0x7f751b203c0c (12)
GPU #0 memory: used 1658585088 bytes (3.2%), free 49391730688 bytes, total 51050315776 bytes
GPU #0 has 49152 bytes of shared memory per block
transposing the samples...
transpose <<<(4, 1), (32, 8)>>> 25, 128
performing kmeans++...
kmeans++: dump 25 128 0x555973d97f80
kmeans++: dev #0: 0x7f751b200000 0x7f751b203200 0x7f751b203800
step 1[0] dev_dists: 0x7f751b203e00 - 0x7f751b203e40 (64)
step 2[0] dev_dists: 0x7f751b203e00 - 0x7f751b203e40 (64)
done
too few clusters for this yinyang_t => Lloyd
plans: [(0, 25)]
planc: [(0, 3)]
iteration 1: 25 reassignments
iteration 2: 2 reassignments
iteration 3: 3 reassignments
iteration 4: 1 reassignments
iteration 5: 1 reassignments
iteration 6: 2 reassignments
iteration 7: 1 reassignments
iteration 8: 0 reassignments
return kmcudaSuccess
arguments: 1 0x7ffede2239ec 0.010 0.10 0 4 128 3 1234 1 1 3 0x5559771795d0 0x555978956050 0x555895ca28c0 (nil)
reassignments threshold: 0
yinyang groups: 0
[0] *dest: 0x7f751b200000 - 0x7f751b200800 (2048)
[0] device_centroids: 0x7f751b200800 - 0x7f751b200e00 (1536)
[0] device_assignments: 0x7f751b200e00 - 0x7f751b200e10 (16)
[0] device_assignments_prev: 0x7f751b201000 - 0x7f751b201010 (16)
[0] device_ccounts: 0x7f751b201200 - 0x7f751b20120c (12)
GPU #0 memory: used 1656487936 bytes (3.2%), free 49393827840 bytes, total 51050315776 bytes
GPU #0 has 49152 bytes of shared memory per block
transposing the samples...
transpose <<<(4, 1), (32, 8)>>> 4, 128
performing kmeans++...
kmeans++: dump 4 128 0x555954856b10
kmeans++: dev #0: 0x7f751b200000 0x7f751b200800 0x7f751b200e00
step 1[0] dev_dists: 0x7f751b201400 - 0x7f751b201440 (64)
step 2[0] dev_dists: 0x7f751b201400 - 0x7f751b201440 (64)
done
too few clusters for this yinyang_t => Lloyd
plans: [(0, 4)]
planc: [(0, 3)]
iteration 1: 4 reassignments
iteration 2: 0 reassignments
return kmcudaSuccess
4%|████▌ | 2/50 [00:25<11:21, 14.20s/it]
| Global Training Round : 3 |

arguments: 1 0x7ffede2239ec 0.010 0.10 0 232 128 3 1234 1 1 3 0x5559736ccd30 0x555977a18c90 0x555954115a90 (nil)
reassignments threshold: 2
yinyang groups: 0
[0] *dest: 0x7f751b200000 - 0x7f751b21d000 (118784)
[0] device_centroids: 0x7f751b21d000 - 0x7f751b21d600 (1536)
[0] device_assignments: 0x7f751b21d600 - 0x7f751b21d9a0 (928)
[0] device_assignments_prev: 0x7f751b21da00 - 0x7f751b21dda0 (928)
[0] device_ccounts: 0x7f751b21de00 - 0x7f751b21de0c (12)
GPU #0 memory: used 1654390784 bytes (3.2%), free 49395924992 bytes, total 51050315776 bytes
GPU #0 has 49152 bytes of shared memory per block
transposing the samples...
transpose <<<(8, 4), (8, 32)>>> 232, 128, xyswap =================> Here stuck on

Thanks in advance for your analysis. If you need more detailed information about this, please let me know.