-
Notifications
You must be signed in to change notification settings - Fork 146
Description
cmake -DCMAKE_BUILD_TYPE=Release -D DISABLE_R=y -D CUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-10.2 . && make
-- Configuring done
-- Generating done
-- Build files have been written to: /root/kmcuda-master/src
[100%] Built target KMCUDA
root@waston-Default-string:/kmcuda-master/src# /usr/bin/python3.8 /root/kmcuda-master/src/test.py/kmcuda-master/src# python3 test.py
arguments: 1 0x7ffee79f4edc 0.010 0.10 1 1000 256 10 3 0 0 3 0x7c73f9599010 0x1f023a0 0x1f04bb0 (nil)
reassignments threshold: 10
yinyang groups: 1
[0] *dest: 0x7c7377e00000 - 0x7c7377efa000 (1024000)
[0] device_centroids: 0x7c7377efa000 - 0x7c7377efc800 (10240)
[0] device_assignments: 0x7c7377efc800 - 0x7c7377efd7a0 (4000)
[0] device_assignments_prev: 0x7c7377efd800 - 0x7c7377efe7a0 (4000)
[0] device_ccounts: 0x7c7377efe800 - 0x7c7377efe828 (40)
[0] device_assignments_yy: 0x7c7377efea00 - 0x7c7377efea28 (40)
[0] device_bounds_yy: 0x7c7377efec00 - 0x7c7377f00b40 (8000)
[0] device_drifts_yy: 0x7c7377f00c00 - 0x7c7377f03428 (10280)
[0] device_passed_yy: 0x7c7377f03600 - 0x7c7377f045a0 (4000)
reusing passed_yy for centroids_yy
GPU #0 memory: used 1485701120 bytes (8.8%), free 15368454144 bytes, total 16854155264 bytes
GPU #0 has 49152 bytes of shared memory per block
transposing the samples...
transpose <<<(32, 8), (8, 32)>>> 1000, 256, xyswap
performing kmeans++...
kmeans++: dump 1000 256 0x2c37040
kmeans++: dev #0: 0x7c7377e00000 0x7c7377efa000 0x7c7377efc800
step 1[0] dev_dists: 0x7c7377f04600 - 0x7c7377f04640 (64)
step 2[0] dev_dists: 0x7c7377f04600 - 0x7c7377f04640 (64)
step 3[0] dev_dists: 0x7c7377f04600 - 0x7c7377f04640 (64)
step 4[0] dev_dists: 0x7c7377f04600 - 0x7c7377f04640 (64)
step 5[0] dev_dists: 0x7c7377f04600 - 0x7c7377f04640 (64)
step 6[0] dev_dists: 0x7c7377f04600 - 0x7c7377f04640 (64)
step 7[0] dev_dists: 0x7c7377f04600 - 0x7c7377f04640 (64)
step 8[0] dev_dists: 0x7c7377f04600 - 0x7c7377f04640 (64)
step 9
F^Z
[5]+ Stopped /usr/bin/python3.8 /root/kmcuda-master/src/test.py
root@waston-Default-string:
arguments: 1 0x7ffccc2d07dc 0.010 0.10 1 1000 256 10 3 0 0 3 0x7703000db010 0x2dfbfe0 0x2dfe7f0 (nil)
reassignments threshold: 10
yinyang groups: 1
[0] *dest: 0x770279e00000 - 0x770279efa000 (1024000)
[0] device_centroids: 0x770279efa000 - 0x770279efc800 (10240)
[0] device_assignments: 0x770279efc800 - 0x770279efd7a0 (4000)
[0] device_assignments_prev: 0x770279efd800 - 0x770279efe7a0 (4000)
[0] device_ccounts: 0x770279efe800 - 0x770279efe828 (40)
[0] device_assignments_yy: 0x770279efea00 - 0x770279efea28 (40)
[0] device_bounds_yy: 0x770279efec00 - 0x770279f00b40 (8000)
[0] device_drifts_yy: 0x770279f00c00 - 0x770279f03428 (10280)
[0] device_passed_yy: 0x770279f03600 - 0x770279f045a0 (4000)
reusing passed_yy for centroids_yy
GPU #0 memory: used 1614020608 bytes (9.6%), free 15240134656 bytes, total 16854155264 bytes
GPU #0 has 49152 bytes of shared memory per block
transposing the samples...
transpose <<<(32, 8), (8, 32)>>> 1000, 256, xyswap
performing kmeans++...
kmeans++: dump 1000 256 0x3c87010
kmeans++: dev #0: 0x770279e00000 0x770279efa000 0x770279efc800
step 1[0] dev_dists: 0x770279f04600 - 0x770279f04640 (64)
step 2[0] dev_dists: 0x770279f04600 - 0x770279f04640 (64)
step 3[0] dev_dists: 0x770279f04600 - 0x770279f04640 (64)
step 4[0] dev_dists: 0x770279f04600 - 0x770279f04640 (64)
step 5[0] dev_dists: 0x770279f04600 - 0x770279f04640 (64)
step 6[0] dev_dists: 0x770279f04600 - 0x770279f04640 (64)
step 7[0] dev_dists: 0x770279f04600 - 0x770279f04640 (64)
step 8[0] dev_dists: 0x770279f04600 - 0x770279f04640 (64)
step 9
F