cephi/ampere_research
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Max core

Browse Source
This commit is contained in:
cephi 2024-12-12 22:33:14 -05:00
parent 46eadfabff
commit 644a6c9954
72 changed files with 1702 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
pytorch/output_max_core/altra_10_2_10_100000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 84217, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 999958, "MATRIX_DENSITY": 9.99958e-05, "TIME_S": 13.840779781341553, "TIME_S_1KI": 0.1643466257565759, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1274.119412755966, "W": 90.91756741033639, "J_1KI": 15.129004984218936, "W_1KI": 1.0795631215827728, "W_D": 80.75756741033639, "J_D": 1131.737103128433, "W_D_1KI": 0.9589223958385645, "J_D_1KI": 0.011386328126608222}

17
pytorch/output_max_core/altra_10_2_10_100000_0.0001.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 11, 18, ..., 999940, 999952,
999958]),
col_indices=tensor([ 4739, 28215, 31996, ..., 61735, 64755, 95212]),
values=tensor([ 1.1882, 1.3136, -2.0799, ..., 1.5641, 2.5173,
0.8848]), size=(100000, 100000), nnz=999958,
layout=torch.sparse_csr)
tensor([0.8457, 0.0677, 0.2670, ..., 0.4314, 0.6888, 0.0802])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 999958
Density: 9.99958e-05
Time: 13.840779781341553 seconds

1
pytorch/output_max_core/altra_10_2_10_100000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 113612, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 100000, "MATRIX_DENSITY": 1e-05, "TIME_S": 17.586217641830444, "TIME_S_1KI": 0.15479190263203224, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1025.6564905548096, "W": 79.79345590532563, "J_1KI": 9.027712658476302, "W_1KI": 0.7023329921603847, "W_D": 69.44845590532563, "J_D": 892.6829744386673, "W_D_1KI": 0.6112774698564027, "J_D_1KI": 0.0053803952914868395}

17
pytorch/output_max_core/altra_10_2_10_100000_1e-05.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 2, ..., 99998, 99998,
100000]),
col_indices=tensor([10815, 45605, 72128, ..., 22455, 22018, 68720]),
values=tensor([ 0.5455, 0.6676, -0.6078, ..., 0.0308, 0.3015,
-0.0823]), size=(100000, 100000), nnz=100000,
layout=torch.sparse_csr)
tensor([0.5616, 0.2401, 0.2358, ..., 0.8210, 0.1278, 0.2310])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 100000
Density: 1e-05
Time: 17.586217641830444 seconds

1
pytorch/output_max_core/altra_10_2_10_100000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 90108, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 499987, "MATRIX_DENSITY": 4.99987e-05, "TIME_S": 11.732282161712646, "TIME_S_1KI": 0.130202447748398, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 729.1587126636505, "W": 75.92243332338597, "J_1KI": 8.092053010428048, "W_1KI": 0.8425715066740574, "W_D": 65.62243332338598, "J_D": 630.2375583791733, "W_D_1KI": 0.7282642309604694, "J_D_1KI": 0.008082126236965302}

17
pytorch/output_max_core/altra_10_2_10_100000_5e-05.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 6, 9, ..., 499981, 499984,
499987]),
col_indices=tensor([ 374, 17783, 22787, ..., 11489, 22480, 43858]),
values=tensor([ 0.3839, 0.4559, -0.2166, ..., -0.4979, -0.2092,
-1.9683]), size=(100000, 100000), nnz=499987,
layout=torch.sparse_csr)
tensor([0.7376, 0.2825, 0.9197, ..., 0.3562, 0.5840, 0.6413])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 499987
Density: 4.99987e-05
Time: 11.732282161712646 seconds

1
pytorch/output_max_core/altra_10_2_10_10000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 128629, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 9998, "MATRIX_DENSITY": 9.998e-05, "TIME_S": 12.030954122543335, "TIME_S_1KI": 0.09353220597643871, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 732.8341610717773, "W": 69.4445813277392, "J_1KI": 5.697270141816988, "W_1KI": 0.539882773929201, "W_D": 58.8845813277392, "J_D": 621.3966868591308, "W_D_1KI": 0.457786201616581, "J_D_1KI": 0.003558965720145387}

16
pytorch/output_max_core/altra_10_2_10_10000_0.0001.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 9996, 9996, 9998]),
col_indices=tensor([7791, 7249, 1656, ..., 9391, 6622, 8506]),
values=tensor([ 0.7435, -0.8659, -0.1431, ..., -0.4350, 0.7354,
-0.2244]), size=(10000, 10000), nnz=9998,
layout=torch.sparse_csr)
tensor([0.1857, 0.8917, 0.3893, ..., 0.2671, 0.5475, 0.0496])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 9998
Density: 9.998e-05
Time: 12.030954122543335 seconds

1
pytorch/output_max_core/altra_10_2_10_10000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 114237, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 1000, "MATRIX_DENSITY": 1e-05, "TIME_S": 13.55590295791626, "TIME_S_1KI": 0.11866473172366448, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 822.1178264427186, "W": 71.60600793772721, "J_1KI": 7.196598531497839, "W_1KI": 0.6268197513741364, "W_D": 61.37600793772721, "J_D": 704.6658750391007, "W_D_1KI": 0.5372690804006339, "J_D_1KI": 0.004703109153782347}

375
pytorch/output_max_core/altra_10_2_10_10000_1e-05.output Normal file
View File

@ -0,0 +1,375 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 1000, 1000, 1000]),
col_indices=tensor([ 204, 4263, 8930, 3170, 3463, 5980, 2442, 8696, 330,
4236, 6038, 8177, 2215, 489, 870, 5998, 9869, 1706,
8379, 8305, 813, 6005, 6901, 7610, 4000, 4910, 5851,
548, 7637, 7846, 5396, 6995, 5740, 1560, 7554, 8679,
5623, 1365, 9449, 3903, 3785, 6022, 966, 7129, 4967,
213, 7596, 796, 1919, 7188, 603, 55, 4993, 5596,
8312, 4981, 4902, 6841, 5937, 6969, 8659, 5360, 7095,
7950, 6265, 9406, 9471, 7952, 7450, 7632, 1496, 2931,
8785, 4004, 7864, 2850, 9769, 5198, 9270, 7352, 2983,
9606, 2885, 4659, 7296, 2176, 1691, 9550, 2730, 1553,
7726, 8049, 3029, 3167, 4757, 7057, 6029, 4786, 8598,
6854, 363, 8316, 5608, 5850, 1798, 2257, 3345, 4097,
282, 8907, 4531, 1212, 2278, 2004, 7972, 8969, 538,
249, 573, 7892, 5270, 7003, 9253, 8273, 572, 4680,
8347, 2988, 6997, 8591, 3749, 9110, 8183, 5319, 305,
8529, 6762, 6873, 9325, 7151, 7385, 3535, 2073, 4062,
6420, 95, 1370, 8646, 5561, 5196, 6468, 7719, 4848,
1512, 4331, 2233, 7183, 8223, 4333, 9795, 2334, 3679,
4850, 6912, 2105, 9560, 8267, 6423, 6455, 9049, 6385,
6120, 6628, 5938, 1280, 4956, 6123, 8569, 4756, 5628,
9103, 6235, 1332, 6631, 2291, 5595, 5929, 9208, 7752,
4894, 5321, 4345, 7022, 6677, 3711, 574, 8100, 4595,
4463, 7922, 4912, 2676, 6831, 8181, 2708, 5108, 3561,
800, 883, 6884, 9487, 5116, 9227, 7842, 5188, 2219,
6649, 9986, 1599, 274, 903, 4163, 3023, 1658, 669,
8401, 6814, 5339, 8275, 8146, 7881, 9976, 5323, 9924,
4141, 7356, 9501, 5110, 6392, 1699, 8355, 4655, 8258,
4935, 9626, 1823, 6221, 2879, 9173, 5168, 9849, 6965,
1630, 4951, 6402, 3834, 418, 5776, 6418, 3460, 9805,
6733, 7984, 6084, 5913, 9294, 2133, 339, 811, 7708,
7645, 6023, 1901, 3142, 4658, 1978, 3084, 2207, 4589,
3830, 9552, 317, 3349, 5714, 2321, 566, 167, 8363,
9320, 1688, 2873, 6902, 2636, 4806, 2306, 444, 2398,
8431, 4795, 3132, 1017, 5347, 2035, 9781, 9913, 3921,
8836, 3816, 2819, 3584, 8221, 5118, 2609, 9174, 9579,
5294, 6011, 8987, 5499, 9193, 4111, 2376, 3493, 1984,
6903, 3190, 980, 5367, 1390, 9417, 8451, 4509, 399,
7409, 564, 6726, 9325, 5146, 4727, 3186, 6314, 8561,
6306, 4168, 2687, 7208, 8765, 883, 1506, 2200, 7530,
8380, 1003, 9825, 2447, 9710, 9458, 3413, 4417, 7304,
6643, 7646, 2360, 7672, 7599, 729, 3876, 7895, 674,
7759, 2222, 6176, 7560, 3270, 7084, 7066, 5130, 2290,
6390, 5273, 7365, 923, 9495, 7289, 6616, 4048, 6165,
1329, 5559, 9777, 1338, 397, 9588, 5260, 9288, 5092,
5054, 8910, 3414, 265, 3967, 1787, 8712, 9059, 9367,
8746, 7426, 98, 6606, 2640, 2539, 1528, 5698, 7001,
6802, 4010, 5165, 9416, 9813, 4190, 6572, 1576, 4719,
7659, 6595, 6900, 5241, 2896, 293, 9633, 5039, 6074,
9406, 5933, 841, 2008, 7351, 7884, 7942, 8477, 3897,
3300, 8446, 4008, 9006, 3400, 733, 250, 9706, 9018,
1417, 4873, 4517, 122, 4101, 7725, 449, 6878, 5572,
5486, 5782, 7693, 7860, 5134, 275, 7829, 7986, 7353,
1948, 4168, 7574, 91, 4523, 5710, 916, 4122, 7519,
8038, 888, 5480, 4974, 7688, 2543, 9042, 7305, 3908,
7833, 5401, 5452, 4733, 6528, 9053, 4734, 4771, 7023,
4290, 3185, 373, 5050, 5200, 9748, 9018, 5183, 9899,
26, 5278, 2165, 7364, 7311, 6002, 1075, 2349, 5357,
7663, 3703, 712, 752, 4841, 767, 1348, 6933, 1919,
8089, 7306, 1835, 843, 8723, 204, 9955, 2133, 3560,
9491, 9437, 4560, 5785, 3350, 747, 7468, 8095, 3367,
6735, 5830, 6725, 15, 7467, 8811, 7511, 4009, 6847,
7632, 2814, 7376, 3919, 9454, 9136, 7619, 1720, 5246,
3851, 8177, 172, 579, 6813, 5046, 2128, 6144, 1017,
2463, 6856, 8124, 1031, 49, 6558, 9659, 8610, 6994,
8901, 9109, 6857, 5055, 6257, 8083, 9157, 4183, 2523,
3470, 4561, 3284, 9907, 3108, 8346, 8762, 8302, 3927,
5178, 6439, 4411, 3505, 6209, 16, 1964, 1277, 6573,
5070, 6709, 4043, 1328, 9572, 3151, 3226, 6015, 6309,
4053, 3853, 115, 7568, 9320, 9278, 9158, 1926, 1978,
9337, 6376, 6573, 6757, 3045, 9164, 6811, 563, 9890,
62, 7399, 734, 914, 8452, 5596, 8973, 3062, 1257,
3825, 1043, 2207, 4443, 8327, 6539, 666, 1701, 4288,
8155, 4276, 7611, 4328, 1175, 9602, 9486, 1815, 7524,
3229, 8789, 5576, 7885, 6662, 1323, 4877, 3139, 711,
5251, 795, 9304, 4571, 5538, 9111, 2557, 4351, 8921,
2777, 6397, 9990, 9948, 9933, 1860, 3129, 4965, 7430,
9295, 1328, 6462, 3724, 9570, 8000, 2324, 2525, 6924,
2863, 5424, 2067, 2913, 4003, 4837, 7782, 6732, 277,
2912, 7806, 4763, 6522, 4314, 3567, 8499, 5705, 6868,
9541, 8162, 9014, 9903, 2692, 4744, 9853, 1852, 603,
6943, 7014, 8978, 2568, 824, 4775, 2418, 9338, 964,
6894, 604, 399, 9110, 440, 8672, 8941, 3259, 8549,
5942, 7719, 7365, 3760, 171, 609, 8941, 4345, 6970,
9533, 8387, 924, 899, 2239, 9992, 92, 8129, 5847,
2598, 3525, 8656, 4504, 8621, 5129, 9831, 6593, 8206,
3964, 8042, 796, 7581, 3042, 3993, 2138, 4714, 8396,
5460, 7271, 3019, 6049, 6331, 9374, 3301, 8344, 9637,
9534, 5358, 190, 4965, 5350, 2621, 3080, 7222, 3932,
6810, 2170, 8369, 4956, 6082, 9260, 5010, 5033, 7900,
4563, 6959, 5701, 2654, 5263, 3500, 9724, 383, 4629,
9363, 6860, 8420, 789, 829, 3071, 9159, 7994, 4369,
3098, 132, 3537, 5441, 2099, 8655, 883, 7886, 5682,
2351, 2370, 4534, 1157, 7587, 9720, 3058, 4625, 5302,
6978, 6456, 2060, 4761, 8846, 8140, 7136, 6136, 6633,
3898, 4325, 7320, 995, 1722, 7821, 2382, 5812, 2781,
5138, 6078, 105, 4909, 9621, 2666, 3217, 218, 3428,
1897, 5008, 5912, 8347, 1152, 6302, 5761, 7004, 3301,
9704, 1183, 2492, 4608, 6222, 1256, 5493, 2994, 8786,
4700, 2873, 7123, 1610, 2700, 5717, 9144, 7511, 6328,
729, 1857, 7843, 8337, 7678, 3217, 4459, 9239, 9907,
2824, 5039, 4370, 505, 6471, 3212, 3612, 7376, 4386,
5744, 2696, 7915, 7448, 9787, 322, 3505, 3425, 3045,
7347, 8228, 4804, 1906, 5318, 8830, 9897, 5331, 0,
5023, 5628, 1932, 4099, 7202, 4080, 4144, 5127, 3957,
3312, 2355, 8165, 9331, 933, 9244, 8147, 2347, 2870,
7704, 3566, 4828, 7686, 975, 7274, 2895, 9862, 4456,
8989, 7548, 9899, 3046, 9921, 7777, 5669, 5902, 8086,
5680, 3995, 5582, 1471, 9044, 3141, 3042, 5974, 2977,
4641, 8909, 1185, 4728, 9130, 3775, 3604, 9661, 2945,
5378, 1896, 4135, 8154, 5735, 2894, 9212, 9997, 165,
1717, 2673, 3120, 4581, 8127, 2922, 3799, 4559, 9975,
8163]),
values=tensor([-5.8057e-01, 3.2745e+00, 7.4622e-01, 5.5744e-01,
-1.3203e+00, -8.5108e-01, 8.8052e-01, 2.4144e-01,
1.0569e+00, 9.5777e-01, -5.8080e-01, 7.0562e-01,
1.7083e+00, 7.6064e-01, 1.2484e+00, -9.3739e-01,
-2.7445e-01, -6.6623e-01, -6.5677e-02, -2.8206e-01,
4.3174e-02, -5.4541e-01, 8.0391e-01, 1.7737e-01,
-2.2026e+00, 4.9869e-01, 3.9421e-02, 2.3896e-01,
1.6738e+00, -1.1715e-01, -4.0740e-03, 6.6337e-01,
1.8015e-01, -7.7665e-01, -2.4325e+00, 1.5665e+00,
8.9376e-01, -1.4290e+00, 9.2200e-01, 2.3189e-01,
1.6833e+00, 6.2490e-01, -1.3063e+00, -3.9870e-01,
-1.5751e+00, 4.0348e-01, -4.3671e-01, -1.1659e+00,
-9.2007e-01, -3.3272e-01, -1.4684e+00, 1.7280e-01,
6.0268e-01, 6.7042e-01, -7.0918e-01, 5.1928e-02,
8.8111e-02, -5.4201e-02, 8.1404e-01, 7.0407e-01,
-2.0609e-01, -1.5587e-01, -1.3066e+00, -5.8263e-01,
2.5682e-01, -4.2616e-01, 2.1908e+00, -1.3538e+00,
-2.2664e-01, 1.6582e-01, 6.3466e-01, 2.4797e-01,
9.4842e-01, -3.2687e-02, -1.0331e+00, -2.0797e+00,
1.2979e+00, 5.5178e-01, 3.2251e-01, -3.1843e-01,
2.0627e+00, -1.7122e+00, -1.6317e-01, 3.7836e-01,
-5.7027e-01, 3.5318e-01, 5.6229e-01, 7.2274e-01,
1.9296e+00, 1.2992e+00, -6.0050e-01, -6.8523e-02,
-9.1011e-02, 4.5224e-01, -6.7661e-01, -1.7651e+00,
-1.0164e+00, 6.3010e-01, -9.0819e-01, 2.6852e-01,
-1.1101e+00, -1.1341e+00, -1.0416e-01, 1.6700e-01,
2.4347e-01, -1.1652e+00, 3.2844e-01, 1.2205e+00,
1.3566e+00, 1.0432e+00, -1.2288e+00, -1.4258e+00,
3.1534e-01, -9.3054e-02, 7.2088e-01, -7.5739e-01,
6.0152e-02, 8.5843e-01, -1.1471e-01, 1.0136e+00,
1.5148e+00, 5.8906e-01, -2.5861e-02, 8.2152e-01,
1.2733e+00, 7.2485e-01, -1.9038e+00, -1.5749e+00,
1.9325e-01, 1.3688e+00, -9.1114e-01, 1.2748e+00,
7.9432e-01, -1.1032e+00, -3.7215e-01, 1.2352e+00,
-6.9823e-01, -2.4130e+00, -5.6550e-01, 9.9392e-01,
-2.1227e+00, 1.5763e+00, -3.3680e-01, -9.8841e-01,
2.6561e-01, -4.1007e-01, 1.1557e+00, 1.8539e-02,
-1.1562e+00, -9.6544e-02, 2.0315e+00, 4.8441e-01,
-1.1283e-02, 1.2218e+00, 2.3421e-01, -8.3243e-01,
2.4328e-01, -5.5248e-01, 2.0275e-01, 1.5952e+00,
9.1178e-01, 1.0176e-01, -5.0845e-01, -1.2482e+00,
1.5095e-01, -2.8977e-01, -1.3504e-02, 1.3189e+00,
1.3034e+00, 6.1404e-01, 2.0866e-02, 1.7327e+00,
6.6260e-01, 2.6118e-01, -8.5127e-01, -6.1754e-01,
-6.1340e-01, 1.8085e+00, 7.4095e-01, 4.1421e-01,
8.4981e-01, 1.0836e+00, 1.3601e+00, -5.4654e-01,
-2.2819e-01, 7.3789e-01, -8.1231e-01, 1.4852e+00,
1.7564e-01, -5.2916e-01, -1.5794e+00, 1.4462e+00,
-6.5425e-01, 5.6403e-01, 1.4336e+00, -7.6091e-01,
3.5890e-01, -1.7265e-01, -1.9162e-01, 8.9323e-01,
-1.3776e-01, 4.0367e-01, -2.1826e-01, 3.9444e-01,
-2.6825e-01, -1.4549e+00, -7.5284e-01, 2.0457e+00,
2.0535e-01, 1.0528e+00, -3.6190e-01, 4.3973e-01,
-7.9614e-01, 1.1119e+00, 4.3706e-01, -8.7439e-01,
-2.2426e+00, 1.1710e+00, 2.2170e-01, -9.3849e-02,
-1.3578e+00, -1.3310e+00, -5.7746e-01, -3.3060e-01,
6.1651e-01, -1.5016e+00, 5.3588e-01, -5.8881e-01,
6.2382e-01, 2.5806e-01, 2.4626e+00, -1.3744e+00,
-7.8302e-01, -1.9451e+00, -2.1342e-01, -8.9730e-01,
5.3032e-01, -1.6068e+00, -8.6662e-01, -9.3414e-01,
1.2584e-02, -3.0036e-01, -4.1805e-01, -1.8076e+00,
-1.4278e+00, 4.4359e-02, -4.0102e-01, 8.9323e-01,
-2.5893e+00, -6.8023e-01, 1.3421e-01, 2.2449e-01,
4.4586e-01, 3.2777e-01, 1.9186e+00, 3.3344e-02,
1.1016e+00, -3.1397e-01, -1.1586e+00, 1.3143e+00,
4.3242e-01, -3.2677e-01, 9.2788e-03, -3.4349e-01,
-1.0923e+00, -1.5645e+00, 1.3428e+00, 6.1153e-01,
-4.4706e-01, 1.4545e+00, -2.1224e+00, -3.4038e-01,
2.4216e-01, 2.1429e+00, 8.6517e-02, 2.3721e-01,
6.0589e-01, 1.3776e+00, 4.2767e-02, 2.4178e-01,
3.7076e-01, 7.6391e-01, 1.6323e+00, 2.9605e-01,
2.1303e+00, -3.5909e-01, 1.3037e+00, -1.2112e+00,
4.4075e-01, 5.7285e-01, -1.8578e-01, 3.0394e-01,
-1.0855e+00, -1.3178e+00, -1.0935e-01, -2.3246e-01,
-3.4023e-02, 1.8324e+00, 2.2272e-01, -1.0789e+00,
-4.3993e-01, 7.2599e-01, -2.9995e-01, -4.9894e-02,
6.3480e-02, -1.3263e+00, -1.3134e+00, -2.7143e-01,
-1.1475e+00, -1.3231e+00, -7.7779e-01, -4.6316e-01,
1.2629e+00, 6.3358e-01, -1.6417e+00, -1.2622e+00,
2.3594e-01, 1.2069e-01, -8.4417e-01, 8.4940e-01,
1.8353e+00, -4.7340e-01, 7.3490e-01, -7.1791e-02,
1.0794e-01, -3.7219e-02, 1.1707e+00, -2.9455e-01,
2.5262e-01, 2.3097e+00, 8.7748e-01, 8.4395e-01,
1.4948e+00, 4.9304e-01, 5.8037e-01, 1.2633e+00,
-8.2680e-01, 7.9194e-01, 8.0205e-01, -1.1203e+00,
-1.3304e+00, -1.4195e+00, 7.3381e-01, -3.4613e-01,
-2.6486e-01, 3.3594e-01, -2.4489e+00, -1.5419e+00,
9.2464e-01, 2.7592e+00, 1.1442e+00, -7.3859e-01,
-1.8809e-01, 1.3966e+00, -3.5501e-01, -3.9923e-01,
2.2557e+00, -1.5468e+00, -5.2415e-01, -2.7345e-01,
-1.6970e-01, -6.2082e-01, -6.1455e-01, 2.8619e-01,
-1.1362e+00, -1.3642e+00, 6.6954e-01, -4.5456e-01,
1.7896e-01, 1.3603e+00, -5.4009e-01, -1.4494e+00,
-1.2638e+00, -5.7037e-01, -2.6938e-01, 1.3731e+00,
3.1355e-01, -9.5876e-01, 3.0276e-01, -6.3430e-01,
-2.0512e+00, 6.0794e-01, 1.4151e+00, -2.4121e-02,
-1.1652e+00, -1.3686e+00, -5.4924e-01, -1.6385e-01,
1.3135e-01, -7.6438e-02, 8.8156e-01, 1.1203e+00,
6.4284e-01, -9.8434e-01, -1.1857e+00, -1.1198e+00,
-6.1699e-03, 1.8244e+00, -5.7553e-01, -3.0129e-01,
-1.1995e-01, 1.6199e+00, 3.1169e-01, 1.7213e+00,
-7.2977e-02, 2.0596e+00, -1.4830e+00, 3.3914e-01,
-5.5543e-01, 1.9406e+00, 1.0497e+00, 2.9452e-01,
-2.6102e-01, -1.0521e+00, 5.9684e-01, -1.6700e+00,
2.2799e+00, 9.2103e-02, -9.1363e-01, -1.8654e+00,
-4.4513e-01, -1.4915e+00, 2.6626e-01, -3.3857e-01,
-6.4206e-01, 4.6132e-01, 2.6807e-01, 5.8230e-02,
-5.5815e-01, 9.2433e-02, 1.2304e+00, 1.3601e+00,
-1.8100e+00, -2.8175e-01, -1.4489e+00, 5.7361e-01,
8.7140e-01, 5.9861e-02, 1.2264e-01, -4.2743e-01,
-1.5406e+00, 1.7147e+00, -1.6347e+00, -7.9608e-01,
-7.7060e-03, -1.6637e+00, 1.3828e-01, 1.5533e-01,
9.7930e-01, -1.4815e+00, 6.0441e-01, 1.9755e+00,
-8.2869e-02, 3.4760e-01, -2.3241e-01, 4.1308e-01,
-2.7169e-01, -1.3484e+00, 6.9031e-01, -6.2154e-01,
-5.4025e-01, 2.0401e-01, -6.8680e-01, -9.3677e-01,
-5.3733e-01, -7.0802e-01, 1.4392e+00, -7.7876e-01,
-1.6313e+00, -3.9798e-01, 5.9074e-01, 9.5578e-01,
1.6084e+00, -4.0742e-01, -3.9755e-01, -7.6211e-01,
-8.7877e-01, 1.4363e+00, -1.0129e+00, 3.6504e-01,
4.1917e-01, -2.4226e-01, 9.2537e-01, 2.3566e-01,
1.2886e+00, 1.9324e+00, -6.4127e-01, 1.7214e+00,
1.5407e+00, -4.5775e-01, 5.7949e-01, -1.2643e-01,
1.4741e+00, -1.2439e+00, -1.7828e+00, 6.6937e-01,
-6.6326e-01, 1.2327e+00, 4.6475e-01, 1.4165e+00,
4.6574e-01, 1.4612e+00, -5.4139e-01, 5.6293e-01,
-3.8138e-01, 1.2663e+00, 2.0042e-02, 1.3303e-01,
-1.0378e+00, 6.1238e-01, 2.1972e+00, 2.1507e-01,
1.3862e+00, 1.2746e+00, 2.6223e+00, -8.1597e-01,
-8.2116e-01, -7.3659e-01, -9.2594e-01, -1.7470e-01,
2.7039e-02, -7.6973e-01, 2.4903e+00, 4.1810e-01,
2.7132e+00, 4.8549e-02, 5.5688e-01, -2.7176e-01,
7.2752e-01, -1.3830e+00, -1.7083e+00, -9.1716e-01,
-1.3355e+00, 2.6338e-01, -4.4779e-01, -1.1385e+00,
7.9882e-01, -2.2198e+00, -5.1253e-01, -3.1148e-01,
7.6823e-01, 1.5365e+00, 7.8355e-01, 1.4040e+00,
4.9512e-02, -2.9482e-01, -2.4342e+00, -1.9444e-05,
9.1415e-01, 4.2319e-01, -1.0324e+00, 1.3540e+00,
1.1204e+00, 4.6601e-01, 1.0324e+00, -3.2695e-01,
8.9230e-01, -4.2988e-01, 3.0006e-01, -5.3584e-01,
4.2397e-01, 5.0670e-01, -1.1385e-01, -4.0557e-01,
-2.0720e+00, 8.9273e-01, -1.5016e+00, -1.4142e-02,
-3.3954e-01, 1.0224e+00, -1.3160e-01, 3.8621e-01,
9.7496e-02, 1.0362e+00, -7.7884e-01, 6.1015e-01,
-1.0727e+00, 2.6818e-01, 1.3017e+00, -7.9516e-02,
-7.8573e-01, 8.0817e-01, 2.2825e+00, -1.0096e+00,
1.2440e+00, -8.5393e-01, -4.6929e-01, 5.4000e-01,
3.8111e-02, -4.9862e-01, 2.4955e+00, -1.1128e+00,
-5.3435e-01, -2.9269e-02, 6.5858e-01, 1.2936e+00,
1.9218e-01, 3.0575e-01, 5.7645e-02, 1.5694e+00,
8.4436e-01, 1.6478e+00, -6.6566e-01, 3.7266e-01,
-7.1747e-01, 7.1670e-01, 1.3989e+00, 7.0936e-02,
-1.0004e+00, -4.4010e-01, -7.0143e-01, -1.2976e+00,
1.2683e+00, -7.9554e-01, -1.4615e+00, 2.1725e+00,
2.4801e+00, -7.1164e-01, 1.4826e+00, -5.8458e-01,
7.1482e-01, 8.2402e-02, -1.7162e+00, 6.7785e-01,
-2.7357e-01, 1.0521e+00, -8.3013e-01, -2.4952e-01,
-1.1864e-01, 1.0675e-01, -2.0809e-01, 1.4979e+00,
-4.9142e-01, 4.2028e-02, 1.7482e-01, -2.1030e-01,
1.6613e+00, 1.2894e-01, 2.7035e+00, 1.1323e+00,
-4.5184e-01, 2.9033e-02, -1.1740e+00, 6.6560e-01,
1.2952e+00, -7.7957e-02, 1.3908e+00, 3.5068e-01,
-9.5930e-01, -1.3995e+00, 3.5166e-01, 1.6498e-01,
2.2008e+00, -2.6724e+00, 1.5839e+00, 1.0270e+00,
6.3679e-01, 7.5355e-01, -1.0124e+00, 1.0162e+00,
1.2156e+00, -1.3413e+00, 5.2420e-01, 4.9867e-01,
-8.6964e-01, -6.9166e-01, 6.3485e-01, 8.9134e-01,
1.4915e+00, 1.8118e-01, -1.0352e+00, 6.9274e-01,
1.0694e+00, 3.9906e-01, 1.6450e+00, -7.8844e-02,
1.4868e+00, 5.4787e-01, 1.2179e+00, 8.3875e-01,
1.8397e+00, -3.7309e-01, -1.5918e-01, -4.1885e-01,
-2.3334e-02, 5.3439e-01, 1.2795e+00, -5.1847e-01,
3.9250e-01, 2.6775e-01, 1.2099e+00, -6.6824e-01,
-1.4180e-01, 1.7346e+00, 4.8250e-01, 4.7158e-01,
3.7317e-01, -1.9602e+00, -1.8783e+00, 2.7278e-01,
-3.0196e-01, 8.5099e-01, -1.3338e+00, -7.3483e-01,
3.0283e-01, -1.2627e-01, -1.3816e+00, 8.1513e-02,
-5.3633e-01, 2.7753e-01, 5.1545e-01, 8.2115e-01,
-4.8921e-01, -5.1688e-01, 9.0242e-01, 2.1956e+00,
1.1571e+00, -1.8859e+00, -6.0886e-01, -6.3269e-01,
1.6371e-01, 1.2471e+00, -1.6294e-01, 7.3368e-01,
-1.2793e+00, -1.4308e+00, 1.1876e+00, -5.4812e-01,
-2.6111e-01, 8.2727e-01, -2.9034e+00, 1.7822e+00,
-2.0522e-01, 1.6649e+00, -1.3560e+00, 9.9972e-01,
1.6951e-01, -2.3309e+00, 2.4207e+00, -6.9441e-01,
-5.6229e-02, -3.0195e+00, 2.3961e+00, -1.2494e+00,
7.1937e-01, 1.4122e+00, -4.8333e-01, 3.4736e-02,
8.7187e-01, 1.4946e+00, 1.9533e-01, -6.1233e-01,
5.9861e-01, 1.8633e-01, -2.7917e-01, 1.0876e+00,
9.1793e-01, 1.6403e+00, -1.5253e+00, 9.1821e-01,
3.5092e-01, 2.5064e+00, 6.5185e-01, -2.2226e+00,
1.3760e+00, 2.3776e+00, -9.1211e-01, -1.1011e+00,
-1.0562e+00, 2.2037e-01, -1.0005e+00, -6.2230e-01,
-4.7244e-01, -1.2379e+00, 1.2417e+00, -3.0415e+00,
-7.1942e-01, 1.5569e+00, 7.7018e-01, 4.8289e-01,
-1.2474e+00, -1.3318e-01, -1.1053e+00, 1.3295e-01,
-1.0419e-01, 7.9604e-01, 4.0443e-01, 3.9375e-01,
-4.9862e-02, 1.1707e+00, 4.2326e-01, 1.3690e+00,
1.0255e+00, -1.8492e+00, -1.7939e-01, -6.8027e-01,
9.3390e-02, 2.2862e+00, 5.4699e-01, -7.9402e-01,
-6.0203e-01, -1.6840e+00, 4.3145e-01, -1.5633e+00,
-3.6103e-01, 1.1163e+00, 9.3996e-01, -1.4514e+00,
-1.2327e-01, 4.2094e-01, -9.0919e-01, 9.1369e-01,
-5.3441e-01, 7.1357e-01, -1.2426e-01, -7.0871e-01,
-1.2839e+00, 7.1327e-01, 8.2112e-01, -6.3594e-01,
5.3437e-01, -2.4992e-01, 1.1300e+00, 7.8631e-02,
3.6491e-01, -8.3755e-01, -1.3907e+00, -6.7980e-01,
-1.6985e+00, 2.9771e+00, 9.6965e-02, 6.8434e-01,
9.4595e-01, -3.7084e-01, 1.5932e-01, 1.3212e-01,
-8.0803e-01, 6.6915e-01, -4.6427e-01, 1.1018e+00,
-6.7640e-01, 4.1715e-01, 1.4843e+00, -5.5934e-03,
8.4507e-01, -9.3781e-01, 1.8580e-01, 2.8702e-01,
1.9062e+00, -2.9965e+00, -7.5127e-01, -2.0902e-01,
-9.5840e-01, 7.2294e-01, 1.3191e+00, -1.0796e+00,
2.0425e+00, 1.4681e-01, 5.0829e-01, 2.8279e-01,
7.0198e-01, 1.2190e+00, 5.9683e-01, 7.8600e-01,
7.3204e-01, -1.1617e+00, 1.0280e+00, -7.2023e-01,
2.4889e-01, 1.8015e+00, 9.8026e-02, 1.8320e+00,
8.7494e-01, 3.6755e-01, -2.9441e-01, -2.0004e+00,
-7.5414e-01, -7.9335e-01, -1.5935e-01, 2.7505e+00,
-6.1145e-01, -9.7560e-02, -1.4664e+00, 1.3932e+00,
6.5238e-01, 2.3698e+00, 4.2610e-01, 5.4787e-01,
9.5760e-01, -9.5063e-01, -1.4945e+00, 2.7731e-01,
9.0244e-01, 1.5433e+00, -4.5351e-01, 3.3418e-01,
-9.2268e-01, 4.1958e-01, 7.8572e-01, 5.4617e-01,
-3.4244e-01, -5.7557e-01, 7.1616e-01, 1.6418e+00,
1.8550e+00, 9.2130e-01, 1.5370e+00, -3.9326e-02,
1.9495e-01, -1.5660e+00, -7.6299e-01, -6.7067e-01,
-1.0029e+00, 5.6061e-01, -1.2435e+00, 9.6545e-01,
1.2522e+00, -7.1200e-01, 1.0962e+00, -9.0625e-01,
-1.6041e+00, 1.3440e-01, 7.5016e-01, -8.9344e-01,
1.4220e+00, 8.9624e-01, 1.4860e+00, 4.7710e-01,
6.0615e-01, 8.1113e-02, 4.8093e-01, 1.0426e+00,
-8.4289e-01, -2.0973e+00, 1.4694e+00, 8.1465e-01,
5.0596e-01, -9.1039e-01, 1.9932e-02, -1.6439e+00,
-9.6835e-01, -1.9809e+00, 6.3412e-01, -4.1401e-01,
-9.8539e-02, -5.5056e-02, 1.2799e+00, -4.2906e-01,
-1.0369e-01, -3.9120e-01, -1.3900e+00, 1.2275e+00,
1.3327e+00, -4.7165e-01, -9.6054e-02, 1.3523e-01,
9.6209e-02, -1.0132e+00, 8.0297e-02, -1.9039e+00,
1.2108e+00, -2.9279e-01, -4.8785e-01, -4.3345e-01,
-1.4850e-01, -5.8455e-01, 6.1613e-01, -4.5960e-01,
6.2385e-01, 3.9640e-01, -2.7009e-01, 3.0861e-01,
-2.7755e+00, 5.5387e-01, -1.2093e+00, 2.3832e+00,
1.6573e+00, 8.8648e-01, -1.5635e+00, 5.6755e-01,
2.2215e+00, -1.0175e+00, -2.0150e+00, 1.2504e+00,
6.0568e-01, -1.2960e+00, 8.0207e-01, 3.9288e-01,
-2.4731e-01, -3.6153e-01, 1.3366e+00, -2.6135e+00,
1.6461e-01, 2.5829e+00, 7.8608e-01, 1.4415e+00,
-2.0807e-01, -3.3511e-01, -7.2562e-01, 1.7444e+00]),
size=(10000, 10000), nnz=1000, layout=torch.sparse_csr)
tensor([0.0034, 0.1810, 0.4190, ..., 0.7648, 0.0071, 0.9275])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 1000
Density: 1e-05
Time: 13.55590295791626 seconds

1
pytorch/output_max_core/altra_10_2_10_10000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 141912, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 5000, "MATRIX_DENSITY": 5e-05, "TIME_S": 11.136729955673218, "TIME_S_1KI": 0.0784763089497239, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 762.7767884826661, "W": 73.15736656251944, "J_1KI": 5.374998509517632, "W_1KI": 0.5155121946172236, "W_D": 62.79236656251944, "J_D": 654.7059026086332, "W_D_1KI": 0.44247397374795255, "J_D_1KI": 0.003117946147950508}

15
pytorch/output_max_core/altra_10_2_10_10000_5e-05.output Normal file
View File

@ -0,0 +1,15 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 4999, 5000, 5000]),
col_indices=tensor([9413, 261, 7246, ..., 8062, 3966, 6079]),
values=tensor([0.5421, 0.3227, 1.2683, ..., 1.2444, 0.6712, 1.2899]),
size=(10000, 10000), nnz=5000, layout=torch.sparse_csr)
tensor([0.0413, 0.5194, 0.3898, ..., 0.3926, 0.4036, 0.8183])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 5000
Density: 5e-05
Time: 11.136729955673218 seconds

1
pytorch/output_max_core/altra_10_2_10_20000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 126855, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 39996, "MATRIX_DENSITY": 9.999e-05, "TIME_S": 11.35659384727478, "TIME_S_1KI": 0.08952421147983745, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 713.6331734085084, "W": 68.3690498724268, "J_1KI": 5.625581754038141, "W_1KI": 0.5389543169163754, "W_D": 58.0540498724268, "J_D": 605.9656513726711, "W_D_1KI": 0.4576410064437886, "J_D_1KI": 0.0036075913952448744}

16
pytorch/output_max_core/altra_10_2_10_20000_0.0001.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 8, ..., 39994, 39994, 39996]),
col_indices=tensor([ 3304, 2257, 6792, ..., 5265, 9578, 11711]),
values=tensor([-1.6347, -1.4269, -0.0725, ..., -0.3851, -2.3655,
-1.2084]), size=(20000, 20000), nnz=39996,
layout=torch.sparse_csr)
tensor([0.4129, 0.9449, 0.8749, ..., 0.0510, 0.5936, 0.6265])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 39996
Density: 9.999e-05
Time: 11.35659384727478 seconds

1
pytorch/output_max_core/altra_10_2_10_20000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 166002, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 4000, "MATRIX_DENSITY": 1e-05, "TIME_S": 13.42890977859497, "TIME_S_1KI": 0.08089607220753348, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1000.7195114231109, "W": 78.87462952277818, "J_1KI": 6.028358160884272, "W_1KI": 0.4751426460089528, "W_D": 68.55462952277819, "J_D": 869.784818983078, "W_D_1KI": 0.4129747203213105, "J_D_1KI": 0.0024877695468808235}

16
pytorch/output_max_core/altra_10_2_10_20000_1e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 4000, 4000, 4000]),
col_indices=tensor([ 3448, 7160, 12825, ..., 18574, 10830, 15045]),
values=tensor([ 1.8380, 0.6299, -0.7420, ..., 1.2355, -0.0735,
-1.7277]), size=(20000, 20000), nnz=4000,
layout=torch.sparse_csr)
tensor([0.0324, 0.5478, 0.6339, ..., 0.9725, 0.3076, 0.7119])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 4000
Density: 1e-05
Time: 13.42890977859497 seconds

1
pytorch/output_max_core/altra_10_2_10_20000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 129547, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 19999, "MATRIX_DENSITY": 4.99975e-05, "TIME_S": 10.910138130187988, "TIME_S_1KI": 0.0842176054265092, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 763.2134310245514, "W": 73.3484674791233, "J_1KI": 5.891401815746805, "W_1KI": 0.5661919417595414, "W_D": 62.9784674791233, "J_D": 655.3103820347786, "W_D_1KI": 0.4861437739131227, "J_D_1KI": 0.0037526440126990413}

16
pytorch/output_max_core/altra_10_2_10_20000_5e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 3, ..., 19998, 19999, 19999]),
col_indices=tensor([ 3408, 17814, 18856, ..., 7525, 14693, 7186]),
values=tensor([-0.3733, 1.9481, 0.7711, ..., 0.4398, 0.2745,
1.4792]), size=(20000, 20000), nnz=19999,
layout=torch.sparse_csr)
tensor([0.3457, 0.8868, 0.6712, ..., 0.7459, 0.0711, 0.3442])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 19999
Density: 4.99975e-05
Time: 10.910138130187988 seconds

1
pytorch/output_max_core/altra_10_2_10_50000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 133208, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 249990, "MATRIX_DENSITY": 9.9996e-05, "TIME_S": 12.71963095664978, "TIME_S_1KI": 0.09548698994542205, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 985.8087538051604, "W": 80.58352961679094, "J_1KI": 7.4005221443543965, "W_1KI": 0.6049451205392389, "W_D": 70.19352961679094, "J_D": 858.7039595532416, "W_D_1KI": 0.5269468021199247, "J_D_1KI": 0.003955819486216479}

17
pytorch/output_max_core/altra_10_2_10_50000_0.0001.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 4, 8, ..., 249984, 249985,
249990]),
col_indices=tensor([ 3160, 5078, 16221, ..., 24450, 42207, 48603]),
values=tensor([ 0.5895, 1.6495, 1.1851, ..., -0.0503, -0.0653,
-0.4288]), size=(50000, 50000), nnz=249990,
layout=torch.sparse_csr)
tensor([0.7615, 0.5458, 0.5625, ..., 0.6577, 0.6072, 0.4727])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 249990
Density: 9.9996e-05
Time: 12.71963095664978 seconds

1
pytorch/output_max_core/altra_10_2_10_50000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 130366, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 25000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.279229879379272, "TIME_S_1KI": 0.07884900878587417, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 741.5320628929138, "W": 69.08692092498494, "J_1KI": 5.68807866232694, "W_1KI": 0.529945851870771, "W_D": 58.651920924984935, "J_D": 629.5298637402058, "W_D_1KI": 0.44990197539991206, "J_D_1KI": 0.003451068341438044}

16
pytorch/output_max_core/altra_10_2_10_50000_1e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 3, ..., 24999, 25000, 25000]),
col_indices=tensor([ 7797, 37092, 5414, ..., 528, 18590, 20600]),
values=tensor([-1.9658, 0.6630, 0.7723, ..., 0.1582, -1.1384,
-1.5153]), size=(50000, 50000), nnz=25000,
layout=torch.sparse_csr)
tensor([0.4818, 0.0474, 0.9164, ..., 0.8815, 0.1094, 0.4529])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 25000
Density: 1e-05
Time: 10.279229879379272 seconds

1
pytorch/output_max_core/altra_10_2_10_50000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 108159, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 124996, "MATRIX_DENSITY": 4.99984e-05, "TIME_S": 10.306652784347534, "TIME_S_1KI": 0.09529167969699733, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 731.3704210281371, "W": 81.3855206944779, "J_1KI": 6.761993186217857, "W_1KI": 0.7524618450103819, "W_D": 71.08052069447791, "J_D": 638.7646095228195, "W_D_1KI": 0.6571854463750396, "J_D_1KI": 0.0060761050525156455}

17
pytorch/output_max_core/altra_10_2_10_50000_5e-05.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /space/jenkins/workspace/Releases/pytorch-dls/pytorch-dls/aten/src/ATen/SparseCsrTensorImpl.cpp:55.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 3, ..., 124993, 124994,
124996]),
col_indices=tensor([35906, 45670, 29546, ..., 25799, 6739, 9431]),
values=tensor([-0.1187, 0.3153, -0.5399, ..., -0.0908, 1.6164,
0.1624]), size=(50000, 50000), nnz=124996,
layout=torch.sparse_csr)
tensor([0.8602, 0.3600, 0.9355, ..., 0.2525, 0.8589, 0.5645])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 124996
Density: 4.99984e-05
Time: 10.306652784347534 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_100000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 99216, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 999942, "MATRIX_DENSITY": 9.99942e-05, "TIME_S": 10.639978885650635, "TIME_S_1KI": 0.10724055480618686, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1508.5351157140733, "W": 144.33, "J_1KI": 15.204554867300367, "W_1KI": 1.45470488630866, "W_D": 124.57125000000002, "J_D": 1302.0169405764343, "W_D_1KI": 1.2555560595065314, "J_D_1KI": 0.01265477402340884}

17
pytorch/output_max_core/epyc_7313p_10_2_10_100000_0.0001.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 8, 12, ..., 999913, 999928,
999942]),
col_indices=tensor([31827, 39989, 40960, ..., 92246, 96901, 99105]),
values=tensor([ 0.1922, -1.7217, -0.3618, ..., -0.5679, -1.6956,
-0.8413]), size=(100000, 100000), nnz=999942,
layout=torch.sparse_csr)
tensor([0.1453, 0.8510, 0.4991, ..., 0.4999, 0.3000, 0.5090])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 999942
Density: 9.99942e-05
Time: 10.639978885650635 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_100000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 148750, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 100000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.243898630142212, "TIME_S_1KI": 0.06886654541272075, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1237.544745504856, "W": 116.25, "J_1KI": 8.319628541209116, "W_1KI": 0.7815126050420168, "W_D": 96.435, "J_D": 1026.6032475936413, "W_D_1KI": 0.6483025210084035, "J_D_1KI": 0.004358336275686746}

17
pytorch/output_max_core/epyc_7313p_10_2_10_100000_1e-05.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 99998, 99998,
100000]),
col_indices=tensor([80120, 15447, 42285, ..., 16971, 5943, 65967]),
values=tensor([-0.4609, 0.4429, -0.6032, ..., 1.6776, 0.1248,
-0.2813]), size=(100000, 100000), nnz=100000,
layout=torch.sparse_csr)
tensor([0.6209, 0.5769, 0.0503, ..., 0.5899, 0.8007, 0.3193])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 100000
Density: 1e-05
Time: 10.243898630142212 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_100000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 135703, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 499989, "MATRIX_DENSITY": 4.99989e-05, "TIME_S": 11.035610914230347, "TIME_S_1KI": 0.08132179033794644, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1575.6648641180993, "W": 144.33, "J_1KI": 11.611127713595861, "W_1KI": 1.0635726549892044, "W_D": 124.54000000000002, "J_D": 1359.6154796457292, "W_D_1KI": 0.9177394751774097, "J_D_1KI": 0.006762853254367329}

17
pytorch/output_max_core/epyc_7313p_10_2_10_100000_5e-05.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 10, 14, ..., 499981, 499985,
499989]),
col_indices=tensor([ 6332, 7243, 12909, ..., 64154, 80886, 88555]),
values=tensor([-1.0337, -2.3858, -1.2258, ..., 0.4265, -1.3399,
0.3314]), size=(100000, 100000), nnz=499989,
layout=torch.sparse_csr)
tensor([0.1034, 0.8164, 0.1667, ..., 0.0323, 0.1870, 0.4890])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 499989
Density: 4.99989e-05
Time: 11.035610914230347 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_10000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 393946, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 10000, "MATRIX_DENSITY": 0.0001, "TIME_S": 10.475416898727417, "TIME_S_1KI": 0.026590996986204752, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1043.1735118150712, "W": 98.17, "J_1KI": 2.648011432569619, "W_1KI": 0.24919659039563802, "W_D": 78.4225, "J_D": 833.3327363789082, "W_D_1KI": 0.19906916176328734, "J_D_1KI": 0.0005053209367864817}

16
pytorch/output_max_core/epyc_7313p_10_2_10_10000_0.0001.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 3, ..., 9999, 10000, 10000]),
col_indices=tensor([ 799, 3531, 4424, ..., 2152, 3390, 5971]),
values=tensor([-1.0371, -0.7496, 0.5134, ..., -0.2092, -1.3121,
1.0092]), size=(10000, 10000), nnz=10000,
layout=torch.sparse_csr)
tensor([0.6037, 0.2302, 0.5312, ..., 0.8081, 0.7734, 0.2639])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 10000
Density: 0.0001
Time: 10.475416898727417 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_10000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 521200, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 1000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.794004440307617, "TIME_S_1KI": 0.020709908749630884, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1063.2545353436471, "W": 96.01000000000002, "J_1KI": 2.040012539032324, "W_1KI": 0.18420951650038375, "W_D": 76.17125000000001, "J_D": 843.5519948473575, "W_D_1KI": 0.14614591327705298, "J_D_1KI": 0.0002804027499559727}

375
pytorch/output_max_core/epyc_7313p_10_2_10_10000_1e-05.output Normal file
View File

@ -0,0 +1,375 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 1000, 1000, 1000]),
col_indices=tensor([2651, 5194, 7832, 3269, 3153, 6193, 1893, 284, 7078,
5667, 7959, 1064, 8594, 8144, 7397, 5538, 6515, 6301,
4177, 3693, 1345, 1158, 6445, 3512, 9067, 4839, 3233,
1863, 9340, 4228, 2096, 3070, 885, 4769, 7787, 2699,
1244, 3774, 3211, 7783, 8928, 9715, 7481, 6309, 8598,
9034, 5559, 7173, 4518, 3263, 9720, 3447, 1238, 7158,
1593, 1979, 7581, 5806, 3514, 9434, 1684, 6486, 1786,
8770, 5621, 6381, 8294, 7955, 6573, 3175, 5913, 1848,
7208, 5423, 2043, 5218, 2668, 3472, 358, 2022, 902,
8056, 3604, 1267, 4576, 4232, 4721, 6474, 7978, 632,
298, 9688, 6751, 8813, 3023, 9619, 8379, 2803, 9252,
3847, 1645, 1326, 4836, 2701, 4068, 8407, 1903, 2534,
4352, 7716, 2476, 5190, 3192, 9979, 6610, 1291, 5849,
921, 2601, 6271, 8908, 3887, 9189, 906, 2865, 6650,
5741, 7359, 7633, 941, 2424, 320, 793, 1936, 5267,
4305, 5250, 6401, 4517, 5487, 2100, 6943, 4750, 6680,
4793, 9189, 1945, 309, 2134, 6608, 2883, 6694, 5042,
9560, 56, 2564, 5699, 5829, 6813, 3872, 7571, 9357,
9610, 7918, 74, 8918, 2539, 7850, 2625, 1672, 3744,
2412, 2187, 3933, 508, 3556, 6460, 9343, 618, 1089,
7523, 7481, 8611, 6876, 3324, 8615, 9595, 8275, 20,
3768, 405, 7423, 7544, 716, 9896, 6867, 1021, 8781,
7209, 7205, 6561, 6960, 5753, 3644, 1543, 2611, 2351,
5711, 8308, 6886, 3678, 3277, 1208, 8907, 2478, 1932,
1119, 1954, 3279, 9307, 4003, 1233, 6291, 6637, 8934,
7038, 7739, 3067, 79, 1093, 719, 1820, 304, 7986,
924, 4665, 8779, 6072, 9776, 3072, 2711, 778, 9468,
7006, 5226, 882, 2938, 9940, 1980, 2322, 1826, 2073,
4396, 7895, 8298, 2095, 7949, 2956, 8183, 1474, 2670,
7574, 8647, 7958, 8841, 2482, 6017, 872, 8517, 2208,
1475, 1494, 369, 6764, 285, 329, 9296, 3929, 3469,
6555, 3525, 6358, 5176, 7339, 8635, 2987, 8977, 8963,
7204, 739, 5175, 4362, 9121, 8149, 317, 3075, 2936,
5244, 310, 7412, 1679, 3564, 5850, 4454, 5900, 8870,
6803, 3329, 183, 7506, 2212, 6552, 5702, 7244, 7983,
6440, 565, 5188, 3881, 7669, 6646, 7453, 1118, 2039,
9612, 4238, 3109, 9030, 6686, 6278, 9201, 4705, 9721,
1141, 8704, 9169, 200, 9298, 3778, 3950, 9955, 4235,
9613, 9876, 7019, 3201, 3529, 5888, 3814, 7970, 1736,
8457, 8378, 3518, 5236, 2666, 1473, 2897, 4577, 8950,
8047, 306, 8113, 3897, 9585, 6396, 2539, 2939, 4877,
1477, 7243, 8467, 7868, 6703, 6951, 3607, 4211, 2396,
8409, 8208, 750, 9613, 1384, 5535, 8840, 3597, 337,
9464, 6497, 4147, 2265, 6662, 9448, 4647, 4655, 8169,
494, 8487, 202, 9324, 2748, 2860, 5436, 8977, 7887,
7711, 1984, 9456, 5654, 9799, 7066, 7667, 3229, 990,
6865, 5047, 6222, 4219, 6807, 4545, 3822, 3110, 9720,
8822, 5465, 3638, 5484, 2409, 5509, 8403, 4557, 5249,
3950, 4964, 7420, 8500, 44, 7725, 4249, 6174, 6462,
1728, 2924, 2929, 2869, 5126, 3252, 2904, 1396, 2125,
4314, 3802, 5409, 6010, 9957, 7503, 132, 228, 1685,
70, 9029, 416, 7872, 1934, 9754, 2905, 9318, 9134,
8078, 5829, 5516, 1169, 9643, 497, 1800, 4984, 5249,
1947, 7731, 7218, 790, 6676, 2892, 2689, 4744, 2890,
4724, 3271, 666, 5869, 3262, 6655, 389, 9760, 2632,
5043, 6605, 9169, 7248, 4803, 216, 1871, 505, 975,
2529, 4807, 4849, 5236, 3351, 3030, 7924, 3378, 5319,
8854, 710, 3683, 6667, 8360, 2600, 7307, 4080, 4391,
8124, 7041, 3004, 3204, 6221, 3353, 6082, 8606, 8432,
8028, 2614, 9954, 3117, 2912, 2145, 4031, 2587, 7452,
7836, 7659, 7757, 7321, 9874, 4413, 4343, 6004, 3952,
9870, 9886, 6752, 9250, 2306, 3979, 5797, 9929, 6148,
5635, 658, 8624, 1764, 4054, 7010, 2351, 346, 6319,
6374, 7423, 6333, 3381, 9249, 2896, 1416, 9921, 4586,
2156, 6411, 1439, 7234, 1244, 8778, 9364, 7668, 6955,
704, 6760, 8996, 4060, 4331, 3926, 2543, 9957, 3486,
2067, 3935, 1577, 3067, 6048, 6888, 6111, 9085, 67,
538, 1702, 8694, 6790, 2510, 4231, 4806, 8893, 4258,
3782, 7750, 3163, 8178, 2067, 5742, 6621, 6475, 1780,
3441, 6292, 1224, 3485, 7090, 4891, 4164, 5573, 738,
7459, 5237, 7191, 7406, 94, 9686, 131, 2332, 5956,
2558, 2007, 6257, 4602, 4044, 8890, 7079, 1519, 564,
8797, 3038, 7290, 3511, 3621, 2631, 2735, 7340, 3940,
1494, 2415, 6568, 7892, 6974, 1840, 4656, 7822, 3670,
7309, 4967, 1811, 4851, 5396, 5820, 9572, 277, 9165,
1908, 5008, 6932, 6402, 6329, 5978, 2524, 4615, 7263,
5901, 8028, 1777, 464, 8304, 6573, 723, 2249, 3126,
5156, 6870, 9480, 791, 5869, 5140, 4196, 4312, 1061,
4077, 3490, 6730, 2518, 2553, 3502, 7350, 9702, 4258,
1039, 790, 4687, 6151, 8379, 944, 4541, 5894, 7201,
4968, 2121, 5453, 7979, 3283, 1350, 5018, 1680, 9014,
511, 6697, 4504, 6424, 9184, 7661, 3651, 1139, 8233,
3776, 1844, 6354, 8654, 4346, 4376, 4750, 3424, 936,
4602, 3779, 2291, 7523, 4924, 2835, 8498, 5614, 1370,
7334, 254, 514, 7197, 6862, 9855, 1729, 6216, 8283,
1512, 647, 6134, 7060, 5279, 8779, 6926, 6476, 5943,
494, 2966, 5020, 3186, 650, 7401, 1778, 9972, 7032,
4666, 4522, 8461, 8027, 846, 8378, 8465, 2390, 3627,
3324, 7310, 4774, 4014, 7797, 1969, 4981, 2315, 4106,
1402, 4777, 713, 3700, 8049, 492, 7351, 9671, 6163,
9896, 1062, 2117, 2353, 9147, 6912, 9091, 6015, 1423,
1003, 3195, 2892, 4535, 5259, 8724, 7366, 3561, 1174,
8520, 8026, 7422, 5966, 2944, 1884, 4088, 1653, 6522,
1208, 6919, 6142, 8990, 8251, 5638, 7924, 769, 5154,
909, 3759, 8769, 8248, 4069, 3843, 5658, 2566, 252,
1821, 7937, 2729, 7319, 198, 400, 4632, 3791, 8892,
1411, 5597, 1311, 7035, 7950, 1888, 1993, 261, 1865,
4452, 3418, 2392, 7166, 8283, 3045, 429, 3624, 1176,
1787, 4977, 287, 2193, 36, 4827, 3698, 4450, 9134,
2953, 2538, 9415, 8005, 3228, 7608, 7833, 7284, 3599,
8025, 2333, 6150, 5561, 1852, 8506, 9043, 1005, 1193,
9130, 790, 451, 9369, 5833, 3365, 4841, 3595, 8961,
357, 6923, 7299, 5662, 1866, 7766, 5069, 4715, 8848,
4973, 4030, 9499, 1106, 3051, 4782, 2997, 6843, 39,
2610, 9794, 8714, 1729, 1668, 3168, 5988, 4404, 8837,
7047, 665, 4242, 2750, 3365, 8548, 8224, 3822, 4872,
618, 7517, 9791, 14, 7008, 1076, 3686, 1606, 2679,
8052, 9459, 1489, 7531, 2764, 8732, 1522, 863, 1124,
1921, 3501, 5916, 9759, 5122, 4643, 9454, 142, 239,
8713, 3520, 5896, 9646, 3673, 3368, 5613, 4791, 867,
140, 1725, 2245, 3354, 4704, 9100, 5639, 5542, 6002,
9408]),
values=tensor([ 4.2350e-01, 8.2305e-01, 8.0390e-01, 7.0087e-01,
3.1686e-01, 2.4294e-01, 2.8762e-02, -1.1092e+00,
4.8117e-01, 3.6740e-02, 4.5364e-01, -1.9907e+00,
2.0109e+00, -1.5982e+00, -3.6439e-01, 2.5415e-02,
1.9981e+00, 1.8933e-01, 3.5725e-02, -1.4839e-01,
5.3192e-01, -2.0954e+00, -1.3819e-01, -2.7449e-01,
-1.0437e+00, -1.8112e+00, 9.4652e-01, -7.6702e-01,
1.9586e+00, -9.2641e-02, -1.5099e+00, -9.1718e-02,
-1.1535e+00, 4.1261e-01, 6.4407e-01, -8.5845e-01,
5.4943e-01, 1.8077e+00, -9.9713e-02, 3.9199e-01,
-3.4435e-01, -8.7162e-01, -5.0063e-01, 3.1002e-02,
-6.8512e-01, 7.0021e-01, 7.4635e-01, 3.7226e-01,
-3.9490e-02, -9.6599e-01, 6.4621e-01, 1.4763e-01,
1.2444e+00, -2.7501e-01, -1.4905e+00, 4.0848e-01,
-1.3125e+00, 3.0412e-01, 8.4391e-01, -5.8470e-01,
-6.8490e-01, 2.6322e-01, 2.2643e+00, -1.1067e+00,
8.3951e-01, -1.2184e+00, 9.7983e-01, 2.7194e+00,
-1.1018e+00, -7.6620e-01, -1.7438e+00, -6.8291e-01,
9.4094e-01, -1.4346e+00, -2.9768e-01, 1.8586e+00,
-9.0226e-01, -1.2493e+00, 8.0037e-01, 8.2669e-01,
-5.9105e-01, 3.9759e-02, 7.5499e-01, 8.6660e-02,
7.4967e-01, -1.4903e-01, 5.5296e-01, 1.5878e+00,
1.2122e+00, -3.5624e-01, -1.5508e+00, 1.7619e+00,
-5.2438e-01, 3.3455e-01, -2.2791e+00, -5.4712e-01,
4.2771e-01, 2.6654e+00, 1.0616e+00, -4.0254e-02,
1.3212e+00, 2.1155e-01, 1.1903e+00, 7.4973e-01,
2.6022e-01, 2.2801e+00, -2.4106e-02, 3.5802e-01,
-3.5074e-01, -8.7930e-01, -1.0800e-01, 1.3316e-01,
-8.8432e-02, 7.5543e-01, 5.8688e-02, 6.8472e-02,
1.1337e+00, 5.3228e-01, -8.1816e-01, 7.0191e-02,
1.3373e-01, 4.1116e-01, 2.2647e+00, 1.0910e+00,
7.5338e-01, -1.6459e+00, -1.4604e+00, -9.6625e-01,
3.6223e-01, 1.2401e+00, 7.6158e-01, 9.3188e-01,
7.1686e-01, 1.7391e-01, 5.2883e-01, 1.2099e+00,
5.9034e-02, 1.3963e+00, 1.0756e+00, -9.7138e-02,
5.0201e-01, -8.3973e-01, 2.7205e-01, 2.6536e-01,
-1.7811e+00, 1.1338e+00, 1.7738e-01, 5.7491e-01,
-4.5298e-01, -1.3104e+00, -4.3938e-02, -1.5059e+00,
-5.3613e-01, -1.4593e+00, 5.9351e-02, -4.1981e-01,
9.6031e-01, 1.1638e+00, -3.0889e-01, -7.0186e-01,
-5.5037e-02, 6.1385e-01, 1.8960e+00, -4.7311e-02,
-3.6615e-02, 1.1164e+00, -1.0588e+00, 1.3009e+00,
-5.3125e-01, 6.8888e-01, 4.7661e-01, 7.1533e-01,
-1.8427e-01, 1.7306e+00, 4.7338e-01, -2.6252e-01,
-2.7663e-01, 1.5924e-01, -1.3103e+00, -1.5974e-01,
-9.1565e-01, 8.7656e-01, 9.5346e-01, -7.4285e-01,
5.6968e-01, 3.0531e-01, 1.5388e-02, 1.7168e+00,
9.1235e-01, -1.6179e+00, -1.6084e+00, -1.8412e+00,
4.7027e-04, 1.0582e-01, 2.7876e-01, 1.2288e+00,
-2.8224e+00, -6.0475e-01, 1.6841e-01, 5.3345e-01,
8.2495e-01, 1.3362e-01, 2.6938e-01, -9.8203e-01,
4.2429e-01, 9.7438e-01, -5.9620e-01, 8.0944e-01,
-7.1189e-01, -7.2273e-01, -1.7291e+00, -9.7576e-01,
-2.6846e-01, 2.2400e+00, 2.1584e+00, 1.3572e-01,
1.2866e+00, 1.2151e+00, -1.0459e+00, 3.9743e-01,
3.4964e-01, -1.8948e+00, -1.2763e+00, -1.5092e+00,
-2.5270e-01, -1.5031e+00, 6.0375e-01, 7.6886e-01,
-1.2586e+00, -2.3434e-01, 1.4694e+00, 1.9085e-01,
4.9615e-01, -9.7999e-01, 5.7358e-01, -7.6271e-01,
6.9127e-01, -2.3102e+00, 1.2069e+00, 3.4485e-01,
6.3022e-01, 3.1535e-01, 2.4482e-01, 6.8610e-01,
9.3086e-01, -1.1471e+00, -7.4297e-01, -3.2452e-01,
9.7733e-01, 6.8480e-01, 1.1135e+00, 2.8077e-01,
-5.2943e-02, -5.8024e-01, 2.5193e+00, 5.0969e-01,
9.0888e-01, -1.2671e+00, 1.6900e+00, 3.8673e-02,
-2.3817e+00, -1.1011e+00, -1.5120e-01, 6.5088e-01,
4.6697e-01, -1.0980e-01, 4.8098e-01, -7.2025e-01,
-2.1700e-01, 5.5756e-01, 8.0716e-01, -1.7352e-01,
-2.6723e-01, 7.0377e-02, -8.1685e-01, 9.8720e-01,
-4.3314e-01, -3.6320e-01, -1.3450e+00, 2.2864e+00,
-2.2796e+00, 2.1142e+00, 1.5656e+00, 6.6128e-01,
1.8285e+00, -9.0319e-02, 1.0089e+00, 8.2049e-02,
-1.1231e+00, 1.6845e+00, -7.6854e-01, -7.6907e-01,
-1.7463e+00, -1.5006e+00, 4.0063e-01, 1.1385e+00,
-7.5925e-01, 5.4816e-01, 7.0085e-01, 6.0949e-01,
2.9928e-01, 9.3619e-01, 1.1161e+00, -3.1338e-01,
1.1210e+00, 1.0418e+00, 2.8270e-01, -5.6078e-01,
-1.2497e+00, -2.6737e-01, 7.0499e-01, -5.6402e-01,
-1.6575e+00, 7.5868e-01, -9.5563e-01, 1.4457e+00,
1.2540e+00, 8.3622e-01, -1.4116e+00, 4.9375e-03,
-9.7158e-01, -9.6983e-01, -1.8226e-01, -6.8004e-01,
2.2828e-01, 1.6887e+00, -3.9312e-01, -3.8751e-01,
-5.6816e-01, -2.7335e+00, -1.2852e+00, -1.1903e-02,
4.1618e-01, 2.7233e-01, 1.4742e+00, -1.8005e-01,
-9.9802e-01, -9.4651e-01, 8.1005e-02, 2.4586e+00,
-6.2108e-03, 4.1839e-01, 6.8686e-01, -8.5606e-01,
-1.1499e+00, 1.4359e+00, 9.0809e-01, 1.6598e+00,
-5.8484e-01, -1.8847e+00, 1.9313e-01, 9.1345e-02,
6.4720e-02, -2.0231e+00, 6.0292e-01, 1.8061e+00,
2.7635e-01, -6.2428e-01, -6.7076e-01, -1.1309e+00,
3.9497e-01, 6.0852e-01, 1.1087e+00, 4.0360e-01,
-1.6841e+00, -4.6096e-02, -4.9633e-01, -6.8442e-01,
-2.5152e-01, 2.1814e-01, -6.2868e-01, -8.2658e-01,
-1.2749e+00, 1.6283e+00, 2.1840e+00, -2.6027e-01,
-3.2740e+00, 4.7177e-01, -7.7335e-01, 1.9163e+00,
-8.2562e-01, 1.1457e+00, 9.4549e-01, -1.1589e+00,
-2.1242e+00, -1.1101e+00, 3.8247e-01, 7.1099e-01,
-4.7793e-01, -1.7350e+00, -1.0199e+00, 4.4650e-01,
2.5518e-01, -6.2817e-01, -8.8052e-01, 1.7682e-01,
7.5055e-01, -1.1667e+00, 7.9410e-01, 2.3305e+00,
-2.4745e+00, 1.1300e+00, -1.5619e+00, -1.5633e+00,
9.9999e-01, 4.3773e-01, -4.4796e-01, 3.8636e-01,
1.3738e+00, -1.1670e+00, -1.7768e+00, 7.3444e-01,
-4.6160e-01, -5.7544e-01, 2.1528e+00, 1.2688e-01,
-6.4337e-01, -2.1737e-01, -1.3143e+00, 7.8221e-01,
-2.3779e-01, 2.7542e+00, -8.2604e-02, -9.5540e-01,
-2.2206e+00, 8.3327e-01, -1.7159e+00, -3.6136e-01,
-6.8905e-01, -1.7827e-01, 1.0750e+00, -5.7502e-01,
-6.8854e-01, -6.6768e-01, -1.1729e-01, -1.7126e+00,
2.0674e-01, -1.5499e+00, 2.5799e+00, -2.7295e-01,
-4.9345e-02, -4.6562e-01, -5.3387e-01, -9.1103e-01,
1.9725e-01, 3.3585e-01, 4.4260e-01, -7.0747e-01,
2.3438e-01, -7.3938e-02, -1.2846e+00, -1.1080e+00,
3.4382e-01, -1.2340e+00, -1.6231e-01, 2.0784e-01,
-1.2283e-01, -6.9486e-01, -1.4514e-01, -3.4300e-01,
-6.4899e-01, 8.6680e-01, 1.0230e+00, 1.2419e+00,
-1.1125e+00, 1.4562e-01, -3.2847e-02, -6.8413e-01,
-9.6430e-01, -8.0016e-01, 3.6284e-01, 1.0795e+00,
2.1395e-01, 2.7607e-02, -1.4464e+00, -9.7554e-01,
-2.7388e-01, -6.5513e-01, 8.2153e-03, -6.0212e-02,
-8.4249e-01, -1.3816e+00, 5.7762e-01, 5.6074e-01,
4.5861e-02, 1.6568e+00, -5.8589e-01, 3.3030e-01,
5.4547e-01, -6.3554e-01, 6.0818e-01, 2.9506e+00,
-7.8257e-01, 3.7028e-01, -2.1630e+00, -1.6947e-01,
2.1183e-01, -6.5975e-01, 2.6241e-02, 2.3716e-01,
-1.2279e+00, -5.9922e-01, -4.8802e-01, -4.0882e-01,
3.7267e-01, 7.5337e-01, 1.6799e+00, 1.2138e+00,
3.0877e+00, -7.6341e-01, -6.6775e-01, -4.3837e-01,
3.4017e-01, -8.0369e-01, 9.6854e-01, 1.5868e-01,
-2.1719e+00, -8.2639e-01, -1.4397e+00, -5.4163e-01,
-2.9705e-01, -1.3685e+00, 6.2714e-02, -1.2076e+00,
1.4155e+00, -6.0217e-01, -3.1748e-01, 7.3557e-01,
3.7363e-01, 6.8654e-01, 3.8576e-01, -1.0353e+00,
5.9497e-01, 1.2630e-01, 3.3457e-01, -3.9127e-01,
1.2481e+00, -9.5252e-01, 9.9410e-01, -1.2564e+00,
-1.1452e+00, 8.7360e-01, -1.2440e+00, 6.0314e-01,
7.5591e-01, -1.0538e+00, 1.6165e+00, -1.4793e+00,
2.7234e-01, -4.8293e-01, -4.9334e-01, 2.9590e-01,
-7.8710e-01, 4.8719e-01, -7.4265e-01, 4.2771e-01,
-4.3596e-01, 1.2698e+00, -1.8515e-01, 1.4344e+00,
4.3506e-01, -4.0930e-01, 1.3369e+00, -7.8129e-01,
-2.2363e+00, 2.1702e+00, -1.9489e-01, 6.7600e-01,
1.0816e+00, 5.2764e-01, -3.8781e-01, -5.5715e-01,
8.5815e-01, 2.1119e+00, -7.9055e-02, -4.9473e-01,
-5.4395e-01, 6.7140e-01, 6.6374e-01, -4.0398e-01,
2.9237e-01, -2.2886e+00, -1.4391e+00, 1.0957e+00,
-7.1013e-01, 9.1007e-01, 1.6630e+00, 5.4100e-01,
-8.7341e-01, -9.8547e-01, 1.7926e-01, -4.2107e-01,
1.8108e+00, 4.4699e-01, 3.1376e-01, -1.3357e+00,
4.1555e-01, -2.1260e-01, 1.4506e+00, 5.6962e-01,
-4.2664e-01, 5.2509e-01, -4.1844e-01, 8.8892e-01,
-2.4534e-01, -9.1272e-01, -1.2163e+00, 4.0829e-01,
-7.0500e-01, -2.9138e-01, 9.3320e-01, -5.9302e-01,
1.2204e+00, -7.0160e-01, -7.1310e-01, -1.2396e-01,
-4.6402e-01, -1.0622e+00, 8.6719e-02, -2.3789e+00,
1.6143e+00, 1.0193e+00, -5.8930e-01, 6.8309e-01,
-1.4454e+00, 6.3099e-01, -1.3856e+00, 1.1903e+00,
-1.1635e-01, -9.0947e-01, -6.7883e-02, -9.2518e-02,
-6.5952e-01, 2.2510e+00, 2.1039e-01, -1.8113e+00,
-1.1105e-01, -1.6224e-01, -6.2111e-01, -7.4668e-01,
-5.7914e-01, 2.1908e+00, -1.9162e-01, 1.5146e+00,
-9.7142e-01, 6.3919e-01, 3.3003e-02, -8.4065e-01,
1.8641e+00, -1.0655e-01, -7.6993e-01, -1.3961e-01,
-1.8983e-02, 3.2670e-01, -2.7699e-01, 1.1933e+00,
-6.5785e-01, -8.9919e-01, 5.4369e-01, -2.7476e-01,
-3.1674e-01, -1.3948e-01, 2.4452e-01, -9.0463e-02,
-6.0247e-01, -1.4504e+00, -1.0806e+00, -3.0926e-01,
-2.3260e-01, -1.0267e+00, 1.0110e+00, -1.0182e+00,
8.1562e-02, 8.4199e-01, -1.0392e+00, -1.1473e-01,
3.4182e-01, -2.0177e+00, 5.3359e-01, 9.2851e-02,
4.5460e-01, 3.0447e-01, 2.0859e-01, 1.4701e+00,
8.8098e-01, -5.7983e-01, -2.7375e-01, 1.3919e+00,
-9.9096e-01, -8.0167e-01, -1.5783e+00, 1.5359e+00,
-8.0353e-02, 7.2024e-01, -4.1623e-01, -6.3921e-01,
-9.1456e-01, -2.0109e-01, -1.9769e+00, 1.5985e+00,
-9.4449e-01, 6.0733e-01, 6.3369e-01, 4.8258e-01,
-2.2019e-01, -1.5574e+00, -1.9108e+00, 2.9156e-03,
-1.4098e-01, -6.5820e-01, 2.8573e+00, 6.7923e-01,
8.3304e-01, -5.8909e-01, -3.9312e-02, -1.8867e+00,
-1.4274e+00, -1.0916e+00, 1.0863e+00, 7.2159e-01,
-4.0775e-01, -3.2791e-01, 8.6075e-01, -1.0552e+00,
-3.3951e-02, 3.6099e-01, -1.5847e+00, 1.1447e+00,
-1.8201e+00, -1.4713e+00, -1.2462e-01, 3.6974e-01,
9.9296e-01, -1.5389e+00, 1.0387e+00, 5.0785e-01,
-7.4961e-01, 1.0517e+00, 4.4070e-01, -1.3172e+00,
-7.1100e-01, 2.1449e-01, 5.5716e-01, -5.8195e-01,
-9.6947e-01, -5.4557e-01, -1.0905e+00, 1.4218e+00,
5.2130e-01, 3.3826e-01, -2.9048e-01, 1.4046e+00,
5.4845e-01, 5.1258e-01, -1.0342e-01, 9.2472e-01,
7.8219e-01, -7.8439e-01, -3.7683e-01, 4.1409e-01,
-6.8138e-01, -4.3957e-01, 9.8066e-01, -2.3513e-01,
-2.6894e-01, 4.7734e-02, -9.2465e-01, 5.4271e-01,
-6.1918e-01, 8.4573e-01, 5.4417e-01, -4.0264e-01,
-1.0216e+00, -1.3087e+00, 9.0422e-01, -8.2181e-01,
-2.7152e-01, 3.9739e-01, 1.6134e+00, -9.4250e-01,
1.5042e+00, 4.0509e-01, -1.5148e+00, -2.1177e-01,
-2.5565e+00, 7.3811e-02, 3.3827e-01, 1.8427e+00,
1.6709e+00, 2.0232e+00, -3.2041e-01, 1.5436e+00,
4.6471e-01, 2.0516e-01, -1.8678e+00, 5.4434e-02,
-1.2665e+00, -5.2681e-01, -1.1934e+00, -1.8230e+00,
9.1128e-01, -2.0682e+00, 1.0792e+00, 3.4336e-02,
-1.4213e-01, -3.0310e-01, -1.2023e+00, -1.7177e+00,
5.5921e-01, -6.6501e-01, -1.4616e+00, -6.0379e-01,
3.2442e-01, -3.0002e-01, -5.6025e-01, 2.7723e-02,
5.6153e-02, 3.4777e-01, 1.1087e-01, -9.5724e-01,
-4.9099e-01, -1.1327e+00, 7.1617e-01, -4.3637e-01,
-1.8701e+00, -3.2514e-01, 1.0974e-01, -6.3546e-01,
4.1184e-01, 2.9935e-01, -3.9011e-02, 1.0232e+00,
1.1854e+00, -1.2406e-01, -1.9112e+00, 1.4239e-02,
5.5008e-03, 5.2655e-01, -4.2288e-02, -1.0899e+00,
-6.0052e-01, -1.4970e+00, 5.2276e-01, 5.7036e-01,
-8.8558e-03, -1.0395e+00, 5.1679e-01, 2.2410e+00,
-1.0719e+00, 1.7373e+00, 1.2079e+00, -9.5935e-01,
1.5390e+00, 1.0305e-01, -5.0341e-01, -5.5391e-02,
-1.0090e+00, -1.2710e+00, -1.1672e+00, 1.1455e+00,
-9.4122e-01, -2.2943e-01, 9.6993e-01, 2.9264e-01,
1.2410e-01, 4.3780e-01, 1.9627e-01, 1.0140e+00,
-1.4596e-01, 5.5276e-01, -4.0080e-01, 1.3926e-01,
2.2296e+00, -4.8268e-02, 2.3331e+00, -4.5636e-01,
2.1482e-01, -1.7862e-01, -6.2706e-01, 3.1592e-01,
1.7821e-01, 1.2309e-01, 3.4700e-01, -9.2821e-02,
-1.1682e+00, -1.0131e+00, 1.0980e+00, -5.1025e-01,
2.3531e+00, -1.9807e-01, 6.2223e-01, 9.7443e-02,
4.3092e-02, -4.9040e-01, -5.0758e-01, -4.2163e-01,
-3.6345e-01, -1.3856e+00, -1.4304e+00, -4.6348e-02,
7.2282e-01, -9.9001e-01, -5.1510e-01, 2.0239e+00,
-7.7100e-01, -9.4390e-01, 9.3503e-01, -1.6447e+00,
6.8471e-01, 1.3631e-01, -3.6638e-01, -1.3261e+00,
-1.1359e+00, 3.6856e-01, 1.9465e+00, 6.7427e-01,
3.3240e-01, -4.2396e-01, -7.4501e-01, -1.1099e+00,
-1.5779e+00, 9.3200e-01, 3.6808e-01, 5.0433e-01,
5.1410e-01, 3.9565e-01, 8.0376e-01, 2.4901e-01,
-3.4567e-01, 8.1075e-01, 9.8055e-01, -7.5226e-01,
1.6926e+00, -2.2407e-01, 1.6191e-02, 5.8853e-01,
4.6703e-01, 1.0399e+00, 1.8867e+00, 2.4651e-01,
-8.7443e-01, 1.0793e+00, -4.4062e-01, 1.1931e+00,
-8.6199e-01, 1.2470e+00, -1.6580e-01, -2.3975e-01,
-6.8277e-01, -1.0126e-01, -9.4668e-02, -4.5427e-01,
8.1837e-01, 4.3734e-01, -4.5235e-01, 6.5808e-01,
-1.1951e+00, 1.7802e+00, -1.1263e+00, 2.2856e-01,
-3.5871e-01, 7.8428e-01, -9.1706e-01, 9.9905e-01,
-1.3901e+00, 4.2979e-01, 1.1237e+00, 7.0063e-02,
-2.0388e-01, -2.5026e+00, 1.1912e+00, 1.0977e+00,
-7.2767e-01, 9.3196e-01, -6.6033e-02, -2.7606e-01,
1.1659e-01, 5.6413e-01, -2.6184e-01, -3.0312e-02,
2.1504e-01, -9.7928e-01, -3.1102e-01, 1.0925e+00,
-9.8866e-01, -1.0573e+00, 2.3829e-01, 1.2873e+00,
-2.2232e-01, 5.6541e-01, 3.0391e-01, 5.6224e-01,
-1.7313e+00, -1.7872e-01, 4.8552e-01, -6.7033e-01,
1.6749e+00, 6.5086e-01, 2.3849e+00, 3.0598e-01]),
size=(10000, 10000), nnz=1000, layout=torch.sparse_csr)
tensor([0.0187, 0.2801, 0.5106, ..., 0.5237, 0.1804, 0.5149])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 1000
Density: 1e-05
Time: 10.794004440307617 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_10000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 431674, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 5000, "MATRIX_DENSITY": 5e-05, "TIME_S": 10.881745100021362, "TIME_S_1KI": 0.025208247659162613, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1032.9713954544068, "W": 97.11, "J_1KI": 2.3929432753754147, "W_1KI": 0.22496142922668497, "W_D": 77.2975, "J_D": 822.2233183002472, "W_D_1KI": 0.17906452554473978, "J_D_1KI": 0.0004148142476608269}

16
pytorch/output_max_core/epyc_7313p_10_2_10_10000_5e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 5000, 5000, 5000]),
col_indices=tensor([7969, 1077, 8574, ..., 4344, 7728, 6479]),
values=tensor([ 1.1970, 0.6292, -1.4825, ..., 0.5053, 0.7511,
1.2540]), size=(10000, 10000), nnz=5000,
layout=torch.sparse_csr)
tensor([0.3153, 0.9159, 0.2730, ..., 0.3296, 0.1411, 0.6731])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 5000
Density: 5e-05
Time: 10.881745100021362 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_20000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 219739, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 39999, "MATRIX_DENSITY": 9.99975e-05, "TIME_S": 10.565528869628906, "TIME_S_1KI": 0.04808217416857684, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1019.1695631408692, "W": 102.72, "J_1KI": 4.638091386330461, "W_1KI": 0.4674636728118359, "W_D": 82.9075, "J_D": 822.593463357687, "W_D_1KI": 0.37729988759391825, "J_D_1KI": 0.0017170365187514198}

16
pytorch/output_max_core/epyc_7313p_10_2_10_20000_0.0001.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 4, ..., 39994, 39996, 39999]),
col_indices=tensor([ 2925, 8680, 14328, ..., 4405, 11796, 13890]),
values=tensor([ 1.5479, -0.9589, -0.3921, ..., -0.2614, -0.0174,
1.4641]), size=(20000, 20000), nnz=39999,
layout=torch.sparse_csr)
tensor([0.2034, 0.4513, 0.8797, ..., 0.2170, 0.9856, 0.8671])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 39999
Density: 9.99975e-05
Time: 10.565528869628906 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_20000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 361194, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 4000, "MATRIX_DENSITY": 1e-05, "TIME_S": 11.07395601272583, "TIME_S_1KI": 0.03065930223848079, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1069.8790954971314, "W": 98.34, "J_1KI": 2.9620622034062896, "W_1KI": 0.2722636588647652, "W_D": 78.5925, "J_D": 855.0383649873734, "W_D_1KI": 0.21759082376783667, "J_D_1KI": 0.0006024209255077234}

16
pytorch/output_max_core/epyc_7313p_10_2_10_20000_1e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 4000, 4000, 4000]),
col_indices=tensor([ 126, 7865, 4868, ..., 3627, 8985, 3806]),
values=tensor([-0.4233, -1.8593, 1.1855, ..., -0.8652, -1.7564,
-0.6758]), size=(20000, 20000), nnz=4000,
layout=torch.sparse_csr)
tensor([0.6011, 0.6114, 0.7176, ..., 0.1714, 0.6050, 0.3460])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 4000
Density: 1e-05
Time: 11.07395601272583 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_20000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 249877, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 20000, "MATRIX_DENSITY": 5e-05, "TIME_S": 10.674734830856323, "TIME_S_1KI": 0.04271995754253623, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1065.5753146767615, "W": 99.88999999999999, "J_1KI": 4.264399343183892, "W_1KI": 0.399756680286701, "W_D": 79.91624999999999, "J_D": 852.5055885627864, "W_D_1KI": 0.3198223525974779, "J_D_1KI": 0.0012799191306021678}

16
pytorch/output_max_core/epyc_7313p_10_2_10_20000_5e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 19994, 19998, 20000]),
col_indices=tensor([ 4636, 6442, 1488, ..., 13018, 4028, 13752]),
values=tensor([ 0.1049, -0.8678, 1.2934, ..., 0.0596, -2.1283,
-0.0346]), size=(20000, 20000), nnz=20000,
layout=torch.sparse_csr)
tensor([0.8196, 0.5466, 0.8776, ..., 0.8990, 0.2478, 0.3420])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 20000
Density: 5e-05
Time: 10.674734830856323 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_50000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 118279, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 249993, "MATRIX_DENSITY": 9.99972e-05, "TIME_S": 10.438204526901245, "TIME_S_1KI": 0.08825069984444614, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1264.2592313790321, "W": 119.07000000000001, "J_1KI": 10.688788638549802, "W_1KI": 1.0066875776765105, "W_D": 99.29500000000002, "J_D": 1054.2926041805747, "W_D_1KI": 0.8394981357637451, "J_D_1KI": 0.007097609345393055}

17
pytorch/output_max_core/epyc_7313p_10_2_10_50000_0.0001.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 7, 15, ..., 249985, 249988,
249993]),
col_indices=tensor([ 1653, 4741, 5510, ..., 14695, 36652, 38992]),
values=tensor([ 0.8068, 1.5510, -0.1668, ..., -0.5450, 0.7955,
0.4919]), size=(50000, 50000), nnz=249993,
layout=torch.sparse_csr)
tensor([0.9246, 0.1529, 0.0342, ..., 0.0874, 0.9069, 0.9445])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 249993
Density: 9.99972e-05
Time: 10.438204526901245 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_50000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 197149, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 25000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.731060981750488, "TIME_S_1KI": 0.05443122197804954, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1091.9111942291258, "W": 103.59999999999998, "J_1KI": 5.53850739404778, "W_1KI": 0.5254908723858603, "W_D": 83.60624999999997, "J_D": 881.1834004104135, "W_D_1KI": 0.4240764599363932, "J_D_1KI": 0.002151045452608906}

16
pytorch/output_max_core/epyc_7313p_10_2_10_50000_1e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 1, ..., 24997, 24998, 25000]),
col_indices=tensor([26949, 13057, 10329, ..., 34707, 29201, 37428]),
values=tensor([ 0.2360, 0.1898, 0.1500, ..., 0.8914, 0.4213,
-0.5484]), size=(50000, 50000), nnz=25000,
layout=torch.sparse_csr)
tensor([0.1499, 0.5126, 0.3529, ..., 0.5964, 0.6087, 0.6992])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 25000
Density: 1e-05
Time: 10.731060981750488 seconds

1
pytorch/output_max_core/epyc_7313p_10_2_10_50000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 155348, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 124998, "MATRIX_DENSITY": 4.99992e-05, "TIME_S": 11.22078824043274, "TIME_S_1KI": 0.07223001416453857, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 1211.2696959972382, "W": 113.0, "J_1KI": 7.797137368986006, "W_1KI": 0.7273991296959085, "W_D": 93.06875, "J_D": 997.6226240649819, "W_D_1KI": 0.5990984756804077, "J_D_1KI": 0.0038564930071864957}

16
pytorch/output_max_core/epyc_7313p_10_2_10_50000_5e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 6, ..., 124989, 124994,
124998]),
col_indices=tensor([24529, 694, 4562, ..., 43638, 44691, 46909]),
values=tensor([0.6771, 2.3793, 1.5881, ..., 1.3174, 0.4706, 2.2981]),
size=(50000, 50000), nnz=124998, layout=torch.sparse_csr)
tensor([0.0142, 0.9048, 0.7712, ..., 0.9005, 0.3364, 0.2982])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 124998
Density: 4.99992e-05
Time: 11.22078824043274 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_100000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 39596, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 999940, "MATRIX_DENSITY": 9.9994e-05, "TIME_S": 10.394734144210815, "TIME_S_1KI": 0.2625198036218511, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 930.358039624691, "W": 89.69, "J_1KI": 23.49626324943658, "W_1KI": 2.265127790685928, "W_D": 80.13624999999999, "J_D": 831.2565999874472, "W_D_1KI": 2.0238471057682594, "J_D_1KI": 0.0511124130156647}

17
pytorch/output_max_core/xeon_4216_10_2_10_100000_0.0001.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 8, 17, ..., 999925, 999932,
999940]),
col_indices=tensor([ 7625, 39686, 48198, ..., 83333, 88880, 93840]),
values=tensor([-0.3927, 0.0550, -0.6417, ..., 1.8498, -1.3312,
0.0154]), size=(100000, 100000), nnz=999940,
layout=torch.sparse_csr)
tensor([0.3501, 0.8761, 0.7884, ..., 0.1601, 0.5038, 0.3220])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 999940
Density: 9.9994e-05
Time: 10.394734144210815 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_100000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 114937, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 100000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.569062948226929, "TIME_S_1KI": 0.09195527069809487, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 910.5351580810546, "W": 88.88, "J_1KI": 7.92203692528128, "W_1KI": 0.7732931954026988, "W_D": 79.3275, "J_D": 812.6741421318054, "W_D_1KI": 0.690182447775738, "J_D_1KI": 0.006004876130190783}

17
pytorch/output_max_core/xeon_4216_10_2_10_100000_1e-05.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 99999, 100000,
100000]),
col_indices=tensor([87224, 75650, 75610, ..., 15482, 57355, 78029]),
values=tensor([-0.2829, 0.8121, -0.5412, ..., 0.4019, -1.1446,
0.9033]), size=(100000, 100000), nnz=100000,
layout=torch.sparse_csr)
tensor([0.3532, 0.1108, 0.5415, ..., 0.9995, 0.5401, 0.9912])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 100000
Density: 1e-05
Time: 10.569062948226929 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_100000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 68377, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 499996, "MATRIX_DENSITY": 4.99996e-05, "TIME_S": 10.122076511383057, "TIME_S_1KI": 0.14803335202455586, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 895.57832177639, "W": 90.38, "J_1KI": 13.097654500437134, "W_1KI": 1.3217894906181902, "W_D": 68.74625, "J_D": 681.2087984445692, "W_D_1KI": 1.0054002076721704, "J_D_1KI": 0.014703777698234355}

17
pytorch/output_max_core/xeon_4216_10_2_10_100000_5e-05.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 3, 7, ..., 499986, 499994,
499996]),
col_indices=tensor([14914, 62815, 64731, ..., 99079, 38887, 56282]),
values=tensor([ 1.1765, 0.6447, -1.0542, ..., -0.0118, -0.2900,
-0.4401]), size=(100000, 100000), nnz=499996,
layout=torch.sparse_csr)
tensor([0.8771, 0.4718, 0.9944, ..., 0.5160, 0.8764, 0.6956])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 499996
Density: 4.99996e-05
Time: 10.122076511383057 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_10000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 392986, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 10000, "MATRIX_DENSITY": 0.0001, "TIME_S": 10.725168943405151, "TIME_S_1KI": 0.027291478432832597, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 868.2826674485207, "W": 80.73, "J_1KI": 2.209449363205103, "W_1KI": 0.20542716534431255, "W_D": 71.26125, "J_D": 766.4425645449758, "W_D_1KI": 0.18133279557032567, "J_D_1KI": 0.0004614230419667003}

16
pytorch/output_max_core/xeon_4216_10_2_10_10000_0.0001.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 2, ..., 9998, 9999, 10000]),
col_indices=tensor([6878, 8048, 7675, ..., 7567, 8531, 7544]),
values=tensor([-0.7610, 1.1523, 0.6429, ..., 0.9317, 1.0352,
1.9714]), size=(10000, 10000), nnz=10000,
layout=torch.sparse_csr)
tensor([0.5808, 0.7639, 0.0323, ..., 0.7294, 0.8106, 0.4895])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 10000
Density: 0.0001
Time: 10.725168943405151 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_10000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 474527, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 1000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.535431146621704, "TIME_S_1KI": 0.02220196352709478, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 830.747336883545, "W": 79.68, "J_1KI": 1.7506850756301433, "W_1KI": 0.16791457598829992, "W_D": 69.91125000000001, "J_D": 728.8979010504485, "W_D_1KI": 0.147328286904644, "J_D_1KI": 0.0003104739812584827}

375
pytorch/output_max_core/xeon_4216_10_2_10_10000_1e-05.output Normal file
View File

@ -0,0 +1,375 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 0, ..., 1000, 1000, 1000]),
col_indices=tensor([2064, 896, 5767, 8670, 1146, 4125, 9977, 6565, 3378,
9326, 8391, 9599, 4058, 9628, 6143, 440, 6008, 3213,
4853, 6546, 7378, 4351, 3837, 6342, 9332, 8840, 4179,
8003, 9756, 3239, 9195, 5687, 2460, 2064, 4697, 7061,
834, 7416, 7224, 4883, 270, 4001, 3389, 3633, 7885,
9998, 9329, 5526, 6582, 3261, 2015, 8547, 3317, 7933,
1303, 4867, 887, 5151, 5655, 7755, 2233, 7436, 4980,
3353, 9810, 6729, 4256, 4173, 8673, 2685, 55, 5842,
5503, 4989, 6590, 6771, 7338, 8091, 3697, 2563, 2311,
2320, 5217, 5949, 192, 9515, 6139, 9365, 3951, 7306,
2274, 3228, 8616, 7376, 7025, 9484, 4041, 8865, 4783,
1059, 2836, 7745, 4418, 3686, 2493, 2195, 8137, 9779,
7315, 625, 8915, 2898, 9453, 7590, 2597, 4170, 1951,
1134, 6168, 6165, 5456, 198, 3173, 9831, 2754, 4065,
4761, 9241, 2335, 3564, 3845, 4906, 7808, 225, 3126,
3229, 8521, 9071, 1528, 3879, 1771, 4180, 9504, 8533,
6474, 7859, 2797, 2073, 3924, 3030, 3526, 8411, 4490,
8374, 4377, 1788, 4542, 9737, 8798, 8688, 1154, 5275,
2718, 3351, 1972, 1577, 2305, 481, 1168, 8903, 463,
134, 3291, 3927, 5072, 5161, 3859, 5451, 4500, 9910,
7791, 1695, 4418, 8117, 8207, 9881, 8085, 5856, 8691,
7144, 7141, 1500, 5060, 1969, 3148, 2606, 152, 325,
1124, 3092, 5148, 923, 4192, 5152, 7899, 8646, 3495,
8002, 5360, 8608, 1283, 9085, 6831, 6484, 8761, 1589,
3359, 3054, 696, 5677, 3844, 186, 5671, 174, 3019,
8470, 677, 5321, 8264, 4749, 4597, 6363, 9352, 5518,
8210, 1984, 6888, 8632, 2559, 3656, 4214, 3332, 6177,
6882, 3795, 7287, 2018, 7045, 3634, 9716, 9950, 9071,
8732, 3869, 1167, 8382, 3313, 82, 6942, 9361, 4340,
8126, 4549, 3926, 2516, 4706, 7348, 5915, 1336, 9470,
6307, 9110, 6080, 748, 7147, 4636, 6326, 5033, 9230,
2898, 8177, 9212, 6264, 6030, 2722, 6849, 1654, 386,
9283, 3041, 6583, 3017, 5716, 117, 9716, 8835, 9770,
8157, 416, 9759, 3293, 4942, 2025, 6656, 2593, 8406,
936, 4397, 2728, 7013, 6306, 5793, 8090, 6303, 4590,
8937, 9415, 304, 8330, 1071, 5382, 9011, 1218, 1209,
7547, 1022, 2561, 972, 5568, 3231, 3597, 9750, 567,
4660, 9183, 86, 87, 8247, 7273, 2704, 9075, 7187,
9352, 6506, 72, 8611, 7811, 8443, 9657, 3767, 3235,
8577, 4086, 16, 8483, 7745, 8982, 6294, 312, 2187,
9832, 8039, 4525, 4172, 895, 1526, 4293, 8571, 2480,
5868, 8652, 4370, 1634, 4315, 2767, 1090, 4904, 2363,
5693, 3722, 2888, 6538, 1766, 8330, 1449, 950, 1372,
2737, 6498, 8332, 3089, 7008, 4584, 9336, 9726, 5065,
4607, 2000, 7784, 5090, 7773, 7101, 115, 3571, 3680,
6901, 8089, 8449, 4622, 7783, 1822, 3950, 8619, 7459,
4621, 3825, 1141, 6518, 8781, 5615, 3873, 8811, 5039,
3599, 7050, 3812, 5371, 6907, 3094, 2729, 7651, 5282,
1831, 1807, 704, 4703, 7350, 1108, 5823, 2825, 5269,
6808, 827, 6831, 6679, 329, 8701, 228, 9776, 608,
2767, 1099, 3855, 959, 3908, 6244, 1086, 3323, 5008,
9675, 444, 1343, 3375, 6111, 4474, 1791, 2253, 3115,
7074, 7289, 5612, 9637, 3530, 3029, 3731, 5380, 6945,
9011, 6048, 8782, 3458, 1615, 5799, 9033, 2448, 7920,
5226, 2000, 1435, 2487, 206, 9197, 679, 7688, 3341,
2288, 5947, 8998, 8424, 8786, 5546, 9630, 3723, 4066,
7682, 178, 601, 1048, 1928, 2494, 5153, 5423, 7378,
1947, 8604, 5223, 8914, 7071, 4417, 7855, 5259, 76,
4045, 4129, 6232, 1970, 2495, 9058, 252, 8853, 2862,
6344, 5111, 7004, 3216, 7874, 9545, 9955, 8347, 8827,
7677, 9945, 1576, 6437, 8984, 429, 7260, 4469, 4182,
2000, 2724, 6546, 6602, 6254, 4196, 707, 9234, 8976,
6368, 5915, 6428, 1200, 9801, 114, 8583, 4852, 7375,
4020, 4171, 4473, 2727, 9996, 5707, 1656, 2207, 2636,
7612, 828, 2375, 9479, 8274, 8858, 7786, 6463, 3229,
4066, 3193, 602, 1105, 4380, 2194, 6685, 2638, 1304,
1549, 4182, 8719, 7761, 3721, 4979, 4458, 1691, 6124,
9800, 2853, 103, 6882, 4441, 5970, 4948, 8246, 2799,
2996, 4383, 3163, 2476, 9375, 9069, 9941, 3422, 7187,
6452, 6977, 5417, 2024, 2635, 9082, 5499, 1852, 9777,
1674, 8698, 4110, 1476, 9983, 7913, 2138, 6281, 8764,
8927, 8149, 4133, 4457, 1384, 1380, 8195, 6569, 4401,
8001, 5877, 164, 8456, 835, 1547, 7686, 7689, 4377,
7439, 6753, 1783, 659, 9616, 1690, 6517, 5549, 4412,
3204, 2642, 9748, 6609, 7333, 7493, 100, 6704, 777,
5989, 4031, 5779, 4282, 6464, 3615, 6082, 3810, 3202,
4262, 7116, 9918, 904, 9054, 5223, 3791, 9999, 5265,
8337, 458, 2617, 8988, 6230, 3093, 2226, 3754, 2801,
2805, 5540, 8035, 4084, 5548, 8434, 5978, 1686, 4265,
4741, 1157, 2332, 37, 3027, 2807, 6481, 9870, 3202,
5792, 8718, 7810, 4352, 9049, 3010, 4953, 7273, 7458,
7740, 2103, 3512, 674, 770, 255, 9659, 2536, 2145,
8111, 4044, 3755, 725, 5620, 5744, 1996, 4931, 6471,
909, 3092, 1067, 3661, 9403, 1758, 9399, 2183, 476,
9173, 6595, 7168, 2227, 9224, 7673, 7552, 8652, 1989,
5717, 3259, 1802, 3415, 8837, 9072, 4773, 6026, 4552,
4317, 5882, 1386, 3443, 8094, 7409, 9971, 2207, 303,
9540, 6713, 7156, 9717, 5027, 835, 5289, 9195, 2378,
3998, 6789, 3179, 6311, 5133, 4951, 4633, 2802, 7799,
3487, 3809, 7357, 5858, 7473, 5429, 5581, 4950, 2570,
5998, 8269, 2311, 4457, 295, 6305, 6664, 3115, 8134,
1839, 5204, 1763, 8028, 6046, 172, 6607, 3103, 5951,
8126, 6101, 5488, 8930, 1839, 4608, 9796, 2697, 1799,
409, 5024, 9633, 6700, 7750, 5007, 9587, 1380, 1289,
4541, 7399, 7574, 7815, 7520, 8836, 2650, 2189, 7536,
7441, 9664, 2911, 3858, 3684, 5146, 941, 4379, 2064,
5983, 2476, 8663, 5292, 2448, 5998, 5389, 5160, 7551,
7291, 1891, 6539, 1394, 4338, 8398, 8434, 5536, 2314,
4692, 2358, 3892, 2382, 4459, 4191, 1073, 8666, 3741,
5285, 9232, 6519, 4596, 5872, 6831, 5271, 6354, 2322,
8655, 8453, 8623, 525, 2599, 5547, 578, 5158, 2102,
2812, 883, 4262, 6080, 1022, 1191, 3005, 9295, 2954,
4285, 2736, 6347, 2364, 7360, 9969, 632, 5653, 5020,
2911, 7012, 5166, 1335, 8815, 5645, 3553, 2299, 4631,
2213, 5413, 8269, 5682, 6797, 7402, 2569, 7872, 8640,
8335, 4609, 8735, 7876, 1460, 7805, 5772, 4685, 8428,
5887, 6774, 2017, 8311, 1892, 7927, 2540, 6370, 8054,
2212, 2197, 435, 1774, 8849, 6930, 7269, 2056, 67,
5416, 3162, 7657, 7954, 9827, 1631, 7037, 4965, 5541,
7188, 4608, 5800, 569, 6708, 7090, 7671, 2087, 9022,
5827, 7712, 1579, 4214, 6582, 3285, 3930, 4932, 8773,
8018]),
values=tensor([ 5.1176e-03, 6.2289e-01, -5.9111e-01, -2.5507e-01,
-1.4720e+00, 2.1104e+00, 2.7731e-01, -8.1725e-01,
-1.0859e+00, -5.9596e-01, -9.4164e-01, 8.6929e-02,
-2.0503e+00, -2.0446e-01, -3.4785e-01, -7.4825e-01,
1.0931e+00, 9.0092e-01, -1.8690e+00, -2.4736e-01,
-2.2635e+00, 1.3374e+00, -7.4686e-01, 1.0888e+00,
8.7548e-01, -1.3074e+00, 1.3396e+00, 4.0301e-01,
-2.0779e+00, 5.5662e-01, -6.4851e-02, -9.8291e-01,
-2.9455e-01, 1.1513e-01, 8.7757e-01, -3.0308e-01,
4.5232e-01, -6.4838e-01, 2.5072e-01, -5.0454e-02,
4.3009e-01, 7.8334e-01, 2.0967e+00, -1.3320e+00,
7.9890e-01, 1.5557e+00, 1.0932e+00, -1.1216e+00,
1.2690e+00, -2.8509e-01, -6.4031e-01, -3.2623e-01,
-1.4615e-01, 6.4304e-01, -1.3330e+00, 7.7970e-01,
-1.2297e+00, 6.2389e-01, 1.1891e+00, 1.2133e-01,
1.0769e+00, -2.9181e-01, 6.4164e-02, 4.6208e-01,
-9.7863e-01, 1.6631e+00, 1.9835e+00, 7.1690e-02,
-6.2036e-01, 3.4780e-01, -1.7133e+00, 1.6404e+00,
1.3791e+00, 1.1154e+00, 9.6124e-02, -1.9122e+00,
1.0818e+00, -7.8482e-01, 3.6367e-01, 2.9507e-03,
4.8142e-01, 1.2850e+00, -3.2870e-01, -8.2834e-01,
-3.8943e-01, 1.6432e+00, 1.6087e+00, 6.0348e-01,
8.4750e-01, 4.9759e-01, -6.1120e-01, 1.0870e+00,
8.6964e-01, -2.0507e+00, -5.8551e-01, 1.0480e+00,
5.0482e-01, 5.5354e-01, 1.9551e-01, 8.0672e-02,
6.5498e-01, 2.2414e-01, 7.7403e-01, -1.5414e-01,
1.6763e+00, -8.1938e-01, 1.5216e+00, 9.5708e-01,
2.9318e-01, -1.4612e+00, -4.7755e-01, 2.6082e-02,
4.5652e-01, -2.6663e-01, 6.8555e-02, 4.0440e-01,
-4.1091e-01, 2.2081e+00, 1.7953e+00, -1.0637e+00,
1.0679e+00, 6.5029e-01, -6.5510e-01, -3.0343e-01,
-4.5691e-01, -1.2553e+00, -2.3115e-01, -8.1682e-01,
-7.4068e-01, -1.4591e+00, -9.1966e-01, 2.9885e-01,
-7.5216e-01, 1.2117e-01, -9.0220e-02, 1.7545e+00,
3.7314e-01, 4.5704e-02, 3.6135e-01, 3.6057e-01,
2.3844e-01, 3.5260e-02, 1.8173e+00, 3.5356e-01,
-1.2007e-01, 1.4829e-01, 4.6137e-01, 1.0923e+00,
5.9000e-02, -2.1910e-01, 1.7719e+00, -7.1297e-01,
2.6004e-01, 1.0586e+00, -9.6132e-01, -1.2622e+00,
3.2865e-02, 2.0779e-02, 1.3664e-01, -6.1847e-01,
-1.2978e+00, -2.1403e+00, -1.0474e+00, 1.3773e+00,
1.0813e+00, 4.1608e-01, 2.5728e+00, -5.6230e-01,
-2.3324e+00, 1.3553e+00, -1.1738e-01, 3.3412e-01,
-1.0006e-02, 7.8864e-02, -2.4070e-01, -1.7962e+00,
2.8447e-01, -9.9219e-01, -1.3962e+00, -5.5047e-01,
1.7451e+00, 4.6817e-01, -5.6460e-01, 1.1095e+00,
4.5170e-01, 1.9359e+00, 2.1421e+00, -3.1258e-02,
-6.6312e-01, -1.5769e-01, 1.0456e+00, 5.0859e-01,
-1.3109e-01, 5.9817e-01, -3.2773e-01, 7.3564e-01,
1.0850e+00, 8.9115e-01, 3.3494e-01, 8.0545e-01,
-9.1958e-01, -2.3899e+00, 3.1819e-01, 1.4076e+00,
-6.1097e-01, -1.1595e+00, 3.2595e-01, -1.0156e-01,
5.0663e-01, -1.1824e+00, 6.7624e-01, 1.0416e-01,
3.1769e-01, -1.4271e-01, -8.4177e-01, -3.0395e-01,
-1.5177e+00, 1.5794e-01, -3.9557e-01, 1.3717e-01,
-1.2789e+00, 4.4911e-01, -1.5576e+00, 5.0415e-01,
2.6894e-01, 5.1204e-01, -4.5413e-01, 2.3626e-01,
-1.3894e-01, 1.9825e+00, -8.8944e-02, 1.7209e-01,
-1.9078e-01, 1.4524e+00, 1.5368e+00, 2.1474e+00,
-1.1639e+00, -1.6242e+00, 5.1981e-01, -1.8869e+00,
-1.5769e+00, -1.5433e+00, 1.5690e+00, -2.6449e+00,
1.3426e+00, 2.2369e-01, -1.3951e+00, -1.3783e-01,
-2.1294e-01, 3.4394e-02, 6.4447e-01, -1.3189e+00,
4.9956e-01, 2.5216e+00, -4.0129e-01, -1.1481e+00,
2.2363e+00, 5.5840e-01, 6.6156e-01, -9.1227e-01,
-2.7352e-01, 1.0849e+00, 1.0310e+00, 5.3216e-01,
1.0654e+00, 1.4128e+00, 1.3360e-01, -1.9503e+00,
-1.0174e+00, -4.9270e-01, -6.2186e-01, 4.2811e-01,
8.1613e-01, 6.4204e-01, -1.8830e+00, -4.3700e-01,
1.5779e+00, 1.1972e+00, 5.5414e-01, -6.2922e-01,
-9.4805e-01, 1.8216e+00, 6.9553e-01, 1.4644e-01,
5.0118e-01, -2.2532e+00, -3.0712e-01, -7.4262e-01,
1.5857e+00, -2.9102e-01, -4.7648e-01, 6.9386e-01,
2.3474e-01, -2.2929e-01, 3.0291e-01, -5.6191e-02,
4.6064e-02, -7.8415e-01, 1.3120e+00, 4.4333e-01,
-1.1856e+00, 8.1742e-01, -8.9201e-01, 1.6149e+00,
-1.5586e+00, 4.4039e-01, 2.2504e+00, 2.0967e+00,
-6.0852e-01, -4.2439e-01, -1.0565e-01, -8.6721e-02,
-1.2385e-01, 2.0413e-01, -2.0152e-01, 2.9321e-01,
4.3864e-01, -7.9733e-01, 6.1725e-01, 6.8618e-01,
1.7286e-01, 8.7514e-01, 8.1121e-01, 1.6886e+00,
4.4982e-01, 8.2114e-01, -6.6332e-01, -3.3915e-01,
-1.3392e+00, -1.3559e+00, -3.0024e-01, -8.2625e-01,
-1.1558e-01, -4.1349e-01, -1.1355e-01, 7.4312e-01,
8.7396e-02, 5.5143e-01, -6.3531e-01, 1.0142e+00,
-5.3449e-02, -6.4588e-01, -1.2576e+00, 6.6464e-01,
-6.2769e-01, -2.0827e-01, 1.3067e+00, -7.8261e-01,
-5.3868e-01, 1.0844e+00, 2.6355e-01, 8.3691e-03,
-6.8119e-01, -6.0625e-01, 1.9778e+00, -8.8911e-01,
1.8064e+00, -9.9170e-01, -1.3547e+00, 5.8447e-01,
1.5567e-01, 5.8077e-02, -7.0511e-01, -4.4424e-01,
2.2578e+00, -1.8438e+00, 4.5780e-01, 2.0270e-01,
7.5270e-01, -1.5868e+00, 1.4997e+00, 1.6320e+00,
-4.0540e-03, -1.4310e+00, -5.2528e-01, -1.0270e+00,
-9.9479e-01, -9.6034e-01, 2.6018e-01, 1.3418e-01,
6.0085e-01, 7.3431e-01, 8.2201e-01, -8.6782e-02,
-1.2294e+00, 4.8123e-01, 9.0425e-01, 7.0099e-01,
1.0580e+00, -3.2235e-01, 1.9346e-02, 8.8222e-01,
1.0727e+00, 1.7707e+00, 5.3189e-01, -7.5333e-01,
-1.6764e+00, -8.4388e-01, -1.7006e+00, -2.3754e+00,
4.7701e-01, 9.4191e-01, -2.0101e-02, 9.3544e-02,
1.1411e+00, -1.8846e+00, 5.3039e-01, 1.6656e-01,
-8.8872e-01, -7.6875e-01, -6.5077e-01, 1.0716e+00,
-1.4110e+00, -1.4232e+00, -5.7283e-01, 7.5621e-01,
3.0423e-03, 1.4533e+00, -1.9396e+00, -4.1976e-02,
-5.7776e-01, -3.3142e-01, 1.7215e-02, 1.5328e+00,
-8.9899e-01, 6.2952e-01, 1.1086e+00, 3.0312e-01,
-2.9763e-01, 5.2234e-01, 1.2894e-01, 8.6425e-02,
-4.6012e-01, -5.5764e-01, -2.1423e+00, 5.6418e-01,
2.2690e-02, 9.3365e-01, -5.6453e-01, 1.9111e-02,
-1.1241e+00, -9.8589e-01, 2.5865e-01, -5.6524e-01,
-1.9172e+00, -1.7404e+00, 9.7353e-01, 1.8232e+00,
1.7236e+00, -5.9302e-01, 1.0921e+00, -8.3600e-01,
-3.6960e-01, 1.5560e-01, -7.8885e-01, -6.2177e-01,
5.1471e-01, -1.5484e+00, 1.0792e+00, 6.9489e-02,
4.6114e-01, -1.3740e+00, 1.3734e-01, -8.4312e-02,
4.9399e-01, -9.5972e-04, -1.4442e-01, -2.7174e-01,
5.6055e-01, 1.6019e+00, -1.5295e+00, -2.6186e-01,
6.1166e-01, 1.5279e-02, 5.0534e-01, -1.3195e-01,
-1.2799e+00, -2.4007e-01, 5.5499e-01, -3.8357e-01,
7.1599e-02, -2.8463e-01, 8.5677e-01, -1.3222e+00,
1.9603e+00, -1.2649e+00, 7.9247e-01, 6.0402e-01,
-9.9299e-01, -6.0136e-01, 7.2921e-01, -1.8042e+00,
6.1274e-01, 2.7469e-01, 1.4178e-01, -1.2825e-01,
1.1177e+00, 5.8273e-01, 7.8379e-03, 1.0223e+00,
1.2932e+00, -1.0284e-01, -4.6627e-01, -2.2479e-01,
-8.1058e-01, -8.8612e-01, -8.6120e-02, 5.8665e-01,
6.4738e-01, -3.7189e-01, 4.0850e-01, -1.2885e+00,
1.2335e+00, 8.7128e-01, -1.1833e-01, 7.8872e-01,
4.5309e-01, 1.1389e+00, -1.8056e-01, -2.9537e-04,
-1.2649e+00, -9.6021e-01, -5.8812e-01, -1.1100e+00,
-3.2434e-03, -5.6212e-01, 2.4569e-01, -5.6202e-01,
-3.6063e-01, -8.4566e-02, -3.2929e-02, 7.0057e-01,
2.5338e-01, -5.9464e-02, -7.9073e-01, 1.2599e+00,
-1.2089e-01, -9.6083e-02, 8.6271e-01, -1.4373e+00,
6.1328e-01, -1.2486e+00, -1.4293e-01, 2.3740e+00,
-1.4389e+00, 1.3352e+00, 9.5627e-01, -1.1649e+00,
6.0582e-01, 5.3372e-01, 2.4707e-01, -1.7497e-01,
1.9852e-01, -2.2340e+00, 6.9369e-01, 4.6373e-01,
4.2469e-02, -7.8337e-01, 1.2421e+00, -7.1923e-01,
8.2962e-01, 1.3541e-02, -1.0611e+00, 1.3721e+00,
1.1379e+00, 7.9544e-02, -1.3090e-01, -8.5308e-01,
-4.1788e-01, -6.9907e-01, -6.2450e-01, 5.2602e-01,
6.7118e-01, 5.2459e-01, -3.4489e-03, 1.0659e+00,
3.7503e-01, 1.7830e-01, -1.4473e+00, -6.2335e-01,
-9.5762e-01, -1.7749e+00, 1.8064e+00, 6.3341e-01,
1.1488e-01, -1.4937e+00, 2.5676e-01, -2.3030e+00,
-6.9894e-01, -1.6161e+00, -2.6768e-01, -3.6864e-01,
-9.3841e-01, 6.2890e-01, -5.6017e-01, -2.1432e-01,
-6.0240e-01, 7.0760e-01, 1.0771e+00, -1.3131e+00,
9.9285e-02, 7.2822e-01, 2.9600e-01, 2.1027e-01,
-1.3769e+00, -6.5128e-01, 3.0721e+00, 4.8480e-01,
7.8544e-01, -7.4187e-01, 7.2858e-02, 1.0210e+00,
1.1748e+00, -7.8627e-01, 3.4766e-01, -1.6835e+00,
-1.2558e-01, -4.5460e-01, -1.9943e-01, -4.7788e-01,
2.8479e+00, 2.4072e-01, 5.8907e-01, 3.0832e-02,
5.6127e-01, -1.0550e+00, -7.5944e-01, -2.5691e-01,
1.0152e+00, -1.7151e+00, 5.1657e-02, -2.2317e+00,
-2.7442e-01, 1.6575e+00, -4.8239e-01, -6.8510e-02,
-2.2088e-01, 9.4249e-01, 4.2983e-01, 1.3687e-01,
2.7978e-01, -1.9037e-02, 7.4878e-01, -4.5084e-01,
-7.3930e-01, -6.5768e-01, -7.7505e-01, 3.0629e-01,
2.3599e+00, -5.9463e-01, -5.7614e-01, 1.6851e+00,
-1.5688e+00, -1.5518e+00, 3.8544e-01, -5.8053e-02,
-6.4847e-01, -1.2411e-01, 1.6862e+00, 2.8125e+00,
1.8134e+00, -2.8095e-01, 1.8266e-02, -1.2539e-01,
1.4937e+00, 1.3322e+00, 9.9373e-01, -1.1804e+00,
-5.3589e-01, -2.0803e+00, 3.2541e-01, 1.1863e+00,
-9.2727e-01, 4.2853e-01, -2.9106e-01, -1.0845e+00,
-6.3716e-02, -1.0573e+00, 1.3986e+00, -8.8973e-01,
-6.9403e-01, 4.6183e-01, -1.1415e+00, 2.0759e-01,
-7.0258e-01, -4.8256e-01, 4.0456e-01, 1.7699e-01,
1.1080e+00, -1.4632e+00, -3.6598e-01, -1.2913e+00,
-4.4764e-01, -9.2151e-01, 9.4327e-01, 4.3638e-01,
3.3328e-01, -5.5500e-01, 1.2042e+00, -4.9143e-01,
-6.7832e-01, 1.3598e+00, -1.3175e-01, -2.6515e-01,
2.4990e+00, -1.6045e+00, 1.0892e+00, -2.8213e-02,
-6.8271e-02, 2.9169e-01, -6.4943e-01, 9.1979e-01,
1.6024e+00, -4.0994e-01, -1.3501e-01, 4.5618e-01,
-1.1601e+00, -2.4095e-01, 4.6394e-01, 3.3332e-01,
-3.4791e-01, 5.9388e-01, 2.3165e-01, -1.2965e+00,
1.3697e+00, 8.2290e-01, -6.8788e-01, -1.1432e+00,
-1.6388e+00, -7.0863e-01, 9.6457e-01, -1.2794e+00,
-1.9228e+00, 2.5467e-01, 6.1895e-01, 6.1878e-01,
-2.3920e-01, 3.2942e-01, -2.8376e-01, 3.8738e-01,
1.4079e+00, 4.8655e-01, 3.4556e-01, 5.1759e-01,
-3.9168e-01, -4.8498e-01, 1.0735e+00, 6.4221e-01,
3.4057e-01, 2.4713e-01, 8.1715e-01, 3.0356e-01,
-4.8601e-01, 9.3761e-01, 2.3024e+00, -2.6290e-01,
-7.1281e-01, 7.0608e-01, 1.2795e+00, 5.6569e-01,
-2.1299e+00, 1.2594e+00, -2.4123e-01, -1.2446e+00,
1.3988e-01, 7.1608e-01, 4.1270e-03, 1.8908e+00,
-1.6568e+00, 4.5292e-01, -2.0565e-01, -1.8981e+00,
-8.3570e-01, -1.8283e+00, -2.2133e+00, -5.8650e-01,
1.7818e+00, -1.1402e-01, -1.2449e+00, 5.9425e-01,
3.6433e-01, -4.0243e-01, -1.3119e+00, -6.8683e-01,
8.0328e-01, 1.9601e-01, -6.6878e-01, -1.2971e+00,
1.3374e+00, -3.2712e-01, -5.4555e-01, -2.2302e+00,
-7.9162e-01, 7.4796e-01, -2.5288e+00, 9.0571e-01,
2.6581e+00, 7.3679e-01, 5.8728e-01, -7.2713e-01,
-4.7163e-01, -1.5709e+00, 1.0270e+00, 1.6614e+00,
4.6532e-01, 2.0610e+00, -6.5204e-01, 7.3615e-02,
1.2622e+00, -2.3829e+00, 9.3167e-01, 1.4601e+00,
5.9831e-01, -1.6535e+00, 5.2549e-01, 2.9321e-01,
1.3734e+00, -5.9871e-01, -7.2233e-01, -5.7314e-01,
1.1992e+00, 5.0744e-01, -3.9399e-01, -9.7251e-01,
1.0566e+00, -7.9516e-01, 1.6912e+00, -2.3620e-01,
-1.0808e+00, -6.4005e-01, 6.4249e-01, -1.0656e+00,
1.7176e-02, 3.3609e-01, -5.9559e-01, 3.7801e-01,
2.1926e+00, -8.2791e-02, -1.7158e+00, -7.4307e-01,
1.7407e+00, 8.8154e-01, 6.0685e-02, 5.9796e-01,
1.0979e+00, 8.3441e-01, 8.2780e-01, -2.5184e-01,
8.1820e-01, 1.5179e+00, -5.0152e-01, 2.9434e+00,
7.2630e-01, 1.3305e+00, -1.8081e-02, 5.8657e-01,
-1.0041e+00, 3.7161e-01, -1.3397e+00, -1.8702e-02,
-9.7588e-02, 1.5715e+00, 1.3799e+00, 5.4326e-01,
-2.0694e-01, 3.9686e-01, 6.1641e-01, 9.5431e-01,
-1.4744e+00, 6.1865e-01, 2.1068e+00, -1.1166e+00,
-9.4844e-01, 8.9873e-01, -2.0723e-01, 1.0102e+00,
2.0038e+00, -8.0166e-01, 8.3775e-01, -2.2144e-01,
1.0481e+00, -8.9134e-01, -2.6371e-01, -1.7186e+00,
-1.9667e+00, 2.9143e-01, -1.5039e+00, 1.5723e+00,
5.4978e-01, -2.3599e+00, -4.4650e-01, 8.5021e-01,
-1.6026e-01, -4.6301e-01, -1.3157e+00, 2.7636e-01,
-1.0129e+00, -1.3535e-01, -5.0656e-01, -9.2027e-01,
1.7650e-01, -4.1543e-01, -1.3801e+00, 3.5565e-01,
7.3571e-01, 2.5822e+00, 4.4472e-01, -1.0066e+00,
6.4677e-02, -2.0261e+00, -1.0811e+00, 4.4377e-01,
1.5140e+00, 1.5536e+00, 6.4479e-01, 7.6791e-01,
1.7671e-01, -9.5334e-01, 2.6979e-02, 1.0421e-01,
5.8800e-01, -1.8904e+00, 1.3622e+00, -2.5713e+00,
1.8117e+00, 1.2724e+00, -7.3623e-01, -5.6986e-01,
-1.5304e+00, -1.5178e-01, -1.5929e-01, -9.6405e-01,
5.2847e-01, 2.5668e-01, -8.9141e-01, -8.4867e-01,
-1.4180e-01, 9.0012e-02, -6.2871e-01, -1.7118e-01,
-1.2428e+00, 4.1028e-01, -1.8423e-01, 8.4266e-01,
9.0175e-01, -6.6421e-01, -8.2770e-02, -1.5526e+00,
3.4764e-01, 1.0629e+00, 1.9759e+00, 1.3888e+00,
-5.1999e-01, -1.1585e+00, -1.4768e-02, -7.0795e-01,
1.0929e+00, 2.2102e-01, -6.7230e-01, 1.6674e-01,
-8.3575e-02, -2.3058e-02, -1.7667e-01, -1.6545e+00,
-3.7191e-01, 2.7835e-01, 1.7232e+00, 2.0275e-01,
8.0939e-02, 5.7524e-01, 5.7820e-01, -7.1905e-01,
-7.6870e-01, -7.7104e-01, -1.8581e+00, 1.0294e+00,
2.4562e-01, 4.2704e-01, -5.8096e-01, 6.9295e-01,
4.6361e-01, -8.9403e-01, 3.4198e-02, 8.8148e-02,
-4.6371e-01, -1.1013e+00, 1.0868e+00, -8.6751e-01,
2.1436e-01, -1.9236e+00, 4.0286e-01, -7.8423e-01,
-1.2506e+00, 1.8513e+00, -4.4562e-01, -8.5675e-02,
-2.0843e-01, -2.4018e-02, -1.3220e+00, 2.8676e-01]),
size=(10000, 10000), nnz=1000, layout=torch.sparse_csr)
tensor([0.2195, 0.7596, 0.6061, ..., 0.0611, 0.4137, 0.7286])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 1000
Density: 1e-05
Time: 10.535431146621704 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_10000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 416897, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 5000, "MATRIX_DENSITY": 5e-05, "TIME_S": 10.193748235702515, "TIME_S_1KI": 0.024451478988101412, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 800.20649715662, "W": 79.97, "J_1KI": 1.9194345297678324, "W_1KI": 0.19182196082005867, "W_D": 70.47874999999999, "J_D": 705.2338834747671, "W_D_1KI": 0.16905554609411916, "J_D_1KI": 0.00040550914517043575}

16
pytorch/output_max_core/xeon_4216_10_2_10_10000_5e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 2, ..., 4998, 5000, 5000]),
col_indices=tensor([ 769, 3843, 1664, ..., 4059, 1971, 2017]),
values=tensor([ 2.1966e+00, -1.0540e+00, -4.9323e-01, ...,
-7.9609e-01, -3.2329e-01, 1.6222e-04]),
size=(10000, 10000), nnz=5000, layout=torch.sparse_csr)
tensor([0.1412, 0.2239, 0.7542, ..., 0.6799, 0.9850, 0.1722])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 5000
Density: 5e-05
Time: 10.193748235702515 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_20000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 255458, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 40000, "MATRIX_DENSITY": 0.0001, "TIME_S": 10.185706377029419, "TIME_S_1KI": 0.039872332739743596, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 872.0504049777984, "W": 82.97, "J_1KI": 3.413674282965491, "W_1KI": 0.32478920213890344, "W_D": 73.40375, "J_D": 771.5050007760525, "W_D_1KI": 0.2873417548090097, "J_D_1KI": 0.00112481016374124}

16
pytorch/output_max_core/xeon_4216_10_2_10_20000_0.0001.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 4, ..., 39994, 39996, 40000]),
col_indices=tensor([ 9602, 2401, 7750, ..., 9001, 11170, 12038]),
values=tensor([ 0.8008, 1.4827, -0.1561, ..., -2.2081, -0.7261,
-0.9781]), size=(20000, 20000), nnz=40000,
layout=torch.sparse_csr)
tensor([0.0634, 0.8251, 0.4475, ..., 0.5026, 0.8965, 0.1391])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 40000
Density: 0.0001
Time: 10.185706377029419 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_20000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 367236, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 4000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.614307403564453, "TIME_S_1KI": 0.028903232263624623, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 884.0254954576492, "W": 81.05, "J_1KI": 2.407240835478137, "W_1KI": 0.2207027633456415, "W_D": 71.43875, "J_D": 779.1940328639746, "W_D_1KI": 0.19453090110991297, "J_D_1KI": 0.0005297163162378224}

16
pytorch/output_max_core/xeon_4216_10_2_10_20000_1e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 0, 1, ..., 4000, 4000, 4000]),
col_indices=tensor([ 7988, 5196, 13588, ..., 5202, 7556, 13647]),
values=tensor([-1.8890, -1.2461, -1.7644, ..., 0.1451, 0.6336,
0.5210]), size=(20000, 20000), nnz=4000,
layout=torch.sparse_csr)
tensor([0.2340, 0.0079, 0.3606, ..., 0.2305, 0.1025, 0.9829])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 4000
Density: 1e-05
Time: 10.614307403564453 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_20000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 282595, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 20000, "MATRIX_DENSITY": 5e-05, "TIME_S": 10.546817541122437, "TIME_S_1KI": 0.037321316870866206, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 883.0263031268121, "W": 82.01, "J_1KI": 3.1247060391260004, "W_1KI": 0.2902032944673473, "W_D": 72.47500000000001, "J_D": 780.3600941240788, "W_D_1KI": 0.2564624285638458, "J_D_1KI": 0.0009075264196600996}

16
pytorch/output_max_core/xeon_4216_10_2_10_20000_5e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 19996, 19997, 20000]),
col_indices=tensor([ 9379, 2781, 3564, ..., 1013, 4414, 17652]),
values=tensor([-0.8565, 0.8965, 1.8252, ..., 1.3859, 0.2437,
-0.6571]), size=(20000, 20000), nnz=20000,
layout=torch.sparse_csr)
tensor([0.3960, 0.8852, 0.5822, ..., 0.9777, 0.0316, 0.1224])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 20000
Density: 5e-05
Time: 10.546817541122437 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_50000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 135614, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 249987, "MATRIX_DENSITY": 9.99948e-05, "TIME_S": 10.49477481842041, "TIME_S_1KI": 0.0773871047120534, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 944.230057759285, "W": 89.98, "J_1KI": 6.962629652980408, "W_1KI": 0.6635008184995649, "W_D": 80.40625, "J_D": 843.7652598544955, "W_D_1KI": 0.5929052310233457, "J_D_1KI": 0.004372006068867121}

17
pytorch/output_max_core/xeon_4216_10_2_10_50000_0.0001.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 5, 9, ..., 249978, 249982,
249987]),
col_indices=tensor([15438, 17468, 32484, ..., 20505, 24131, 38813]),
values=tensor([-1.1174, 0.6528, 0.8028, ..., -1.0629, 0.2029,
0.8951]), size=(50000, 50000), nnz=249987,
layout=torch.sparse_csr)
tensor([0.3092, 0.4067, 0.8954, ..., 0.5715, 0.5196, 0.0128])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 249987
Density: 9.99948e-05
Time: 10.49477481842041 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_50000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 206467, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 25000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.296310186386108, "TIME_S_1KI": 0.0498690356637434, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 813.6328146743774, "W": 80.36, "J_1KI": 3.940740237783168, "W_1KI": 0.3892147413388096, "W_D": 70.7925, "J_D": 716.763327934742, "W_D_1KI": 0.34287561692667595, "J_D_1KI": 0.0016606799969325651}

16
pytorch/output_max_core/xeon_4216_10_2_10_50000_1e-05.output Normal file
View File

@ -0,0 +1,16 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 1, 1, ..., 24998, 25000, 25000]),
col_indices=tensor([46334, 2630, 33725, ..., 2029, 4960, 42876]),
values=tensor([ 0.6590, -0.0165, -0.4990, ..., -1.5928, -0.9899,
0.5757]), size=(50000, 50000), nnz=25000,
layout=torch.sparse_csr)
tensor([0.0756, 0.9491, 0.7501, ..., 0.6406, 0.2224, 0.3754])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 25000
Density: 1e-05
Time: 10.296310186386108 seconds

1
pytorch/output_max_core/xeon_4216_10_2_10_50000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 167249, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 124998, "MATRIX_DENSITY": 4.99992e-05, "TIME_S": 10.643674850463867, "TIME_S_1KI": 0.06363969201886926, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 957.751288819313, "W": 87.27, "J_1KI": 5.726499344207218, "W_1KI": 0.5217968418346298, "W_D": 77.35, "J_D": 848.883490204811, "W_D_1KI": 0.4624840806223056, "J_D_1KI": 0.002765242725650411}

17
pytorch/output_max_core/xeon_4216_10_2_10_50000_5e-05.output Normal file
View File

@ -0,0 +1,17 @@
/nfshomes/vut/ampere_research/pytorch/spmv.py:59: UserWarning: Sparse CSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at ../aten/src/ATen/SparseCsrTensorImpl.cpp:53.)
matrix = matrix.to_sparse_csr().type(torch.float32)
tensor(crow_indices=tensor([ 0, 2, 5, ..., 124995, 124996,
124998]),
col_indices=tensor([ 7605, 45645, 5199, ..., 26894, 25887, 26531]),
values=tensor([-0.3773, -0.1946, 1.1156, ..., 0.6896, 0.4060,
1.4589]), size=(50000, 50000), nnz=124998,
layout=torch.sparse_csr)
tensor([0.3321, 0.0883, 0.2123, ..., 0.2938, 0.7846, 0.1527])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 124998
Density: 4.99992e-05
Time: 10.643674850463867 seconds