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

1core runs

Browse Source
This commit is contained in:
cephi 2024-12-12 21:56:03 -05:00
parent 89d3365313
commit 0cbb446bb5
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_1core/altra_10_2_10_100000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 4591, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 999953, "MATRIX_DENSITY": 9.99953e-05, "TIME_S": 10.515824556350708, "TIME_S_1KI": 2.290530288902354, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 273.044737739563, "W": 26.087881391815785, "J_1KI": 59.473913687554564, "W_1KI": 5.682396295320363, "W_D": 16.547881391815785, "J_D": 173.19581712722783, "W_D_1KI": 3.604417641432321, "J_D_1KI": 0.7851051277352038}

17
pytorch/output_1core/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, 5, 11, ..., 999928, 999936,
999953]),
col_indices=tensor([70868, 74790, 90634, ..., 84755, 88122, 91648]),
values=tensor([ 0.4036, 1.0126, 0.6579, ..., 0.6347, -1.2417,
0.8926]), size=(100000, 100000), nnz=999953,
layout=torch.sparse_csr)
tensor([0.1903, 0.5682, 0.1772, ..., 0.5583, 0.7604, 0.7111])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 999953
Density: 9.99953e-05
Time: 10.515824556350708 seconds

1
pytorch/output_1core/altra_10_2_10_100000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 14108, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 99999, "MATRIX_DENSITY": 9.9999e-06, "TIME_S": 10.44682264328003, "TIME_S_1KI": 0.7404892715679068, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 262.4202471351623, "W": 25.1855074061394, "J_1KI": 18.60081139319268, "W_1KI": 1.7851933233725121, "W_D": 15.715507406139402, "J_D": 163.747637515068, "W_D_1KI": 1.1139429689636662, "J_D_1KI": 0.07895824843802567}

16
pytorch/output_1core/altra_10_2_10_100000_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, 0, 0, ..., 99999, 99999, 99999]),
col_indices=tensor([54792, 14606, 48431, ..., 1584, 20255, 32244]),
values=tensor([ 0.2168, 2.2061, -2.0429, ..., -0.0096, -0.3108,
-0.6421]), size=(100000, 100000), nnz=99999,
layout=torch.sparse_csr)
tensor([0.3823, 0.7563, 0.1745, ..., 0.6521, 0.2295, 0.6307])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 99999
Density: 9.9999e-06
Time: 10.44682264328003 seconds

1
pytorch/output_1core/altra_10_2_10_100000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 7167, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 499988, "MATRIX_DENSITY": 4.99988e-05, "TIME_S": 10.57473087310791, "TIME_S_1KI": 1.4754752160050104, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 266.5949284362793, "W": 25.47428412267614, "J_1KI": 37.19756222077289, "W_1KI": 3.5543859526546866, "W_D": 16.09928412267614, "J_D": 168.48314472317693, "W_D_1KI": 2.2463072586404547, "J_D_1KI": 0.3134236442919568}

17
pytorch/output_1core/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, 5, 11, ..., 499974, 499983,
499988]),
col_indices=tensor([15958, 18715, 52510, ..., 53360, 79192, 95801]),
values=tensor([ 0.1102, -0.4713, 0.0240, ..., 1.5482, 1.0325,
1.6719]), size=(100000, 100000), nnz=499988,
layout=torch.sparse_csr)
tensor([0.0554, 0.4470, 0.9893, ..., 0.6567, 0.6732, 0.5744])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 499988
Density: 4.99988e-05
Time: 10.57473087310791 seconds

1
pytorch/output_1core/altra_10_2_10_10000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 175804, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 10000, "MATRIX_DENSITY": 0.0001, "TIME_S": 10.573810815811157, "TIME_S_1KI": 0.0601454507053944, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 214.99873767852785, "W": 20.46577514143255, "J_1KI": 1.222945653560373, "W_1KI": 0.11641245444604532, "W_D": 10.405775141432551, "J_D": 109.31560151100159, "W_D_1KI": 0.059189638127872805, "J_D_1KI": 0.0003366797008479489}

16
pytorch/output_1core/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, 0, 1, ..., 9998, 9998, 10000]),
col_indices=tensor([5607, 1451, 5668, ..., 3387, 9491, 9524]),
values=tensor([-1.8541, 0.1963, 1.5424, ..., -0.3045, -0.4980,
0.1016]), size=(10000, 10000), nnz=10000,
layout=torch.sparse_csr)
tensor([0.8961, 0.6841, 0.9747, ..., 0.8377, 0.2079, 0.9912])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 10000
Density: 0.0001
Time: 10.573810815811157 seconds

1
pytorch/output_1core/altra_10_2_10_10000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 416649, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 1000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.149910688400269, "TIME_S_1KI": 0.024360818550867202, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 217.88464315414433, "W": 21.251971491655738, "J_1KI": 0.522945316451364, "W_1KI": 0.051006894272290916, "W_D": 11.821971491655738, "J_D": 121.20409821033482, "W_D_1KI": 0.028373934634802287, "J_D_1KI": 6.810033057754197e-05}

375
pytorch/output_1core/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([ 838, 471, 8264, 1933, 3490, 7261, 7284, 6998, 2605,
6095, 4691, 5234, 4206, 6889, 2260, 4619, 4582, 7441,
8584, 5173, 5710, 7393, 3829, 1411, 8599, 9358, 3880,
5288, 7826, 470, 6578, 3784, 8496, 5718, 9937, 7145,
7947, 5701, 2585, 8298, 7820, 7940, 6905, 8342, 1771,
6193, 9824, 2790, 987, 326, 5223, 5114, 7951, 9055,
5808, 4607, 1220, 238, 576, 4862, 2157, 2085, 1520,
8969, 5790, 1872, 6800, 1026, 4390, 1668, 7994, 4809,
2257, 5557, 3183, 537, 4913, 5571, 4112, 1179, 7062,
4471, 6404, 8277, 3053, 1499, 3552, 7119, 4976, 7335,
390, 2533, 2953, 1440, 2039, 6469, 3145, 8189, 224,
3491, 6602, 2839, 6424, 4100, 2799, 1949, 7285, 8640,
5367, 4180, 8857, 9367, 9062, 9855, 7430, 5434, 7498,
752, 4909, 2832, 1999, 8614, 2677, 6371, 2620, 1519,
7898, 3164, 3448, 3738, 1870, 467, 3411, 6254, 3225,
6935, 6233, 7327, 8043, 1622, 4383, 1437, 7840, 3983,
9241, 151, 7297, 2131, 3463, 4915, 1505, 1043, 5859,
6279, 5515, 3377, 9474, 3937, 6724, 2211, 2691, 5698,
1914, 7193, 8184, 8514, 5535, 6526, 2695, 6459, 529,
5116, 341, 6332, 4305, 1374, 1661, 1849, 717, 9722,
2053, 2874, 9674, 722, 5604, 5085, 6273, 1294, 9589,
3198, 184, 4889, 3916, 9799, 2416, 1970, 1680, 2290,
2580, 6721, 1515, 2029, 4611, 4463, 5811, 7641, 1318,
866, 1446, 6593, 7854, 3205, 6316, 9445, 136, 8650,
6752, 3702, 2549, 9903, 4600, 1731, 9731, 2315, 5805,
8623, 9234, 892, 7379, 8731, 3748, 3143, 4288, 7346,
709, 9820, 6003, 7528, 632, 6211, 3677, 538, 1468,
7559, 7506, 5840, 4063, 8206, 3153, 4416, 2290, 2541,
6343, 1796, 1324, 5922, 2008, 3232, 657, 7507, 8558,
9583, 8220, 545, 7500, 9154, 2896, 2819, 5318, 761,
7901, 8015, 2706, 1654, 743, 407, 3066, 5011, 9713,
9160, 2841, 9366, 6323, 6992, 5479, 3012, 4610, 598,
5275, 1609, 2075, 1738, 2516, 5426, 2144, 466, 5653,
3010, 6606, 899, 3575, 4315, 7388, 8261, 6927, 5499,
1695, 8182, 751, 2418, 7054, 7948, 4112, 336, 4058,
3069, 4909, 5209, 4807, 9285, 3360, 9718, 2819, 4245,
4899, 2610, 4884, 8904, 9644, 8026, 9557, 2778, 1881,
1643, 1721, 8448, 8476, 8718, 4673, 4482, 2001, 8029,
254, 4052, 8036, 9038, 5373, 7937, 8974, 9998, 2104,
3453, 3289, 3031, 4384, 2771, 8378, 719, 5773, 980,
819, 120, 2661, 7991, 6988, 5961, 7464, 9375, 4093,
343, 2777, 1575, 9786, 8842, 1295, 1579, 3673, 7846,
1398, 7928, 8497, 5848, 9794, 1974, 104, 6776, 6966,
9934, 231, 9806, 7246, 8321, 4819, 2845, 5304, 4357,
8236, 2295, 5424, 9209, 7782, 9817, 2932, 1538, 2119,
8058, 4107, 8587, 1238, 2732, 6728, 3150, 7824, 5113,
9509, 2105, 8379, 1356, 326, 8651, 6952, 4550, 7630,
1201, 2886, 8155, 6759, 4305, 6399, 7067, 6893, 2270,
2479, 6170, 6559, 7368, 2690, 8454, 2144, 6817, 6418,
457, 9592, 9702, 2386, 8936, 4590, 8245, 9153, 4600,
209, 7062, 2933, 8854, 718, 6094, 6166, 4721, 1821,
6266, 3494, 5336, 1610, 5639, 3582, 3046, 7690, 6011,
7581, 3104, 3918, 8756, 4977, 9583, 2866, 9656, 5506,
9231, 5163, 4546, 3828, 6315, 4257, 5286, 3323, 2938,
3942, 9564, 4667, 2797, 1358, 2635, 5577, 595, 1255,
8671, 2794, 330, 9412, 3559, 4903, 5614, 3554, 748,
6316, 421, 528, 7219, 9970, 5636, 6319, 7308, 3237,
9312, 5858, 3201, 4818, 239, 76, 9282, 4724, 703,
4737, 8273, 4310, 5798, 9261, 7639, 1144, 5709, 9331,
2011, 7517, 5733, 708, 7448, 7928, 7186, 6761, 1852,
7824, 3340, 1554, 5981, 1905, 8003, 2575, 8085, 2988,
5612, 6335, 8501, 6019, 9426, 2712, 6278, 3610, 7452,
427, 6093, 4708, 6677, 9913, 4255, 7185, 9991, 1049,
5202, 881, 5836, 7253, 3392, 9977, 4191, 2018, 4849,
8458, 8678, 5200, 8499, 5327, 6480, 3459, 6761, 9986,
6397, 9099, 5147, 3259, 4454, 2894, 8997, 7050, 2986,
6685, 8675, 4718, 4683, 9403, 8492, 8541, 4716, 6867,
1596, 1383, 4994, 5767, 9040, 5956, 990, 2094, 2831,
9057, 7853, 9270, 4629, 9870, 5521, 5083, 6300, 2194,
5593, 3453, 2009, 446, 1402, 7242, 7107, 6062, 7254,
8486, 6168, 2987, 6278, 257, 7639, 7295, 7000, 2457,
2582, 8342, 6588, 9454, 2212, 4226, 7435, 3877, 9005,
6269, 139, 3059, 5941, 4678, 5026, 3696, 7031, 8779,
3921, 9599, 2383, 7015, 7515, 1217, 7912, 8524, 9179,
5620, 2280, 9073, 6468, 4224, 8194, 7057, 8551, 5030,
7080, 6879, 2091, 4519, 6723, 7999, 4868, 1694, 6273,
6350, 6236, 6959, 9826, 592, 213, 4281, 725, 6627,
6655, 2780, 914, 2658, 7603, 2575, 8499, 1309, 9202,
9921, 8402, 1365, 6110, 1668, 7911, 9288, 1206, 9575,
4406, 1929, 3062, 8156, 6995, 8163, 645, 4720, 4351,
5601, 4550, 9680, 6126, 9985, 8852, 2284, 7696, 8929,
8926, 4030, 8417, 7121, 2913, 7917, 2439, 9926, 339,
8817, 7406, 8787, 7240, 4013, 4911, 1131, 7410, 9105,
5834, 1942, 4689, 3750, 8396, 9533, 1318, 979, 5662,
7758, 3672, 9738, 7184, 9401, 6995, 4388, 3392, 5173,
1736, 685, 4414, 2075, 8432, 1698, 6947, 2115, 4438,
6568, 9811, 6535, 7297, 5329, 7644, 5544, 8575, 9540,
7142, 7664, 3683, 6962, 9342, 3764, 9782, 7664, 9317,
2418, 101, 1090, 4794, 9445, 2918, 7374, 7723, 149,
9450, 2109, 3569, 7564, 9712, 4154, 5601, 9520, 9675,
4414, 2467, 3973, 6697, 6650, 5164, 4768, 283, 6683,
8892, 2924, 7016, 5421, 8302, 5535, 4166, 9605, 6416,
5332, 9150, 3855, 153, 7530, 5787, 3126, 4659, 2183,
4312, 2384, 8600, 5917, 9906, 9770, 6759, 6645, 1613,
5813, 7068, 2231, 9973, 4770, 9999, 7469, 501, 3679,
9882, 5581, 7800, 8826, 4851, 5507, 907, 6635, 3608,
9442, 3810, 7396, 366, 1566, 2725, 8026, 2951, 9378,
9503, 4633, 3616, 8758, 2369, 1637, 9362, 7785, 8005,
5341, 7973, 6824, 5273, 3028, 9927, 7310, 7380, 6306,
7972, 9984, 2622, 3638, 9315, 4536, 7260, 286, 1052,
9442, 8819, 3014, 1790, 514, 900, 740, 328, 865,
6981, 2659, 659, 292, 5355, 8865, 7398, 1667, 5597,
9425, 9717, 9941, 9128, 7200, 5639, 7449, 7148, 2722,
2263, 9360, 8141, 6225, 5422, 1933, 7752, 1920, 286,
4260, 5632, 2661, 8413, 7874, 4209, 5606, 1726, 9154,
948, 709, 7066, 7612, 5991, 9367, 9547, 7071, 5443,
864, 4818, 2436, 1478, 6575, 9374, 7114, 3494, 6878,
5875, 3999, 2645, 9336, 1164, 4840, 464, 8629, 5180,
1528, 4170, 5258, 6431, 4711, 2301, 4037, 1270, 5835,
645, 4315, 8822, 1849, 1484, 8416, 6040, 21, 1250,
7902, 3030, 2112, 271, 4813, 2721, 2561, 156, 4328,
5712]),
values=tensor([ 1.1369e+00, 1.5339e+00, 1.4096e+00, -1.4045e+00,
-7.9136e-01, -5.5989e-01, -2.4266e+00, 1.4632e+00,
1.7832e+00, 1.9926e+00, 2.1189e+00, -3.8447e-01,
9.0111e-03, 1.3569e+00, 7.1095e-01, 6.2696e-02,
1.0649e+00, -1.6543e+00, -8.5607e-01, 1.2649e+00,
-4.8175e-01, 1.2370e+00, 6.6874e-01, 3.7960e-01,
3.4366e+00, 2.9047e-01, 6.8067e-01, -1.0781e+00,
3.6723e-01, 1.1424e+00, -1.2012e+00, 7.6023e-01,
7.8300e-01, 2.0663e-01, 8.4355e-01, 1.8765e+00,
-4.2665e-01, -2.4471e-01, 1.6922e-01, -2.9572e-01,
1.2670e+00, -4.6533e-02, -1.3133e+00, -2.3986e-01,
-7.8042e-01, -1.1267e-01, -5.6011e-01, 3.1524e-01,
-1.2697e+00, -1.3715e+00, 1.0906e-01, 9.5301e-01,
-2.2326e+00, 1.3419e-01, -8.9309e-02, -3.9298e-02,
-1.5041e-01, -7.4336e-01, -1.0359e+00, -1.1712e-02,
8.8678e-01, -7.2629e-01, 4.7687e-01, -4.8069e-01,
-1.5071e+00, 8.5944e-01, 1.3078e+00, 1.0323e+00,
1.6828e+00, -5.2993e-01, 2.7024e-01, -5.4029e-01,
-1.6405e+00, 8.1030e-02, -1.0246e+00, 2.3359e-01,
1.4334e+00, -4.0079e-01, 2.8060e-01, 1.1613e+00,
7.5058e-01, 8.9685e-01, -1.7582e-02, 5.3912e-01,
-8.5417e-01, 8.3129e-01, 1.6617e-01, 2.4709e-01,
1.3890e+00, -4.1124e-02, -5.4919e-01, -2.1670e-01,
-1.2025e+00, 1.3619e+00, -1.3874e+00, 5.5751e-01,
4.4365e-01, 1.5242e-01, 4.5550e-01, 3.7989e-01,
4.8720e-01, -9.6641e-01, -9.0280e-01, -1.7995e-02,
2.3738e+00, -2.3607e-01, 6.6421e-01, -1.2567e-01,
1.3051e+00, -7.7247e-02, -1.5838e-01, 1.2846e+00,
2.1946e+00, 9.3870e-01, -5.0840e-01, 1.1144e-01,
-9.6993e-01, -6.0428e-01, 6.8721e-02, -2.9233e-01,
7.6383e-01, 8.4709e-01, -5.2885e-01, 1.2977e+00,
5.7832e-01, -2.2131e-01, -1.6796e-01, 3.9574e-02,
-6.4392e-01, -1.0344e+00, 9.7976e-02, -2.6981e-01,
-3.2509e-01, 1.2237e+00, -4.7060e-01, -8.5785e-01,
7.9468e-01, 3.9761e-01, 4.1171e-03, -8.0639e-01,
4.0837e-01, -1.1441e+00, 8.1129e-01, 9.9810e-01,
-6.4338e-02, 6.6963e-02, -9.1013e-01, -1.6663e+00,
-9.6067e-01, -3.9883e-01, -1.2884e-01, -1.4008e+00,
9.6259e-01, 1.9116e-01, 5.4006e-01, 4.4848e-01,
-2.3094e-01, -1.4227e-01, 6.4775e-01, -2.5803e-01,
-1.1213e+00, 8.3679e-01, -1.1746e+00, -2.5890e-01,
-3.4803e-01, -6.5672e-01, -6.8248e-01, -3.4701e-01,
-1.1117e+00, 1.1076e+00, -7.7863e-01, 5.1784e-01,
5.0642e-01, -2.0743e-01, -5.2042e-01, 1.3131e+00,
6.3399e-01, 2.3691e-01, 7.2556e-01, -4.2061e-01,
2.0851e-01, -1.8054e-01, 1.0162e+00, -3.8179e-01,
4.8977e-02, -1.5620e+00, -6.0892e-01, 1.8965e+00,
-1.3371e+00, 1.1450e+00, 1.5963e+00, -1.9053e+00,
7.7885e-01, -8.1566e-01, -6.6321e-01, 3.1558e-01,
-5.8205e-01, 5.9067e-01, -3.0311e-01, 3.0478e-01,
1.2894e+00, -2.5772e-01, -3.4630e-01, -4.7768e-01,
-4.6345e-01, 1.1209e-01, -2.1818e+00, -3.1455e-01,
5.5983e-01, 2.1632e-02, 1.1908e+00, -2.5395e-01,
-3.0246e-01, 4.8415e-01, -8.1615e-01, 6.7366e-02,
-1.3745e-01, -4.2006e-01, 2.6579e-01, -1.2086e+00,
-2.2033e+00, -1.8500e+00, 6.1013e-01, 1.6159e+00,
-6.9113e-01, 4.9113e-01, 4.2234e-01, -4.2230e-01,
-2.2090e-01, -5.4914e-01, 1.0050e+00, 8.9124e-01,
2.2515e-01, -1.3628e+00, 1.1496e+00, -2.4832e-01,
3.6596e-01, 1.3870e+00, -9.1264e-01, 1.0841e+00,
2.1155e-01, 4.2133e-01, 5.9313e-02, 6.3489e-01,
1.2981e+00, -1.0236e-01, 1.4692e+00, 3.6281e-01,
-1.5933e+00, -6.2590e-01, -3.6216e-01, -3.5763e-01,
-9.7007e-01, -6.0164e-02, 1.2426e+00, 4.8412e-01,
-4.6559e-02, -6.9849e-01, 1.5833e-01, 3.0430e+00,
2.4278e+00, 1.4754e+00, -3.4257e-01, 3.3453e-01,
1.1375e+00, 9.7826e-01, -1.4202e+00, 1.7953e+00,
5.4297e-01, 1.4735e+00, -1.0038e+00, -1.3657e+00,
-2.7892e-01, -8.1743e-01, -1.5464e+00, 2.2762e-01,
-2.4116e-01, 5.6781e-01, 1.0303e+00, -3.1346e-01,
-8.9577e-01, 8.5765e-01, 1.5978e+00, 1.8341e+00,
-1.8859e-01, 7.2785e-01, 2.3277e-01, -1.2581e+00,
-3.0438e-01, 6.8164e-01, -3.9202e-01, 1.2624e+00,
2.8774e-01, 2.1528e-01, 9.5611e-01, -5.5930e-02,
4.8346e-01, -2.7975e-01, -1.0903e+00, -9.4228e-03,
-5.8527e-02, 3.2908e-01, 5.4624e-01, -3.2579e-01,
5.7246e-01, 4.7989e-01, -6.5144e-01, 5.7807e-01,
3.2267e-01, 8.8225e-01, -1.2185e+00, 5.0256e-01,
-1.2965e+00, 6.6939e-01, -9.6043e-01, 1.4849e+00,
-1.0684e+00, 7.6762e-01, 1.2612e+00, -7.2934e-01,
1.8094e-01, -7.8679e-01, 1.1524e-01, -1.0763e+00,
-1.8886e+00, 1.7719e+00, 1.3995e+00, -6.2670e-01,
-1.3764e+00, -8.5481e-01, 3.5463e-01, 1.4917e+00,
-2.5663e+00, 1.3411e+00, -4.5171e-01, -1.1377e+00,
3.0532e+00, -3.4094e-01, -5.2497e-01, 1.1965e+00,
3.4147e-02, 2.4035e+00, -9.9713e-01, -6.7185e-01,
-2.2684e-01, -1.3027e-02, 1.6016e+00, -7.9835e-01,
-1.0888e+00, 2.7909e-01, 1.6163e+00, 5.1288e-01,
-2.0013e+00, -1.1853e+00, 3.1276e-01, -5.1879e-01,
-4.0567e-01, 1.1196e+00, 1.5833e-01, -7.3907e-02,
4.8382e-01, -9.9747e-01, 4.9709e-01, 1.2453e+00,
1.0332e+00, -1.2473e+00, -2.2498e-01, -8.0507e-01,
-2.5696e-01, -1.0351e+00, -2.2509e+00, 6.9375e-01,
-1.3717e+00, 1.4816e-01, 5.1258e-01, 8.8223e-01,
2.3615e-02, 3.2715e-01, 1.6091e+00, 1.4051e+00,
5.2218e-01, 7.6995e-01, -9.2637e-01, -7.2946e-01,
9.0335e-01, 8.3630e-01, 7.0489e-01, -3.2802e-01,
1.6625e-02, -4.1946e-01, 3.4437e-01, -7.6515e-01,
1.0795e+00, 1.7134e-01, 9.0791e-01, 6.6335e-01,
1.1088e+00, 1.3046e+00, 6.7427e-01, -2.6491e-01,
1.5726e+00, -2.8366e-03, 1.2024e+00, 1.0614e+00,
-1.2109e+00, -3.9723e-02, -1.0169e+00, 2.2484e+00,
7.2299e-01, -1.8025e+00, 1.0507e+00, -9.1671e-01,
8.1536e-01, 6.3220e-03, -3.8879e-01, 3.7404e-01,
2.0012e+00, -1.0277e+00, -4.2646e-01, -5.4685e-01,
-1.2244e+00, 3.5843e-01, -7.0020e-02, -8.5100e-01,
-2.6113e-01, -2.1377e-01, 8.0506e-01, 1.4043e+00,
1.0853e+00, 4.3254e-01, -6.1877e-01, -8.8055e-01,
-2.0446e+00, -5.5927e-01, -2.3441e+00, 1.0840e+00,
5.0527e-01, -1.0776e+00, -1.7421e+00, -3.2694e-01,
6.4818e-01, -4.9212e-01, -1.8325e-01, 1.0429e+00,
1.8449e+00, -2.2105e+00, -1.4949e+00, 5.5683e-01,
-3.2010e-01, -7.9960e-01, 5.6682e-01, 4.5947e-01,
1.2198e+00, 3.3389e-01, -4.7855e-01, -7.1867e-01,
-1.0509e+00, -4.4068e-01, 5.7454e-01, 4.3343e-01,
8.9727e-01, 2.0940e-01, -1.1906e+00, 8.1320e-01,
-3.4610e-02, -1.7007e+00, 7.1738e-01, 6.2155e-01,
-3.5714e-01, 3.6710e-01, 2.3023e+00, 1.8531e+00,
-6.1109e-01, 2.3105e-01, 2.5151e-01, 1.8025e+00,
3.1952e-01, 5.6062e-01, -3.2077e-01, 6.5552e-01,
-4.1318e-02, -3.2553e-01, 3.6305e-01, -6.1358e-01,
8.2575e-01, 4.8602e-01, -6.0998e-02, -1.8869e-01,
5.5177e-01, -1.2154e+00, 1.6820e+00, 1.1305e+00,
7.2566e-01, 7.3846e-01, -1.2996e+00, 5.3428e-01,
2.1085e-01, 1.6163e+00, 2.2423e+00, 1.3255e+00,
1.3378e+00, -4.8311e-01, -3.4563e-01, 2.9579e-01,
-5.2395e-01, -7.6477e-02, -1.4480e+00, -1.3787e+00,
-1.1171e+00, 1.4660e+00, -4.3653e-01, -1.1789e+00,
2.1326e+00, 2.6186e-01, -1.0269e+00, 9.6927e-01,
7.4635e-02, 1.2699e+00, 1.0576e+00, -2.1719e-01,
-1.3756e+00, 4.8233e-01, 1.2341e+00, 1.2019e+00,
1.5977e+00, -1.7538e+00, 3.8024e-03, -8.8605e-01,
3.4062e-01, 5.5646e-01, 3.6678e-01, 1.6165e+00,
9.7009e-01, 6.2853e-01, -2.0437e-01, 6.2815e-01,
-2.3869e-01, 1.0789e+00, -4.0854e-01, 2.1375e-01,
-2.7437e-01, 3.3338e-01, -1.5663e+00, -2.4861e-01,
1.8815e-01, 9.3455e-01, -1.8994e+00, -4.9050e-01,
-2.3433e-01, 2.3463e-01, -7.7949e-01, 8.8990e-01,
-3.3897e-01, -5.2663e-01, 2.7658e-01, -4.4943e-01,
9.5837e-01, 7.7637e-01, 9.2963e-01, 1.4249e+00,
-5.0999e-01, 3.9453e-01, 9.5984e-02, 5.8528e-01,
5.2025e-01, 1.0584e+00, 4.0023e-01, -3.4842e-01,
-5.7213e-01, 9.9679e-01, -5.9674e-01, -4.1858e-01,
-1.1104e-01, -7.0216e-01, -2.1382e+00, 1.8053e+00,
-1.0045e+00, -6.7924e-01, -1.1691e+00, -6.6565e-01,
-7.2840e-03, 5.1839e-01, -6.3692e-01, 2.9516e-01,
1.1771e+00, 1.6407e+00, 5.6581e-01, -1.1719e+00,
-3.5027e-01, 1.1088e+00, 2.5556e-01, 2.4099e-01,
6.8503e-01, 4.5017e-01, 2.0006e-01, -2.6993e-01,
9.9829e-01, 3.6412e-01, -1.6085e+00, 1.7407e+00,
1.0631e+00, 9.2874e-01, 1.4398e+00, 8.1982e-02,
-2.6963e-01, 2.7150e-04, 1.5189e+00, -1.4538e+00,
-1.5381e+00, -2.2879e-01, -1.6059e-01, -3.7228e-01,
-1.1967e+00, -3.6770e-01, 7.9067e-01, 3.4917e-01,
3.3837e-01, -2.7175e-01, 7.2868e-01, 2.8289e+00,
8.7240e-01, 2.3808e-01, 1.0081e+00, -1.1530e+00,
1.5262e-01, -7.7727e-02, 5.7539e-01, 1.9946e+00,
-1.1267e+00, -4.9483e-01, 3.7268e-03, 1.3245e-01,
9.8200e-01, 5.2143e-02, -2.7788e+00, -4.6267e-01,
3.5470e-01, -8.3092e-01, -4.7913e-01, -1.2856e+00,
-8.4204e-01, 1.1145e+00, 1.8091e+00, 6.9720e-01,
1.8108e+00, -5.2550e-01, -6.0526e-01, 9.8005e-01,
-1.1457e+00, 2.3116e-01, 1.9249e-01, -1.8658e+00,
-4.7475e-01, 3.8170e-01, 1.3984e+00, 2.9747e-01,
1.5098e+00, 1.2237e+00, 3.6738e-01, -7.9580e-01,
-6.9312e-01, -8.8584e-01, 6.9478e-01, 7.4070e-01,
-2.6889e-01, 2.6489e-01, 5.0775e-01, 3.3952e-01,
-1.5221e+00, 5.2449e-01, -6.6295e-03, -1.2241e+00,
1.6044e-02, -1.0920e+00, -6.3790e-01, 7.0208e-01,
3.9040e-01, -6.1183e-02, -3.2645e-01, -4.0683e-01,
-1.1934e+00, -3.0262e+00, -2.6878e-01, 2.0865e-01,
-5.9991e-01, 6.7066e-01, -1.5210e+00, -2.0601e-01,
1.1857e+00, 1.2120e+00, 7.2776e-01, -2.2840e+00,
-5.5023e-01, -1.3025e+00, -9.8657e-01, -9.4564e-01,
-1.4542e-01, 3.1721e-01, 1.2177e-01, -2.3689e-01,
-6.6915e-02, -2.6412e-01, 4.0012e-01, 5.9813e-02,
-3.3095e-01, -2.2993e+00, 1.0324e-01, 1.5798e+00,
6.0813e-01, 7.6398e-01, 1.2187e+00, -3.6947e-01,
3.7906e-01, 3.7939e-01, 1.0384e+00, -5.5732e-02,
4.7542e-02, 1.2617e+00, -4.3388e-01, -1.4722e+00,
7.8295e-01, -1.0196e+00, -3.9518e-01, -2.4531e-01,
-1.5125e-01, -4.9058e-01, 9.5067e-01, -1.2708e+00,
1.7554e+00, -3.6575e-01, 1.2354e+00, 4.3132e-01,
9.0332e-01, 2.5216e-01, -1.1321e+00, 6.7567e-01,
2.1159e+00, 6.6177e-01, 4.9455e-01, -2.0402e+00,
-1.1083e+00, 1.0495e+00, -3.1532e-01, -9.2323e-01,
-1.3443e+00, 1.8521e+00, -1.2809e-01, -1.0717e+00,
4.4377e-01, -9.4508e-01, -1.4632e+00, -6.3528e-01,
-7.1330e-01, 1.2530e+00, 1.5023e-01, 7.0206e-01,
-4.6564e-01, -8.4533e-01, -1.6425e+00, -5.6744e-01,
3.9033e-01, 5.2305e-01, 6.9952e-01, 2.6702e-01,
1.9561e-01, 4.3401e-01, 8.4666e-02, -1.9149e-01,
-5.6678e-01, 1.4842e+00, -1.7494e-01, 1.7113e-01,
1.3101e+00, -1.2393e+00, 2.4770e+00, -2.3210e-01,
5.4365e-01, 2.3116e-01, -6.2748e-01, -1.2196e+00,
1.6693e+00, -1.3651e-01, 7.7219e-01, -4.8107e-01,
1.0282e+00, 1.4752e+00, -1.4725e-01, 7.5750e-01,
-1.0719e+00, 1.4204e+00, -5.2840e-01, 7.2556e-01,
4.9613e-02, -9.0223e-01, -1.4259e-01, -1.3080e+00,
1.7487e+00, -4.8945e-01, -3.5038e-01, 2.7354e-01,
-8.6033e-01, -2.1169e+00, 7.1351e-01, -1.8134e+00,
-1.1948e+00, -7.5889e-01, 6.3285e-01, 1.0819e+00,
-1.8171e+00, 5.0034e-01, -2.2569e+00, 1.0262e+00,
-2.1513e-01, -1.6599e+00, -5.9996e-02, 1.0105e+00,
-2.1154e+00, -2.7981e-01, 2.9200e-01, -9.6077e-01,
1.3051e+00, -6.3529e-01, 4.4569e-02, 1.2877e-01,
-6.9683e-01, 6.2886e-01, -2.5021e+00, -5.0856e-01,
5.2660e-02, -2.7380e-01, 2.6669e-01, -4.7975e-01,
1.7343e+00, 1.2458e+00, 1.2610e+00, -1.1433e+00,
1.8364e-01, 5.6740e-01, 1.3825e+00, 2.4821e-01,
2.2500e+00, 1.2575e+00, 8.8799e-01, 7.7097e-02,
1.3197e+00, 6.9076e-01, 1.4227e+00, 1.3865e-01,
-1.2342e+00, 1.1481e-01, 3.3851e-01, -1.1594e+00,
1.2351e-01, 8.2577e-01, -4.7048e-01, 6.3113e-01,
-8.9680e-01, -1.2326e+00, 3.5810e-01, 1.5616e+00,
-6.0111e-01, 6.3751e-01, 3.4320e-01, -6.3613e-01,
-1.8328e+00, 2.0378e-01, -2.6458e-01, 1.9648e-01,
-4.3603e-01, 3.6306e-01, -1.0498e+00, -1.7427e+00,
7.3474e-01, 1.9844e+00, 1.2203e+00, -1.3024e-01,
6.9773e-01, -3.2055e-01, 2.5124e-02, -2.2077e-01,
-6.2708e-01, 1.1041e+00, 4.6044e-01, 2.4757e-02,
-6.8674e-02, 2.2164e-01, 1.0517e+00, -6.2095e-01,
-5.5854e-01, 4.5275e-01, 3.9225e-01, -2.7085e-01,
2.3666e-01, -3.0335e-01, -7.3988e-02, 1.5429e+00,
8.6628e-02, -3.0852e+00, -4.4412e-01, -1.4957e-03,
5.6003e-01, -1.0861e+00, 5.1258e-01, -1.5563e-01,
-9.9270e-01, -3.4407e-01, -1.3710e+00, -3.7343e-01,
-1.2692e+00, 1.5319e-01, -1.9604e+00, 2.3194e+00,
1.5912e+00, 3.8704e-02, -1.5021e-01, 1.3297e+00,
3.3195e-01, 6.4505e-01, -3.3755e-01, 1.0111e+00,
4.3147e-01, 1.2171e+00, 8.6452e-01, -1.4805e-01,
-1.1412e+00, -4.5805e-01, -6.9215e-01, 3.4390e-01,
-5.2862e-01, 4.7393e-01, -1.0682e+00, 1.6752e-01,
-9.6358e-01, -9.1412e-01, 1.5162e+00, 6.9533e-01,
1.5635e+00, 4.5192e-01, -8.1156e-01, 1.6523e-01,
-1.0150e+00, 7.7748e-01, 1.0199e+00, 1.7440e+00,
-2.7373e-01, 7.2097e-01, 1.2072e-01, -5.2520e-02,
-2.7958e-01, 1.0542e+00, 7.6258e-01, 2.1059e+00,
4.9881e-01, 3.9517e-01, 1.2090e+00, 9.2054e-01,
1.4122e-01, -6.3346e-01, -2.3927e-01, -1.2589e+00,
3.5968e-01, -6.3301e-01, -2.2351e+00, 1.2034e-01,
4.2830e-01, 1.1810e+00, -7.0758e-01, -2.7252e+00,
3.4989e-02, -7.0880e-01, 2.7555e-02, -6.9174e-01,
4.5584e-01, 1.7592e-01, 1.1455e-01, 7.4348e-01,
1.6374e+00, -9.6609e-01, -3.7679e-01, 5.6164e-01,
-1.8639e+00, 1.3913e+00, -1.0705e+00, 3.4157e-01,
-2.9685e-01, 1.7885e+00, 2.1616e+00, -9.7058e-01,
1.8309e-01, 9.0027e-01, 1.3771e+00, 3.9202e-01,
-1.8969e+00, -8.7223e-01, -5.2439e-01, 8.6957e-01,
6.3362e-01, -8.0012e-01, -4.2359e-01, -1.8762e+00]),
size=(10000, 10000), nnz=1000, layout=torch.sparse_csr)
tensor([0.6078, 0.2670, 0.7163, ..., 0.6248, 0.5701, 0.5212])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 1000
Density: 1e-05
Time: 10.149910688400269 seconds

1
pytorch/output_1core/altra_10_2_10_10000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 232594, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 5000, "MATRIX_DENSITY": 5e-05, "TIME_S": 10.347587585449219, "TIME_S_1KI": 0.04448776660382133, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 211.02262233734132, "W": 20.325620076607365, "J_1KI": 0.90725737696304, "W_1KI": 0.08738669130161296, "W_D": 10.790620076607365, "J_D": 112.02929783344271, "W_D_1KI": 0.046392512603968136, "J_D_1KI": 0.00019945704792027368}

16
pytorch/output_1core/altra_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 /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, ..., 5000, 5000, 5000]),
col_indices=tensor([2868, 7932, 249, ..., 7975, 3408, 5305]),
values=tensor([ 2.4798, 0.2185, 0.9681, ..., 0.4257, -1.2815,
-1.0756]), size=(10000, 10000), nnz=5000,
layout=torch.sparse_csr)
tensor([0.4898, 0.4327, 0.8103, ..., 0.9373, 0.0546, 0.0972])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 5000
Density: 5e-05
Time: 10.347587585449219 seconds

1
pytorch/output_1core/altra_10_2_10_20000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 59076, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 39998, "MATRIX_DENSITY": 9.9995e-05, "TIME_S": 10.14804720878601, "TIME_S_1KI": 0.171779524828797, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 218.31626580238344, "W": 21.429018605550787, "J_1KI": 3.695515366686699, "W_1KI": 0.36273645144476246, "W_D": 11.904018605550787, "J_D": 121.27671069979668, "W_D_1KI": 0.2015034634293247, "J_D_1KI": 0.0034109192130361687}

16
pytorch/output_1core/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, 0, 3, ..., 39998, 39998, 39998]),
col_indices=tensor([ 3923, 4463, 7742, ..., 473, 4406, 15797]),
values=tensor([-1.0869, 0.4083, -0.1118, ..., -0.2764, 0.8806,
-0.5343]), size=(20000, 20000), nnz=39998,
layout=torch.sparse_csr)
tensor([0.4042, 0.1033, 0.9915, ..., 0.8010, 0.0420, 0.3163])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 39998
Density: 9.9995e-05
Time: 10.14804720878601 seconds

1
pytorch/output_1core/altra_10_2_10_20000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 167541, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 4000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.27948784828186, "TIME_S_1KI": 0.06135505845304648, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 206.614094543457, "W": 20.665713760266534, "J_1KI": 1.2332151207373538, "W_1KI": 0.12334720313395846, "W_D": 11.170713760266535, "J_D": 111.6838709640503, "W_D_1KI": 0.0666745080921478, "J_D_1KI": 0.0003979593537829415}

16
pytorch/output_1core/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, 0, 0, ..., 4000, 4000, 4000]),
col_indices=tensor([ 1080, 8729, 18175, ..., 17993, 9127, 87]),
values=tensor([ 0.4773, 1.2327, -0.1343, ..., 0.2605, 1.0707,
0.2212]), size=(20000, 20000), nnz=4000,
layout=torch.sparse_csr)
tensor([0.1046, 0.3432, 0.0078, ..., 0.4521, 0.7459, 0.6533])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 4000
Density: 1e-05
Time: 10.27948784828186 seconds

1
pytorch/output_1core/altra_10_2_10_20000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 77166, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 19999, "MATRIX_DENSITY": 4.99975e-05, "TIME_S": 10.135813474655151, "TIME_S_1KI": 0.1313507694406235, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 212.44719322204588, "W": 20.92446157139477, "J_1KI": 2.753119161574345, "W_1KI": 0.27116167186837165, "W_D": 11.469461571394769, "J_D": 116.45006540775297, "W_D_1KI": 0.1486336154704762, "J_D_1KI": 0.0019261542061332223}

16
pytorch/output_1core/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, 1, 3, ..., 19998, 19998, 19999]),
col_indices=tensor([15328, 1557, 7791, ..., 7811, 7304, 544]),
values=tensor([-0.0434, 0.6146, 0.0336, ..., 2.3152, 2.2408,
-0.3543]), size=(20000, 20000), nnz=19999,
layout=torch.sparse_csr)
tensor([0.6138, 0.3544, 0.9837, ..., 0.2956, 0.2755, 0.2504])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 19999
Density: 4.99975e-05
Time: 10.135813474655151 seconds

1
pytorch/output_1core/altra_10_2_10_50000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 16259, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 249989, "MATRIX_DENSITY": 9.99956e-05, "TIME_S": 10.448587656021118, "TIME_S_1KI": 0.6426340891826754, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 265.63406348228455, "W": 25.74416236713474, "J_1KI": 16.337663047068364, "W_1KI": 1.5833791971913858, "W_D": 16.45916236713474, "J_D": 169.82934300780298, "W_D_1KI": 1.0123108658056916, "J_D_1KI": 0.06226156994930141}

17
pytorch/output_1core/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, 9, ..., 249983, 249986,
249989]),
col_indices=tensor([ 2109, 4915, 9069, ..., 21226, 23887, 48661]),
values=tensor([-0.5327, -2.0009, -0.4522, ..., 0.2326, -0.6829,
0.9097]), size=(50000, 50000), nnz=249989,
layout=torch.sparse_csr)
tensor([0.3317, 0.4011, 0.1693, ..., 0.4876, 0.5926, 0.5720])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 249989
Density: 9.99956e-05
Time: 10.448587656021118 seconds

1
pytorch/output_1core/altra_10_2_10_50000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 38882, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 25000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.414772033691406, "TIME_S_1KI": 0.2678558724780466, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 230.91071128845212, "W": 22.283717353358732, "J_1KI": 5.938756012768174, "W_1KI": 0.5731113973910481, "W_D": 12.918717353358732, "J_D": 133.86771002769467, "W_D_1KI": 0.33225444558815725, "J_D_1KI": 0.008545199464743513}

16
pytorch/output_1core/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, 0, 0, ..., 24999, 25000, 25000]),
col_indices=tensor([22311, 35881, 31128, ..., 24299, 37246, 44423]),
values=tensor([ 1.1649, -0.3542, -0.8853, ..., 0.1551, 0.1696,
-2.7590]), size=(50000, 50000), nnz=25000,
layout=torch.sparse_csr)
tensor([0.4524, 0.0084, 0.2512, ..., 0.5914, 0.1409, 0.6936])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 25000
Density: 1e-05
Time: 10.414772033691406 seconds

1
pytorch/output_1core/altra_10_2_10_50000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Altra", "ITERATIONS": 21272, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 124999, "MATRIX_DENSITY": 4.99996e-05, "TIME_S": 10.406851530075073, "TIME_S_1KI": 0.48922769509566916, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 269.6182607269287, "W": 25.94493503699868, "J_1KI": 12.674796010103833, "W_1KI": 1.2196753966246088, "W_D": 16.629935036998678, "J_D": 172.81732077121734, "W_D_1KI": 0.7817758103139656, "J_D_1KI": 0.03675140138745608}

17
pytorch/output_1core/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, 5, ..., 124997, 124998,
124999]),
col_indices=tensor([34758, 37051, 4283, ..., 36902, 8930, 21854]),
values=tensor([ 0.2027, -0.2421, -0.0122, ..., 2.0567, 0.6418,
-1.9333]), size=(50000, 50000), nnz=124999,
layout=torch.sparse_csr)
tensor([0.7323, 0.4458, 0.6264, ..., 0.0699, 0.5750, 0.3790])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 124999
Density: 4.99996e-05
Time: 10.406851530075073 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_100000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 7096, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 999948, "MATRIX_DENSITY": 9.99948e-05, "TIME_S": 10.168954849243164, "TIME_S_1KI": 1.43305451652243, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 672.870105085373, "W": 66.03, "J_1KI": 94.82385922849113, "W_1KI": 9.305242390078917, "W_D": 46.44625, "J_D": 473.3044543135166, "W_D_1KI": 6.545412908680947, "J_D_1KI": 0.9224088090023883}

17
pytorch/output_1core/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, 17, ..., 999931, 999938,
999948]),
col_indices=tensor([ 4524, 19684, 22420, ..., 85904, 94042, 95795]),
values=tensor([-1.3594, -0.6760, 1.5198, ..., -0.4902, -0.6733,
-0.9072]), size=(100000, 100000), nnz=999948,
layout=torch.sparse_csr)
tensor([0.5407, 0.5422, 0.8589, ..., 0.3802, 0.1065, 0.2777])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 999948
Density: 9.99948e-05
Time: 10.168954849243164 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_100000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 15524, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 100000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.355973243713379, "TIME_S_1KI": 0.6670943857068654, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 670.6344871044159, "W": 64.68, "J_1KI": 43.199851011621746, "W_1KI": 4.166451945374904, "W_D": 45.22375000000001, "J_D": 468.90238692313443, "W_D_1KI": 2.9131506055140433, "J_D_1KI": 0.18765463833509685}

17
pytorch/output_1core/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, 2, 4, ..., 99998, 99999,
100000]),
col_indices=tensor([52367, 87862, 40992, ..., 72156, 81616, 61830]),
values=tensor([-1.2130, 0.3871, 0.5809, ..., 0.0845, 0.0061,
0.2524]), size=(100000, 100000), nnz=100000,
layout=torch.sparse_csr)
tensor([0.1658, 0.3689, 0.6232, ..., 0.0281, 0.1230, 0.4103])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 100000
Density: 1e-05
Time: 10.355973243713379 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_100000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 9636, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 499985, "MATRIX_DENSITY": 4.99985e-05, "TIME_S": 10.401328325271606, "TIME_S_1KI": 1.0794238610701128, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 675.1464500951768, "W": 65.51, "J_1KI": 70.06501142540232, "W_1KI": 6.798464092984641, "W_D": 45.37875000000001, "J_D": 467.67366771876823, "W_D_1KI": 4.709293275217934, "J_D_1KI": 0.48871868775611604}

17
pytorch/output_1core/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, 8, 17, ..., 499979, 499983,
499985]),
col_indices=tensor([ 5105, 20642, 52866, ..., 97652, 88458, 89695]),
values=tensor([ 0.4983, 1.2371, 1.2262, ..., 0.2227, -0.8154,
-1.7583]), size=(100000, 100000), nnz=499985,
layout=torch.sparse_csr)
tensor([0.6479, 0.3012, 0.7174, ..., 0.8126, 0.2989, 0.7422])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 499985
Density: 4.99985e-05
Time: 10.401328325271606 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_10000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 377024, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 9999, "MATRIX_DENSITY": 9.999e-05, "TIME_S": 10.560901880264282, "TIME_S_1KI": 0.028011219127334817, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 693.101114051342, "W": 65.39, "J_1KI": 1.8383474634276387, "W_1KI": 0.1734372347648956, "W_D": 46.042500000000004, "J_D": 488.02734429895884, "W_D_1KI": 0.1221208729417756, "J_D_1KI": 0.0003239074248370809}

16
pytorch/output_1core/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, 1, 2, ..., 9998, 9999, 9999]),
col_indices=tensor([8074, 8881, 2785, ..., 4144, 527, 3612]),
values=tensor([-1.2557, -1.0833, 0.1504, ..., -2.2932, 0.0716,
-0.7273]), size=(10000, 10000), nnz=9999,
layout=torch.sparse_csr)
tensor([0.6353, 0.6903, 0.7256, ..., 0.9307, 0.7404, 0.5188])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 9999
Density: 9.999e-05
Time: 10.560901880264282 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_10000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 634293, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 1000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.566636562347412, "TIME_S_1KI": 0.016658920344931147, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 669.9189710187911, "W": 64.99, "J_1KI": 1.0561664262711257, "W_1KI": 0.10246053480016332, "W_D": 45.294999999999995, "J_D": 466.90228946447365, "W_D_1KI": 0.07141021578355743, "J_D_1KI": 0.000112582380356645}

375
pytorch/output_1core/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([1698, 2743, 1039, 6916, 2774, 2597, 923, 7219, 396,
4866, 834, 6999, 7087, 2820, 3930, 2118, 9590, 8266,
7993, 357, 479, 8961, 9987, 3231, 9668, 9626, 907,
8007, 8206, 2305, 2321, 492, 5876, 7706, 1004, 5699,
2751, 8337, 5455, 4515, 2835, 8478, 2625, 102, 7988,
4049, 5405, 3458, 1977, 9297, 2105, 2993, 1344, 9348,
6100, 2703, 8225, 7647, 3119, 1193, 238, 8458, 5019,
1718, 4649, 207, 1628, 7502, 4336, 5724, 4510, 6851,
468, 9661, 1612, 6178, 2594, 6565, 4936, 9321, 5927,
9620, 3848, 7114, 5578, 2034, 1680, 951, 3358, 475,
3085, 7183, 4423, 9677, 9327, 7724, 1601, 6116, 2386,
3393, 870, 1429, 9229, 5322, 3465, 9681, 5015, 9277,
7854, 7184, 8452, 303, 918, 808, 37, 8062, 2242,
9309, 3842, 7109, 9542, 4679, 3670, 3474, 5566, 2045,
8284, 2694, 5176, 9956, 1600, 5340, 5705, 5611, 5005,
9494, 3225, 2582, 3123, 205, 7942, 22, 70, 8730,
4227, 1530, 2466, 4741, 4748, 863, 8274, 7422, 8375,
4343, 9261, 5463, 8724, 7419, 9441, 5595, 3224, 2774,
7720, 4426, 4297, 4041, 2636, 7923, 9378, 4499, 3509,
1586, 2744, 9571, 9465, 9670, 6418, 248, 3084, 2884,
9343, 1313, 6563, 6117, 5400, 1135, 8658, 9705, 9823,
1756, 8527, 1981, 7281, 1820, 1424, 7948, 160, 7499,
3013, 6153, 1590, 635, 3894, 4402, 2166, 1658, 2161,
6374, 6301, 7327, 9259, 2264, 9556, 7015, 5035, 599,
3847, 59, 1201, 148, 7800, 2667, 7949, 1728, 8410,
5677, 8437, 8740, 2817, 3055, 4034, 6322, 7697, 5774,
5630, 1360, 5998, 9734, 2914, 5395, 4345, 3718, 5656,
9029, 9732, 8692, 2013, 3075, 6479, 916, 1750, 8442,
5889, 6561, 9770, 9419, 2911, 2893, 5215, 2550, 6323,
8339, 1484, 5364, 8290, 5742, 9005, 7842, 4336, 3895,
8637, 3530, 1518, 2833, 1884, 7452, 9481, 3586, 6531,
4775, 3441, 6258, 5368, 12, 9064, 8317, 4028, 7700,
7067, 2154, 9420, 2060, 4918, 9078, 7986, 9327, 3611,
1602, 1982, 9216, 708, 4090, 2030, 4453, 8502, 9074,
6118, 3701, 8413, 2801, 9004, 9491, 9749, 1422, 7912,
1313, 4148, 2697, 4422, 9636, 4100, 8846, 7167, 7699,
9434, 5217, 1092, 4035, 8538, 7222, 4177, 9164, 4665,
6949, 3185, 5507, 8645, 9913, 9242, 1568, 2239, 4045,
3775, 5408, 4363, 7500, 2464, 4141, 3955, 6179, 7755,
9385, 5942, 7523, 7859, 9598, 4592, 9025, 7338, 5823,
9370, 9476, 8349, 1985, 570, 6986, 3629, 7772, 6443,
8223, 4187, 7164, 5720, 5085, 4292, 3161, 7658, 6863,
1399, 7896, 9183, 957, 7476, 9093, 6629, 8202, 7884,
2611, 7042, 239, 1501, 8272, 6788, 8970, 8436, 1953,
5281, 2387, 4812, 712, 1917, 4728, 859, 6723, 5113,
6098, 9844, 635, 5774, 5387, 6625, 4015, 7073, 7578,
411, 9879, 3740, 7748, 8163, 6399, 782, 3571, 8320,
4869, 8307, 3658, 2128, 6029, 4628, 2642, 4208, 7999,
5391, 7891, 4478, 5014, 6046, 5029, 6308, 572, 7644,
5287, 1314, 3654, 5155, 4146, 9555, 6525, 2875, 8012,
5009, 6591, 5553, 9388, 8575, 2536, 5882, 1448, 4968,
5360, 324, 1548, 7256, 4327, 4863, 1503, 354, 3573,
7520, 4509, 1115, 92, 7402, 942, 5361, 5649, 1172,
5677, 4149, 776, 3953, 6332, 6798, 8362, 3986, 8585,
582, 5178, 117, 6369, 5132, 4306, 2165, 610, 3257,
5897, 3238, 4501, 6618, 2401, 6415, 6605, 5290, 3612,
6705, 6069, 5273, 7070, 1694, 2300, 4742, 9537, 6071,
3180, 8165, 7085, 675, 6090, 3796, 3311, 976, 2744,
1370, 9409, 9222, 6552, 237, 4660, 8904, 5711, 8987,
2850, 9114, 2082, 5687, 4830, 6359, 418, 2149, 7270,
5212, 7999, 5935, 473, 4916, 9219, 2956, 4079, 5452,
9620, 76, 9943, 3571, 9825, 3414, 1500, 7367, 4030,
8405, 7378, 9331, 4061, 3830, 2696, 385, 7919, 532,
58, 499, 4567, 3114, 1747, 352, 231, 6899, 9363,
1031, 7552, 4161, 6125, 3043, 8964, 2406, 9004, 1796,
4155, 2544, 5540, 2882, 8486, 7394, 1220, 3545, 5922,
7621, 9879, 8557, 4997, 4485, 6176, 3946, 6282, 9593,
9097, 6196, 6033, 3305, 8937, 278, 9036, 5616, 1140,
3841, 7784, 776, 4808, 1715, 56, 6768, 3305, 9866,
9544, 891, 4148, 9014, 5497, 4101, 9049, 1446, 7542,
8265, 803, 6661, 3848, 400, 7900, 2288, 2713, 4082,
735, 8165, 5139, 6258, 1916, 9570, 183, 1681, 3282,
8939, 4785, 2073, 2863, 8809, 5926, 8134, 2687, 3953,
7610, 2842, 5110, 1997, 1398, 9515, 8135, 2280, 3088,
294, 5772, 4043, 9516, 591, 7950, 2439, 9797, 4098,
8217, 9350, 4608, 1151, 2415, 8900, 1181, 4583, 5086,
9355, 3567, 5768, 9765, 8222, 5235, 2761, 315, 2666,
8412, 2294, 7174, 3768, 4376, 5138, 7852, 1553, 4643,
430, 9900, 9110, 6575, 7861, 5021, 6661, 6136, 3872,
2312, 9455, 3242, 1741, 8812, 2603, 8047, 2574, 2581,
1776, 7289, 7220, 4962, 6398, 2350, 1100, 6727, 2575,
3981, 7536, 4621, 5073, 6706, 4951, 2840, 8397, 7838,
5184, 9312, 3684, 2385, 1853, 4171, 620, 8684, 9506,
461, 2176, 3638, 4013, 3218, 4962, 6482, 1084, 893,
957, 6021, 7355, 3911, 5785, 9077, 2686, 7610, 4190,
8423, 7643, 4541, 1906, 6904, 4191, 8826, 5165, 8669,
8070, 5439, 5393, 6735, 4019, 5386, 6392, 1431, 136,
5461, 9751, 3342, 7688, 3319, 3439, 1969, 7288, 4646,
8165, 7646, 87, 6784, 7215, 6027, 7102, 5001, 9958,
7338, 363, 4482, 2452, 255, 4297, 7352, 9797, 974,
5998, 326, 3176, 2362, 6791, 3349, 4022, 2240, 1371,
6075, 555, 5143, 3869, 9752, 6413, 8984, 1848, 1833,
8393, 68, 2203, 609, 2197, 4975, 7897, 2373, 4914,
2044, 187, 5507, 6227, 6299, 643, 8322, 5214, 4280,
196, 6585, 874, 2498, 2129, 7573, 7368, 2197, 6317,
3119, 8703, 4175, 4858, 5432, 1855, 2122, 2023, 531,
1722, 6243, 8490, 8327, 4485, 6959, 8846, 5720, 3459,
4817, 2750, 4582, 3270, 8488, 1508, 8724, 9082, 2707,
1229, 33, 5717, 4063, 3409, 361, 3463, 1504, 7873,
6873, 8450, 4212, 6209, 6560, 4110, 9813, 6219, 4403,
5970, 9173, 8821, 9991, 7529, 9811, 6025, 8019, 6690,
1302, 2385, 8359, 6449, 4455, 5958, 5620, 3960, 4768,
7897, 7242, 138, 299, 5589, 1117, 3441, 5891, 3646,
6710, 590, 1866, 3534, 4966, 4389, 803, 5172, 2151,
995, 4852, 9394, 4453, 73, 3919, 6164, 2380, 6275,
9817, 8695, 7770, 616, 3744, 515, 7994, 9019, 1921,
8384, 6029, 2689, 7009, 1551, 552, 1931, 4400, 7005,
1930, 4611, 1454, 4846, 2126, 1867, 4058, 2636, 6809,
3918, 4995, 8703, 4866, 8400, 4380, 7698, 3676, 4535,
6062, 5289, 3042, 7195, 1510, 4512, 3851, 2439, 8244,
6500, 5491, 6466, 3737, 9811, 4077, 6034, 7021, 9955,
7579]),
values=tensor([-8.5689e-01, 1.2097e+00, 6.6290e-01, 3.4176e-01,
-2.0309e+00, 5.4900e-01, 2.9127e-01, 2.1012e+00,
4.7353e-01, -8.0716e-01, 4.9650e-01, -6.0049e-02,
-2.6540e+00, 6.3942e-01, 1.0099e-01, -6.3532e-01,
2.7193e-01, 1.1259e+00, 2.4949e-01, 1.4831e+00,
1.9369e+00, 9.5118e-01, 9.5716e-01, -7.0569e-02,
1.5161e+00, -3.8890e-01, -6.8646e-01, -1.4198e-01,
3.2862e-02, 7.7921e-02, 3.6365e-01, 2.3211e+00,
1.3496e+00, -9.1078e-01, -2.7527e-02, 8.4156e-01,
2.7367e-01, 2.6654e+00, 1.0063e-01, 4.6554e-01,
-9.1779e-01, 4.3967e-01, -5.2488e-01, -1.0941e+00,
1.6657e-01, 2.3238e-03, 2.4518e-01, -9.6366e-01,
1.2252e+00, -3.4085e-01, 9.2751e-01, 2.5336e-02,
5.3063e-01, -5.6305e-01, -4.1729e-01, -1.8912e+00,
2.7819e-01, -5.6433e-01, 7.3080e-01, -1.1630e+00,
5.3933e-01, 1.1984e-01, 1.1545e+00, 1.3946e+00,
-1.6702e-01, -2.7158e-02, 7.1098e-01, 6.2466e-01,
3.3148e-01, -3.2809e-01, -3.2689e-01, -1.7592e+00,
9.0920e-01, -1.4030e-01, 2.4868e+00, -1.8536e-01,
-7.3811e-02, -1.6849e+00, 3.7719e-01, -3.1310e-01,
3.1142e-01, 1.0664e+00, 4.0784e-02, -7.5989e-01,
6.7684e-01, 2.7119e-01, -4.9332e-01, -3.0090e-01,
-4.6301e-01, -1.4742e-01, 1.2262e+00, -4.0203e-01,
3.5398e-01, -2.1315e+00, 9.0932e-01, 1.7203e+00,
-6.3822e-01, 1.0841e+00, 1.9889e-01, 1.4332e+00,
6.9287e-01, 1.2077e+00, -7.5407e-01, 5.6201e-01,
1.1909e+00, 6.5611e-01, 8.0740e-02, 1.9173e-01,
4.3212e-01, -7.2225e-02, -4.6629e-01, -1.0897e-01,
5.8335e-01, 1.4646e+00, 1.5695e+00, -1.1510e-01,
-1.0289e+00, -1.3023e-01, 4.9750e-01, 1.0585e+00,
-5.7645e-02, 2.0094e+00, -2.9508e-01, -4.7097e-01,
-5.2683e-02, -3.5319e-01, -1.6821e+00, -8.4147e-01,
1.2827e+00, -7.6760e-02, 7.1096e-01, 1.1954e+00,
5.7789e-01, 1.1579e+00, 1.0533e+00, 2.4522e-01,
8.6762e-01, -1.3501e+00, -4.8093e-01, -1.1414e+00,
-1.5482e+00, 6.8505e-01, 1.9751e+00, -1.0731e+00,
1.6409e-01, -5.3768e-01, -7.3844e-01, -6.4573e-01,
-7.8483e-01, -8.7924e-01, 4.3129e-01, 1.4047e+00,
1.9155e+00, 5.6637e-01, -6.4022e-01, 7.9506e-01,
-7.0854e-01, -1.2834e+00, -4.1176e-01, -1.3780e-01,
-1.6196e+00, -7.1500e-01, 8.6835e-01, 8.0887e-01,
4.6424e-02, 9.1775e-01, 2.2640e+00, 4.9689e-01,
8.2025e-02, 3.0099e+00, 6.3027e-01, -1.7984e+00,
-6.1999e-01, 1.7909e-01, 1.9490e+00, 7.8218e-01,
-1.3883e+00, -1.1214e+00, 2.1615e-01, -6.1819e-01,
-1.1214e-02, -2.0840e+00, 6.6560e-01, 9.1137e-02,
5.7388e-01, 1.8005e-01, 9.8555e-01, 6.0733e-01,
4.7661e-01, -1.2436e+00, -1.1329e-01, 1.2342e-03,
1.1652e+00, 2.4206e+00, 1.7745e+00, -1.4510e+00,
-9.7770e-01, 5.6283e-01, 6.3687e-02, 1.2809e+00,
4.9439e-02, -8.9922e-01, 6.0771e-01, 3.0981e-01,
-8.2659e-02, 3.5419e-01, -4.7433e-01, -4.6397e-01,
-4.0787e-01, 1.2082e+00, -1.4649e+00, 6.1348e-01,
-8.5734e-01, 1.0216e+00, -1.6124e-01, 6.1655e-01,
6.4240e-01, 1.2095e+00, -1.1977e-01, 4.2474e-01,
8.9273e-01, 2.2927e-01, 5.2743e-01, -3.8270e-01,
1.2204e+00, 2.4922e-01, 1.1471e+00, -1.1385e+00,
6.2492e-01, 1.2899e+00, 1.1440e+00, -6.9629e-01,
-8.4045e-02, 2.0530e-01, -5.5081e-01, 5.0499e-01,
-1.5913e-01, -1.8878e+00, 1.6437e-01, -2.1635e+00,
2.6413e+00, 2.0165e-01, 1.1099e+00, 3.8919e-01,
2.6776e-01, -1.3834e-01, -1.9736e+00, -3.6151e-01,
-2.4112e-01, -8.4378e-01, -3.5888e-01, 3.7164e-01,
4.7567e-01, -1.6478e+00, -2.2089e-01, 8.6319e-01,
4.5733e-01, 5.9154e-01, 1.5991e+00, -3.0355e-01,
2.1325e+00, 1.1055e+00, 1.3590e-01, 3.9073e-01,
-1.7295e+00, -5.2981e-01, -2.1835e-01, 9.3592e-01,
-1.4843e+00, 1.3399e+00, 1.9025e-01, 1.0071e+00,
2.5877e-02, 3.2546e-01, -1.5411e-01, 1.0708e+00,
-1.1882e+00, -1.9991e-01, 4.7020e-01, 6.7998e-02,
7.6877e-01, 8.7577e-01, -1.6423e-01, -1.7507e+00,
-7.4582e-01, -6.3328e-01, 2.0928e+00, 1.7218e+00,
-2.2465e-01, -2.0560e-01, -1.4457e+00, -2.4893e-01,
4.1291e-01, 1.0593e+00, -1.0848e+00, 7.6270e-01,
-6.0062e-01, 2.7033e-01, 1.9684e+00, 5.1004e-01,
5.6979e-01, 1.0909e-02, 6.5170e-01, -3.7072e-01,
-2.7624e-01, -2.7221e-01, -3.8037e-02, -5.2022e-01,
1.2878e+00, 6.4494e-01, 5.3458e-01, 1.2673e+00,
-1.2654e+00, -8.2346e-01, -4.7230e-01, 5.2640e-01,
6.2282e-01, -5.4414e-01, -3.0600e-01, -8.2693e-01,
-4.8921e-01, -1.7811e+00, -1.9990e+00, 4.3150e-01,
-1.9346e+00, -3.0803e-01, 6.1017e-01, -8.3369e-02,
1.0629e-01, -3.4583e-01, -5.5604e-01, 2.9426e-01,
8.9995e-01, -5.7986e-01, 9.5170e-02, -6.1116e-01,
-3.9599e-01, -6.9750e-01, 1.7287e+00, -2.7599e-01,
-9.6241e-01, 2.8961e-01, 8.9299e-01, 3.9939e-01,
-1.3965e+00, -1.5788e+00, 3.7016e-02, 2.6522e-01,
9.4617e-01, 1.2279e+00, 1.4730e-01, 2.2211e+00,
2.3586e+00, -5.8135e-01, -1.8943e+00, -1.2611e+00,
-4.6566e-01, -1.4001e-01, 1.2457e+00, 7.0424e-01,
9.8285e-02, -4.8855e-01, -1.1448e+00, 1.9848e-01,
-4.6023e-01, -1.1413e+00, 4.5395e-02, 6.7274e-01,
1.1623e+00, -4.0169e-01, -6.3421e-01, 2.1598e-01,
3.2160e-01, -3.3337e-01, 3.0242e-01, -1.0492e+00,
7.8076e-01, -2.1904e+00, -1.3415e+00, -1.9779e-01,
-1.1301e+00, -4.0326e-01, 6.2447e-01, 2.9531e-02,
1.0303e+00, -2.1968e-01, -9.5831e-01, -1.0706e-01,
1.4897e+00, 7.1324e-01, -7.8904e-01, -6.5349e-01,
3.0886e-01, -1.0471e+00, -1.8985e+00, 1.0577e+00,
-1.4348e+00, -5.2264e-01, 5.0149e-01, -1.0359e-01,
1.4192e+00, 1.6129e-01, -1.4402e-01, 6.6174e-01,
1.3507e+00, -5.6175e-01, 3.5407e-02, 2.7751e-01,
-1.7356e-01, 6.3910e-01, -3.7465e-02, 1.5066e+00,
-3.8304e-01, -7.1819e-02, -3.4109e-01, 1.9834e+00,
-6.3699e-01, 5.2520e-01, -1.3723e+00, -4.2614e-01,
-9.5754e-01, -1.3294e+00, -8.8089e-01, 2.3412e+00,
-2.4791e+00, -1.0486e+00, -7.1954e-01, 6.8718e-01,
7.8290e-01, -1.4284e+00, 1.7077e+00, 6.1097e-01,
-5.0978e-01, 1.1192e+00, -4.3710e-02, -1.1563e+00,
-4.1027e-01, 1.5081e+00, 1.9014e+00, 1.8975e-02,
-2.0564e-02, 5.1113e-01, 2.0979e-01, -1.7490e+00,
-5.1872e-01, 1.1068e+00, 4.4300e-01, -9.0785e-02,
1.8603e+00, -3.2767e-01, 1.5154e+00, -1.0343e+00,
-1.4210e+00, -6.5645e-02, -4.6452e-01, -9.9318e-01,
1.9451e-01, -1.1981e+00, -4.0686e-02, 6.4455e-02,
7.3492e-01, 1.0813e+00, -9.7804e-01, 9.9781e-01,
5.0819e-01, 3.8654e-01, 2.0798e+00, -3.0960e-01,
1.2902e+00, 1.3761e+00, -6.2642e-02, -4.7770e-01,
7.6771e-01, 1.0105e-01, 1.3634e-02, -7.5883e-01,
9.1743e-01, -3.7837e-01, -1.5279e+00, 3.2870e-02,
5.3553e-01, -6.4774e-01, 1.9871e+00, -4.0720e-01,
-9.1741e-01, -1.3450e+00, 1.7506e+00, 2.9707e-01,
-1.0264e+00, 1.7855e+00, 7.0655e-01, 2.7848e-01,
7.2634e-01, 1.1559e-01, -7.3096e-01, -3.0901e+00,
-7.9027e-01, -4.4089e-01, -6.9713e-02, -7.1950e-01,
4.6355e-03, -3.2028e-01, 1.2251e+00, 2.4139e+00,
1.5456e-01, 2.8065e-01, -2.3067e-01, 2.0358e+00,
1.4949e+00, -1.0179e+00, -1.5601e+00, 1.0164e+00,
-3.0133e-01, -4.2909e-01, 8.1046e-01, 2.4521e-01,
1.8104e-01, -4.4590e-01, -7.9742e-01, 7.9268e-01,
1.0385e+00, -6.4065e-01, 1.8532e+00, 4.6001e-01,
-7.1608e-01, -7.1107e-01, -9.6351e-01, -1.3006e+00,
-9.3178e-01, -4.9779e-01, 5.0896e-01, 2.2150e-01,
-1.6357e-01, 1.7560e-01, 3.5630e-01, 3.4051e-01,
3.9700e-01, -3.2117e-01, 1.3720e+00, 1.2224e-01,
1.5750e+00, -1.1416e+00, 1.1359e+00, 1.1440e+00,
-2.3389e-01, -2.2913e-01, 5.9581e-01, 4.8866e-01,
4.1229e-01, -6.1616e-01, 1.3733e+00, -1.3124e-01,
1.1330e+00, -1.2203e-01, -9.2551e-01, -7.5411e-01,
2.5215e-01, -4.1072e-01, 9.6426e-01, -2.9944e-01,
-9.1142e-01, -6.4125e-01, -2.0563e+00, -1.8080e-01,
-1.4041e+00, -1.0324e+00, 6.8227e-01, 4.8331e-01,
-5.4755e-01, -4.3604e-02, 1.0613e+00, 8.8337e-01,
-1.2126e+00, -5.9985e-02, -1.7255e+00, 2.0184e+00,
-7.2165e-02, -2.8088e-01, -1.8624e-01, 8.6376e-02,
-6.7318e-01, -1.0071e-01, 2.2482e-01, 1.6008e+00,
3.5504e-01, -4.5956e-01, 7.7826e-01, 4.1072e-01,
5.0045e-01, 1.0829e+00, -2.5926e-02, 1.5400e+00,
-4.2367e-01, -1.2776e-01, -2.0890e-01, 2.8788e-01,
-4.8441e-01, -1.1165e+00, -7.0298e-01, -3.3063e+00,
1.0891e+00, -1.0817e+00, 1.4369e-01, 8.3073e-01,
1.9647e-01, 1.3475e+00, -9.6423e-01, -1.8997e-01,
6.1890e-01, 2.0889e-01, -1.4500e+00, -4.1832e-01,
-1.4642e+00, 1.4730e+00, 2.9546e-01, -1.2365e+00,
1.5213e+00, 7.4028e-01, -7.5714e-01, 9.3191e-01,
2.6968e-01, -3.9266e-02, 1.9520e+00, 7.6778e-01,
3.0581e-02, 4.4563e-01, 5.5016e-01, 1.8439e+00,
-1.6014e-01, 1.6832e-01, -1.0260e+00, -9.8299e-01,
-1.3377e+00, -5.3785e-01, 4.4521e-01, -2.8417e-01,
-1.3533e+00, 5.7214e-02, -3.6627e-01, -7.1891e-01,
-1.2485e+00, -8.9925e-01, 2.5183e+00, 5.3125e-01,
-2.7903e-02, -8.8786e-01, 1.0826e+00, -1.4403e+00,
-7.5576e-01, 2.4216e+00, -2.7179e-01, -5.4057e-01,
-4.3449e-01, -2.2057e-01, 1.6880e+00, -4.6452e-02,
1.1322e-01, -3.2739e-01, -6.6454e-01, 2.8627e-01,
-9.0947e-01, -1.0433e+00, 7.8364e-01, 6.6598e-01,
-6.4543e-01, 1.1573e+00, -1.0542e+00, -1.3761e+00,
-1.3893e+00, -1.6869e-01, 1.3070e-01, -5.5741e-01,
5.6324e-01, -1.6500e+00, 1.3002e+00, 9.9171e-01,
6.8222e-01, -2.2311e+00, 1.8324e+00, -1.6144e+00,
8.7300e-01, 1.5826e+00, -8.5160e-01, -1.7282e+00,
1.1178e+00, -9.9170e-01, 5.0548e-01, 1.0514e-01,
-1.0839e+00, 9.0679e-01, -1.6280e-01, 5.9919e-01,
1.1061e+00, -1.7264e-01, 8.0276e-01, 2.1544e+00,
8.2835e-01, 8.2847e-01, -1.5155e+00, 1.2644e+00,
-3.4807e-01, 1.2637e+00, 2.3995e-01, -2.2327e-01,
-1.3088e-01, 6.7821e-02, 2.3044e+00, -2.7062e-01,
1.8816e+00, 5.0019e-01, -1.5037e+00, -2.4944e-01,
-1.2903e+00, 5.3685e-01, -2.5310e-01, 9.5933e-01,
2.6886e-01, 5.1943e-01, 2.6789e-01, -7.1040e-01,
-5.4198e-01, 6.2782e-01, -1.1705e+00, 2.6820e-01,
8.3553e-01, 1.2083e-01, -4.9588e-01, -6.2483e-02,
-7.0767e-01, -1.1254e+00, -2.7930e+00, 1.9913e+00,
1.5345e-01, 1.0150e+00, 6.1777e-01, -8.3391e-01,
1.4251e+00, 5.1195e-01, -1.1433e-01, -6.9002e-01,
1.2152e+00, 9.6290e-01, 1.7890e+00, -2.4689e-01,
-7.8309e-02, 1.1296e+00, -7.0446e-01, 8.6161e-01,
2.1772e-01, -3.1510e-01, 4.7449e-02, 3.4586e-01,
5.4482e-01, -3.5457e-01, -1.3334e+00, -8.3614e-01,
-2.7505e-01, 1.9153e-01, 1.1271e+00, 4.2934e-01,
8.7867e-01, 7.2701e-01, -6.3220e-01, 8.7714e-03,
2.3687e-01, 2.2499e-01, 2.5670e-01, -1.3181e-01,
8.9160e-01, 1.1182e+00, -1.0175e+00, 6.6489e-01,
-2.8800e-01, -5.6134e-01, -1.6781e+00, -5.4235e-02,
3.7937e-01, 9.1954e-01, -7.8936e-01, 9.5492e-02,
1.3691e+00, -6.4487e-02, 1.0831e+00, 1.2509e+00,
1.7708e-01, 6.4521e-01, 5.2638e-01, -6.8988e-01,
-2.7349e-01, -3.4422e-01, -2.8844e-01, -1.7878e+00,
6.7360e-02, 3.5657e-02, 2.0005e-01, 6.9236e-01,
1.3433e+00, -6.8289e-02, -9.1876e-01, 1.0596e+00,
2.4374e+00, 1.9350e-01, -1.5351e-01, -2.0267e+00,
-2.6589e-01, -1.4912e+00, -1.7837e+00, -1.0053e+00,
-2.2196e-01, -4.2461e-03, 5.9882e-01, 4.2340e-01,
-4.5188e-01, 1.8994e+00, 1.0790e+00, -2.0139e+00,
1.4898e+00, -1.1257e+00, -1.1923e+00, 7.7116e-01,
-9.6211e-01, -9.2271e-01, 9.4487e-01, 1.8176e-01,
-6.0543e-01, -1.9627e+00, -3.7713e-01, -1.4271e-01,
-2.2043e-01, 1.9470e+00, -3.6987e-01, 6.8263e-04,
-4.6381e-01, 1.7285e+00, -6.6937e-01, -4.3294e-01,
-2.4764e-01, -5.0450e-01, 6.6482e-01, -3.3063e-01,
9.6943e-01, -1.8964e-01, 8.2059e-01, 7.5185e-01,
-1.6748e-01, 2.2382e-01, -1.1664e+00, 1.5390e+00,
-1.9456e-01, -1.9449e+00, -3.5027e-02, -2.9343e-01,
3.9996e-01, -1.8940e+00, 7.3924e-02, -1.1793e+00,
-1.2512e+00, -1.2817e-01, 6.4452e-01, 5.1107e-01,
-3.2113e-01, 1.1179e+00, 7.0618e-02, -7.8027e-01,
-2.5237e-01, 1.3734e+00, -9.6716e-01, 1.2713e+00,
7.8619e-01, -2.1447e+00, 5.1844e-01, -4.7097e-01,
-2.6642e+00, -7.0739e-01, -8.8190e-02, -1.4605e-03,
2.2912e+00, 7.7741e-01, -1.3949e-01, -4.9428e-01,
4.9101e-01, 1.8230e+00, 2.2876e-01, 1.0535e+00,
-4.6515e-02, -4.1397e-01, -6.6887e-01, 6.4614e-01,
-5.9561e-01, -3.7286e-01, 3.7637e-01, -5.7816e-02,
1.4101e+00, 1.3075e+00, -2.2726e+00, -1.3249e+00,
1.1467e+00, 8.6657e-01, 1.0717e-01, -1.0467e+00,
-1.3209e+00, -7.6885e-01, 1.1640e-01, 6.4706e-01,
2.0351e-02, 2.8236e-01, -1.4893e+00, 1.3627e+00,
9.9925e-01, 1.2451e+00, 3.2840e-01, 5.0067e-02,
-2.6641e-01, -7.7992e-01, -1.1319e+00, 1.0208e+00,
-5.9890e-01, 5.6660e-01, -1.4854e+00, 1.1691e+00,
-1.7132e+00, 8.3444e-01, -4.0707e-01, -1.5139e+00,
8.3768e-01, 7.5062e-01, 4.3870e-01, 1.3469e+00,
2.9784e-01, -1.6711e+00, 1.1040e+00, 1.4180e+00,
-2.8145e-01, -1.0619e-01, 9.2538e-01, -1.3258e+00,
7.8037e-01, 1.7529e-02, -3.7982e-01, 1.6881e+00,
-1.5036e+00, -2.4716e+00, 1.4702e+00, -1.1938e+00,
-5.7362e-01, -1.6876e+00, 8.9935e-01, 8.2957e-01,
5.5708e-01, 2.4123e-01, 5.9823e-01, 8.5230e-02,
-1.3881e-02, 1.5465e+00, 2.6701e-02, -1.2181e+00,
-2.7483e-01, -1.5039e+00, 4.2858e-01, -1.9858e-01,
7.1848e-01, -1.8949e+00, -1.5626e-01, -1.5969e+00,
-2.4866e-01, 2.9512e-01, -4.5688e-01, -3.1831e-01,
8.1712e-01, -9.5078e-01, 1.4496e+00, 6.8175e-01,
-3.1556e-01, 7.2929e-01, 1.6336e+00, -5.8629e-01,
1.5029e+00, 1.0770e+00, 5.6100e-01, 2.1107e-01,
9.6809e-01, -1.0419e+00, -9.2013e-02, -1.2955e+00,
6.3738e-01, -5.7441e-01, 1.0233e+00, -2.5539e-01,
1.3183e+00, -7.6584e-01, -2.0945e-01, 6.1365e-02,
-1.1505e+00, 2.4718e-01, 1.1095e+00, -1.2793e+00,
-2.4920e+00, 4.2282e-01, -1.4824e+00, -2.0651e+00]),
size=(10000, 10000), nnz=1000, layout=torch.sparse_csr)
tensor([0.9650, 0.5575, 0.0929, ..., 0.2484, 0.1490, 0.4552])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 1000
Density: 1e-05
Time: 10.566636562347412 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_10000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 472596, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 5000, "MATRIX_DENSITY": 5e-05, "TIME_S": 10.733664512634277, "TIME_S_1KI": 0.02271213576211876, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 687.49480676651, "W": 65.37, "J_1KI": 1.4547199019173036, "W_1KI": 0.1383211030139908, "W_D": 45.042500000000004, "J_D": 473.7109504938126, "W_D_1KI": 0.09530867802520547, "J_D_1KI": 0.0002016705135574687}

16
pytorch/output_1core/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, ..., 4999, 4999, 5000]),
col_indices=tensor([2720, 7565, 9569, ..., 3581, 5033, 9559]),
values=tensor([-0.2858, 1.1812, -0.9537, ..., -0.6696, -0.1596,
1.4045]), size=(10000, 10000), nnz=5000,
layout=torch.sparse_csr)
tensor([0.5283, 0.9547, 0.2607, ..., 0.6208, 0.8258, 0.5120])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 5000
Density: 5e-05
Time: 10.733664512634277 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_20000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 77665, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 39998, "MATRIX_DENSITY": 9.9995e-05, "TIME_S": 10.419386148452759, "TIME_S_1KI": 0.13415806538920697, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 715.0278479003906, "W": 64.64, "J_1KI": 9.206564706114603, "W_1KI": 0.8322925384664907, "W_D": 44.5275, "J_D": 492.5495435857773, "W_D_1KI": 0.5733277538144597, "J_D_1KI": 0.007382060822950617}

16
pytorch/output_1core/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, 1, 2, ..., 39991, 39995, 39998]),
col_indices=tensor([11652, 13274, 207, ..., 7083, 12745, 14601]),
values=tensor([-0.6317, 1.0881, 0.8065, ..., -2.7134, -1.0640,
1.4791]), size=(20000, 20000), nnz=39998,
layout=torch.sparse_csr)
tensor([0.8018, 0.5866, 0.4831, ..., 0.3300, 0.2526, 0.0357])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 39998
Density: 9.9995e-05
Time: 10.419386148452759 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_20000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 343961, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 4000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.51590633392334, "TIME_S_1KI": 0.030572961277363826, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 682.592255680561, "W": 65.41, "J_1KI": 1.9845048004877328, "W_1KI": 0.19016690845764492, "W_D": 44.973749999999995, "J_D": 469.32783150762316, "W_D_1KI": 0.13075246902991908, "J_D_1KI": 0.00038013748369704434}

16
pytorch/output_1core/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, 1, ..., 4000, 4000, 4000]),
col_indices=tensor([15853, 14861, 12215, ..., 3910, 15182, 195]),
values=tensor([-0.0688, 0.2364, -0.1682, ..., 0.7875, 0.1623,
-0.0174]), size=(20000, 20000), nnz=4000,
layout=torch.sparse_csr)
tensor([0.0870, 0.4697, 0.0475, ..., 0.6658, 0.3278, 0.5072])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 4000
Density: 1e-05
Time: 10.51590633392334 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_20000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 94898, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 20000, "MATRIX_DENSITY": 5e-05, "TIME_S": 10.925030946731567, "TIME_S_1KI": 0.11512393250365201, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 615.267878255844, "W": 65.02, "J_1KI": 6.483465175829249, "W_1KI": 0.6851566945562604, "W_D": 45.30499999999999, "J_D": 428.70980043649666, "W_D_1KI": 0.47740732154523796, "J_D_1KI": 0.005030741654673839}

16
pytorch/output_1core/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, 2, 3, ..., 19999, 20000, 20000]),
col_indices=tensor([ 5331, 11927, 15794, ..., 14979, 9049, 4624]),
values=tensor([ 0.0049, -2.8806, 0.2868, ..., -0.5517, 0.2551,
0.4274]), size=(20000, 20000), nnz=20000,
layout=torch.sparse_csr)
tensor([0.4157, 0.0530, 0.1068, ..., 0.7666, 0.0297, 0.3868])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 20000
Density: 5e-05
Time: 10.925030946731567 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_50000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 19780, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 249992, "MATRIX_DENSITY": 9.99968e-05, "TIME_S": 10.32542872428894, "TIME_S_1KI": 0.5220135856566704, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 671.4097343873977, "W": 65.31, "J_1KI": 33.943869281466014, "W_1KI": 3.301820020222447, "W_D": 45.71, "J_D": 469.91485161304473, "W_D_1KI": 2.3109201213346817, "J_D_1KI": 0.11683114870246115}

17
pytorch/output_1core/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, 5, 10, ..., 249979, 249986,
249992]),
col_indices=tensor([ 7217, 18019, 22940, ..., 41080, 45621, 49188]),
values=tensor([ 1.3418, -1.2707, -0.5433, ..., -1.2530, 1.9430,
0.1875]), size=(50000, 50000), nnz=249992,
layout=torch.sparse_csr)
tensor([0.8358, 0.3669, 0.4208, ..., 0.9554, 0.6487, 0.8610])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 249992
Density: 9.99968e-05
Time: 10.32542872428894 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_50000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 43359, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 25000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.284875392913818, "TIME_S_1KI": 0.23720278126603053, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 638.9146102428437, "W": 64.12, "J_1KI": 14.735455389719403, "W_1KI": 1.4788163933670058, "W_D": 44.045, "J_D": 438.8801311314106, "W_D_1KI": 1.0158213980949746, "J_D_1KI": 0.023428155586959445}

16
pytorch/output_1core/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, 0, ..., 24997, 25000, 25000]),
col_indices=tensor([19994, 22049, 48468, ..., 8495, 16023, 32837]),
values=tensor([-1.8163, -0.0299, 0.1597, ..., -2.9027, 0.0447,
0.9203]), size=(50000, 50000), nnz=25000,
layout=torch.sparse_csr)
tensor([0.2325, 0.1458, 0.0013, ..., 0.5389, 0.4890, 0.3122])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 25000
Density: 1e-05
Time: 10.284875392913818 seconds

1
pytorch/output_1core/epyc_7313p_10_2_10_50000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Epyc 7313P", "ITERATIONS": 23583, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 124997, "MATRIX_DENSITY": 4.99988e-05, "TIME_S": 10.195623636245728, "TIME_S_1KI": 0.4323293743902696, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 656.70900785923, "W": 64.5, "J_1KI": 27.846711947556717, "W_1KI": 2.7350209896959674, "W_D": 44.925, "J_D": 457.40546012520787, "W_D_1KI": 1.9049739218928887, "J_D_1KI": 0.08077742110388367}

17
pytorch/output_1core/epyc_7313p_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, 0, 5, ..., 124989, 124993,
124997]),
col_indices=tensor([ 7870, 10492, 13171, ..., 12371, 39744, 40067]),
values=tensor([-0.1686, 0.9622, 0.3859, ..., -0.4609, -1.3867,
-0.0631]), size=(50000, 50000), nnz=124997,
layout=torch.sparse_csr)
tensor([0.9114, 0.0077, 0.0075, ..., 0.9886, 0.9267, 0.5022])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 124997
Density: 4.99988e-05
Time: 10.195623636245728 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_100000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 3828, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 999959, "MATRIX_DENSITY": 9.99959e-05, "TIME_S": 10.317766427993774, "TIME_S_1KI": 2.695341282130035, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 535.4964678955079, "W": 51.84, "J_1KI": 139.88935942933853, "W_1KI": 13.542319749216302, "W_D": 42.282500000000006, "J_D": 436.7694715237618, "W_D_1KI": 11.045585161964473, "J_D_1KI": 2.8854715679113045}

17
pytorch/output_1core/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, 11, 23, ..., 999934, 999944,
999959]),
col_indices=tensor([ 9492, 26542, 45572, ..., 85669, 93860, 99637]),
values=tensor([-0.5032, -0.4425, 1.2901, ..., 0.6217, -1.3072,
1.2570]), size=(100000, 100000), nnz=999959,
layout=torch.sparse_csr)
tensor([0.5683, 0.9148, 0.7465, ..., 0.9189, 0.9836, 0.0134])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 999959
Density: 9.99959e-05
Time: 10.317766427993774 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_100000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 10284, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 99999, "MATRIX_DENSITY": 9.9999e-06, "TIME_S": 10.279484272003174, "TIME_S_1KI": 0.9995608977054816, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 527.2581393003464, "W": 51.17000000000001, "J_1KI": 51.269752946358075, "W_1KI": 4.975690392843252, "W_D": 41.602500000000006, "J_D": 428.67415947318085, "W_D_1KI": 4.045361726954493, "J_D_1KI": 0.39336461755683516}

16
pytorch/output_1core/xeon_4216_10_2_10_100000_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, 2, ..., 99996, 99997, 99999]),
col_indices=tensor([53446, 60239, 57002, ..., 87293, 8614, 73293]),
values=tensor([-1.5600, -1.7444, 1.1495, ..., -0.1715, -1.3206,
-1.4367]), size=(100000, 100000), nnz=99999,
layout=torch.sparse_csr)
tensor([0.8858, 0.7738, 0.0915, ..., 0.6061, 0.1212, 0.2608])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 99999
Density: 9.9999e-06
Time: 10.279484272003174 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_100000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 5951, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [100000, 100000], "MATRIX_SIZE": 10000000000, "MATRIX_NNZ": 499988, "MATRIX_DENSITY": 4.99988e-05, "TIME_S": 10.432047128677368, "TIME_S_1KI": 1.7529906114396518, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 542.8448672485351, "W": 52.16, "J_1KI": 91.21910052907666, "W_1KI": 8.7649134599227, "W_D": 42.12625, "J_D": 438.42060178160665, "W_D_1KI": 7.078852293732146, "J_D_1KI": 1.1895231547188954}

17
pytorch/output_1core/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, 6, 11, ..., 499969, 499979,
499988]),
col_indices=tensor([22455, 23859, 29197, ..., 41507, 47228, 83381]),
values=tensor([ 0.8086, -1.3111, 0.7144, ..., -0.9283, 0.6276,
0.0423]), size=(100000, 100000), nnz=499988,
layout=torch.sparse_csr)
tensor([0.2048, 0.7838, 0.9595, ..., 0.6479, 0.2872, 0.9019])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([100000, 100000])
Size: 10000000000
NNZ: 499988
Density: 4.99988e-05
Time: 10.432047128677368 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_10000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 123377, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 10000, "MATRIX_DENSITY": 0.0001, "TIME_S": 10.433231830596924, "TIME_S_1KI": 0.08456383143208963, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 530.0696156978607, "W": 50.84, "J_1KI": 4.296340612090265, "W_1KI": 0.41207032104849367, "W_D": 41.282500000000006, "J_D": 430.4209069639445, "W_D_1KI": 0.33460450489151145, "J_D_1KI": 0.00271204928707548}

16
pytorch/output_1core/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, 1, 4, ..., 9998, 10000, 10000]),
col_indices=tensor([9176, 176, 6730, ..., 9592, 5827, 9675]),
values=tensor([-0.0157, -0.6431, -0.6454, ..., 0.0062, -1.2344,
0.4342]), size=(10000, 10000), nnz=10000,
layout=torch.sparse_csr)
tensor([0.7658, 0.4719, 0.0181, ..., 0.6359, 0.9564, 0.9840])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 10000
Density: 0.0001
Time: 10.433231830596924 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_10000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 356710, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 1000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.643239498138428, "TIME_S_1KI": 0.029837233321573346, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 540.0778507947922, "W": 50.71, "J_1KI": 1.5140530144789666, "W_1KI": 0.1421602982815172, "W_D": 41.230000000000004, "J_D": 439.1127940893174, "W_D_1KI": 0.11558408791455245, "J_D_1KI": 0.0003240281683007273}

375
pytorch/output_1core/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([3764, 7829, 5913, 9241, 2345, 6075, 4450, 6634, 8729,
4007, 3849, 5461, 1244, 8426, 5722, 9818, 3851, 4820,
8955, 4220, 698, 4092, 5428, 631, 7097, 3582, 7936,
6223, 7975, 9145, 6326, 8587, 2642, 8193, 4524, 5178,
2495, 6920, 2580, 7419, 3189, 3756, 4022, 8147, 9518,
5609, 3664, 4255, 1311, 9267, 4467, 5274, 9296, 8452,
3572, 4174, 4359, 3960, 5997, 9129, 5697, 3838, 1404,
3088, 5778, 3281, 8945, 6617, 3929, 9796, 5842, 3109,
8559, 8895, 6359, 9120, 4308, 1343, 6943, 5789, 8328,
9886, 3202, 2572, 8626, 6452, 2516, 1848, 1451, 7459,
6651, 4759, 2648, 9393, 6515, 9253, 267, 1581, 7821,
2207, 671, 5337, 2470, 7840, 7915, 4010, 2589, 6195,
2640, 1421, 4183, 7620, 5400, 604, 3940, 4042, 624,
5720, 5203, 7094, 6961, 499, 1406, 6808, 8312, 6060,
7588, 7742, 7125, 9988, 5570, 1265, 4122, 6703, 4557,
9657, 2068, 8184, 4926, 1561, 7756, 1800, 8838, 1861,
815, 4967, 4669, 8016, 447, 7317, 3036, 6757, 3847,
3055, 2173, 188, 7462, 3789, 4149, 5190, 4299, 8131,
390, 1021, 164, 1170, 1415, 4078, 8338, 9553, 8994,
65, 3860, 1816, 4194, 8861, 5329, 9880, 4465, 5778,
8974, 7706, 4292, 4804, 609, 943, 2924, 4134, 8858,
2158, 377, 801, 7058, 3537, 4386, 9438, 1331, 4640,
2631, 3988, 5171, 855, 4951, 6678, 8617, 6808, 3039,
4308, 1692, 2356, 4302, 5745, 7586, 1914, 6595, 8610,
3681, 7043, 8579, 7981, 8283, 2210, 9976, 5886, 5303,
9202, 3919, 4655, 2928, 5331, 4165, 306, 7537, 7451,
5430, 3476, 6265, 3679, 5183, 5551, 3046, 8931, 8738,
4189, 7326, 9922, 7770, 7459, 9344, 4573, 1422, 3604,
8270, 3552, 7597, 546, 3373, 6841, 9561, 9881, 2688,
5647, 1280, 5199, 3432, 8614, 3113, 5953, 3644, 1227,
8108, 5902, 6004, 7213, 4129, 4589, 1751, 8767, 8176,
7555, 8268, 2912, 1139, 9103, 5065, 5041, 8906, 2669,
2544, 2377, 3465, 5201, 7618, 4679, 4503, 9320, 4759,
5771, 5860, 7377, 3855, 5350, 7073, 6034, 488, 2255,
1341, 4300, 259, 4735, 9619, 2309, 7776, 5389, 2163,
8148, 9370, 2818, 6918, 2278, 753, 5061, 9138, 3006,
3059, 6683, 3386, 1918, 6800, 2762, 5642, 7519, 6890,
3370, 3622, 4661, 6938, 3902, 3648, 5127, 2576, 4358,
5595, 1129, 745, 3678, 3023, 1100, 2824, 4608, 3362,
6056, 358, 2003, 2783, 118, 2999, 2137, 3548, 7025,
5739, 5103, 8612, 6787, 8661, 1122, 6836, 1815, 3359,
4883, 116, 5784, 4117, 6706, 4254, 7449, 9064, 4499,
517, 3561, 6531, 7058, 2675, 8168, 3706, 9417, 2907,
4328, 6957, 1247, 448, 7714, 4076, 8646, 9255, 3220,
7414, 228, 9408, 2973, 7903, 1691, 3546, 7255, 15,
4152, 704, 6471, 1248, 7243, 9318, 7946, 9730, 7137,
7746, 4517, 2544, 6752, 3343, 3713, 3135, 2239, 6067,
3476, 1882, 7065, 855, 2006, 8695, 80, 2925, 8701,
263, 2227, 8914, 3479, 5990, 5647, 5236, 4096, 9626,
2076, 5557, 3661, 261, 4206, 6509, 3869, 2597, 7645,
8148, 4939, 6993, 9785, 1349, 3685, 4333, 6849, 1908,
5237, 4296, 4308, 4445, 1001, 3087, 7019, 1595, 1620,
2153, 8897, 5288, 4095, 4974, 9515, 1626, 8427, 145,
2236, 2609, 6428, 433, 1025, 4041, 176, 4130, 6920,
4052, 7037, 2701, 3729, 4427, 2888, 2755, 7887, 4983,
97, 3326, 4324, 1674, 9895, 3842, 458, 4220, 2663,
4299, 1295, 4964, 1610, 6603, 8268, 3294, 3912, 7422,
9381, 775, 3556, 9917, 9301, 8097, 1143, 227, 206,
6320, 3178, 9383, 6748, 6280, 4565, 6717, 9994, 3930,
1366, 6141, 4132, 1103, 9490, 6254, 4001, 4247, 7227,
463, 7939, 4758, 1069, 2621, 1191, 3937, 1174, 9167,
9390, 7656, 645, 1916, 8549, 7925, 3489, 8983, 4853,
7402, 2763, 5448, 846, 8825, 3332, 4755, 8946, 1089,
1370, 3543, 2244, 4061, 23, 5963, 4981, 8115, 7693,
3060, 4118, 1293, 5984, 8706, 2532, 9378, 6580, 591,
455, 7304, 6752, 8966, 330, 9348, 4058, 5126, 4197,
636, 8879, 9333, 9130, 1026, 6012, 6436, 178, 9833,
1378, 3336, 3968, 2292, 1574, 4072, 8136, 3630, 7471,
4387, 9097, 8945, 5039, 9864, 438, 8987, 5629, 6952,
5294, 6918, 9755, 7429, 4008, 7618, 7714, 2509, 7409,
2897, 2824, 5640, 6188, 3480, 6375, 3032, 8382, 1016,
6072, 9315, 3073, 9854, 7189, 8107, 7049, 121, 5691,
8261, 6343, 1229, 9403, 2821, 3992, 9818, 5631, 4269,
7641, 4626, 9087, 5283, 3317, 2813, 1707, 1758, 1223,
9098, 4578, 4571, 3558, 2371, 6055, 7195, 3077, 9853,
2136, 5311, 3241, 8441, 1234, 1454, 5967, 3583, 5776,
5921, 1710, 6395, 6139, 7269, 6978, 8283, 1221, 5784,
583, 4934, 7898, 496, 3621, 697, 7358, 5069, 7515,
138, 1268, 245, 8416, 3255, 6467, 8212, 3644, 7221,
126, 5426, 6604, 1395, 7063, 910, 9003, 6565, 6264,
2697, 189, 5692, 5058, 2177, 2088, 4972, 6867, 6055,
3889, 8142, 3840, 1060, 7207, 4836, 2827, 9097, 9816,
9509, 8802, 642, 5912, 2887, 6265, 5350, 3129, 2302,
8038, 6939, 1046, 7443, 1128, 1344, 3324, 7661, 8985,
498, 6847, 9630, 9990, 4173, 2983, 3350, 2343, 3675,
6623, 1096, 8888, 9877, 6732, 3778, 6464, 2627, 505,
4926, 4096, 8448, 1310, 5272, 2393, 7, 9574, 2900,
9065, 4430, 1622, 5945, 4700, 1891, 8690, 8418, 9223,
58, 7892, 1402, 166, 5220, 1108, 7488, 304, 2843,
9173, 3362, 3296, 4178, 2318, 3513, 3143, 5666, 4165,
5427, 9882, 4994, 9390, 4343, 8070, 5528, 6101, 2222,
7217, 4718, 9885, 4761, 6408, 1836, 4206, 5903, 4968,
9804, 9345, 7387, 1130, 9497, 1855, 4282, 1497, 9171,
9854, 3070, 2659, 8019, 9442, 1391, 3191, 7726, 3159,
9344, 5473, 5142, 2610, 6977, 4194, 334, 1559, 2591,
2247, 7909, 6959, 7927, 337, 3115, 9599, 7867, 6727,
143, 4730, 4901, 8232, 239, 2629, 72, 5088, 1382,
571, 5240, 7369, 3615, 6758, 3116, 9669, 9906, 3161,
4861, 7839, 8726, 2805, 3045, 7277, 1527, 5898, 9256,
6585, 2531, 105, 4628, 374, 6324, 3745, 8390, 805,
7562, 2184, 7832, 856, 1300, 4844, 3084, 3959, 9679,
6202, 584, 7435, 2626, 8243, 1632, 635, 7504, 6564,
8676, 4636, 8055, 965, 196, 1341, 6820, 7122, 2278,
8094, 6325, 4210, 7787, 7998, 4191, 9170, 6035, 1771,
2349, 5270, 3823, 3444, 2047, 2381, 6764, 6747, 4933,
6996, 7923, 1729, 7797, 8572, 5641, 2749, 1583, 8124,
5956, 5899, 9446, 3606, 4193, 2994, 8766, 3364, 6391,
7078, 3195, 674, 7905, 1793, 1965, 3891, 2423, 1657,
2827, 1031, 6090, 7336, 551, 1139, 6261, 3163, 6332,
1374, 1290, 2894, 8781, 5226, 5810, 263, 9903, 9037,
2367, 4768, 3073, 1292, 2018, 829, 2614, 1788, 5705,
272]),
values=tensor([-1.1946e+00, 1.0241e+00, -2.6246e+00, 2.2120e+00,
1.6317e+00, 1.1250e+00, 3.0840e-01, -1.3677e+00,
-1.5256e-02, -4.8276e-01, 2.1175e-01, 2.8014e-01,
-1.2104e+00, 3.7719e-01, 2.1108e-01, 7.4261e-02,
-1.6512e+00, -3.8696e-01, 8.6402e-01, -1.2954e+00,
-2.5758e+00, -1.7301e-01, -1.5582e+00, 1.2871e-02,
-1.7587e+00, 1.2048e+00, -4.1745e-03, 2.5752e+00,
7.5364e-01, -4.1424e-01, -1.1954e+00, -9.6286e-01,
3.9663e-01, 2.4725e+00, 2.5615e+00, 7.2358e-01,
9.2695e-01, 2.0094e+00, -1.7622e+00, 8.0226e-01,
-1.1973e+00, 2.1328e+00, -7.3614e-01, 1.7703e+00,
-7.3179e-01, -6.8291e-01, 1.2366e+00, -2.4823e-01,
1.0453e+00, -2.8247e-01, -1.2582e+00, -5.9235e-01,
1.1136e-02, -4.7006e-01, 2.1866e-01, 3.2844e-01,
-9.4612e-01, 4.5357e-01, -7.9006e-01, -3.9919e-01,
1.5186e+00, 7.0875e-01, -2.9790e-04, -4.3449e-01,
6.4513e-01, 2.1426e+00, -1.7150e-01, -1.1076e+00,
1.7367e+00, 2.3431e-02, 2.1602e-01, -1.0393e+00,
1.4506e+00, -1.2431e+00, -3.0248e-01, -2.1092e+00,
1.5054e+00, 1.2465e-01, -1.3473e-01, 2.3430e+00,
5.6974e-01, -7.3644e-01, -2.7770e-01, 1.9890e+00,
-2.4426e-01, 5.2589e-01, -1.3429e-01, -5.5741e-01,
-1.1784e+00, -6.9101e-01, 1.0530e+00, 4.4348e-01,
-2.5901e-01, 1.0358e+00, -1.2586e+00, -1.1275e+00,
-1.4416e+00, 1.4757e+00, 1.6271e+00, -2.4097e-03,
-1.6191e-01, -5.2795e-01, -6.4157e-02, -1.0381e+00,
-7.2013e-01, -5.0349e-01, -3.5789e-01, 8.8352e-01,
4.6719e-01, -1.4023e+00, 1.3586e+00, 3.1369e-02,
-2.0799e-01, 8.6935e-01, 5.1592e-01, -1.0419e+00,
-5.8210e-01, 3.8923e-01, 7.9399e-03, 1.1945e+00,
-9.2725e-01, 2.1726e-02, 1.0525e-01, -8.9964e-01,
4.0951e-01, -9.3387e-01, 1.1991e-01, 8.0212e-01,
-7.4922e-01, 9.8479e-01, -1.3579e+00, -7.5522e-01,
1.8265e-01, 2.9629e-01, 1.5879e-01, 2.1822e-01,
-3.5851e-01, 7.1753e-01, 1.6158e+00, 1.7649e-01,
7.4674e-01, 7.4593e-01, 1.2604e+00, -5.3425e-01,
1.6179e+00, -9.4536e-02, -1.1019e+00, 1.0737e+00,
-1.6092e-01, -2.7323e-01, 5.2809e-01, -2.5319e-01,
-3.2856e-02, -5.4473e-01, -1.7609e+00, -9.9842e-01,
-8.1607e-01, 8.6284e-01, 1.3279e-01, 9.2607e-01,
-9.1620e-01, 3.5447e-03, -1.5657e-01, 5.6398e-01,
-2.1546e-01, 7.8071e-02, 3.0247e-01, 1.1885e+00,
2.1798e-01, 8.3471e-01, 6.3038e-01, -3.3617e-01,
-1.6138e+00, -1.2121e+00, -6.3585e-01, 9.5982e-01,
-1.2351e+00, 6.8745e-01, 2.0249e-01, -1.8767e+00,
-8.1044e-01, 7.0050e-02, 1.7490e+00, 5.6094e-01,
-3.2183e-01, 1.2508e+00, -4.9129e-01, 7.0914e-01,
1.4307e-01, -1.7606e-01, 2.4075e-01, 1.2946e+00,
1.9927e+00, -1.4919e+00, -2.1427e+00, 2.0905e-01,
1.2361e+00, 1.7221e+00, -1.3142e-02, -1.2153e+00,
9.5494e-02, -4.5621e-01, -9.7498e-01, -3.1674e-01,
-3.9258e-01, 1.0503e+00, 1.2826e-01, 5.1997e-01,
-2.9445e-01, -3.8412e-01, 1.1762e+00, -4.0353e-01,
-8.8617e-01, -5.1915e-01, 1.5131e+00, -4.6426e-01,
2.5079e-01, -7.4668e-02, -8.4378e-01, -1.4111e-01,
2.9942e-01, -7.1915e-01, 1.5935e-01, -1.7103e+00,
-1.3010e+00, -5.8621e-01, 7.7494e-01, -8.7947e-01,
3.0187e-01, -9.1227e-01, 2.1805e-01, -8.4239e-01,
6.4713e-01, -1.9645e+00, 6.2310e-01, 5.4112e-01,
-1.4813e+00, 8.0613e-01, -1.2853e+00, -1.1351e+00,
-4.8844e-01, -1.9266e+00, -8.3081e-02, 1.4881e+00,
-7.0505e-01, 6.7337e-01, 1.2632e+00, 8.2593e-01,
-9.7455e-01, 7.4966e-01, 6.8899e-01, -1.1204e+00,
1.3517e+00, -5.4740e-01, -1.5907e+00, -2.1423e+00,
1.3721e+00, 1.1839e+00, 6.1736e-01, -1.0136e+00,
-9.3219e-01, -1.4893e+00, 1.2092e+00, 1.2451e+00,
5.1533e-01, -1.3934e+00, -5.0302e-01, -6.5601e-01,
-2.6717e+00, -4.1049e-01, -1.3137e+00, -3.2452e-01,
-1.5398e+00, 8.7905e-01, -3.1866e-02, 4.7459e-01,
1.0482e+00, -1.1775e+00, 2.9826e-01, -9.5396e-01,
1.3574e+00, -5.2503e-02, 1.7189e-01, 8.6699e-01,
-1.1462e+00, -1.7903e+00, -3.4916e-01, -1.2466e+00,
-3.0682e+00, 2.9241e+00, -4.1416e-01, -3.1102e-02,
-1.8185e+00, -5.7250e-01, -2.1144e+00, -5.5245e-01,
1.2785e+00, -6.1769e-01, -1.1974e-01, -9.9129e-01,
4.5799e-01, 7.9548e-01, 3.0571e-01, 4.1981e-01,
-2.5307e+00, 1.3986e-01, -3.6424e-02, 3.8326e-01,
6.7315e-02, 1.3289e+00, -5.5261e-02, -8.9057e-01,
-8.0899e-02, 1.9045e+00, 3.5334e-01, 1.2352e+00,
1.7655e-01, 1.3114e+00, 3.0444e-01, -5.6223e-01,
3.1563e-01, 1.8755e-01, 1.8943e+00, -3.5483e-02,
1.4423e-01, 1.2998e-01, 2.1779e+00, -1.5407e+00,
1.4057e+00, 1.3072e+00, 1.3040e+00, -2.4900e-01,
-5.7484e-01, 3.6292e-01, 3.9165e-02, 1.1487e+00,
4.7314e-01, -3.0886e-02, 9.1366e-01, 1.0233e+00,
-1.2785e+00, -4.8579e-01, 7.3396e-01, -4.7937e-01,
9.9721e-02, -9.4828e-01, -1.1561e+00, 2.0921e-01,
1.1042e+00, -1.4290e+00, -4.2525e-01, 1.0073e+00,
-6.9314e-01, -8.8086e-01, 2.2425e-01, 1.7684e+00,
-3.7663e-01, -1.8295e-01, -1.9686e-01, 5.1750e-01,
1.1877e-01, -2.5534e-01, 6.0006e-01, -4.9048e-02,
-1.6212e-01, 6.5580e-02, -9.6844e-01, -6.5921e-01,
1.2298e+00, -1.8196e-01, -1.6631e+00, 1.0352e+00,
8.9681e-01, 8.2679e-01, -1.0280e+00, 3.2908e-01,
4.3520e-01, -7.0476e-01, -3.7922e-01, 1.0518e+00,
5.9101e-01, 1.2338e+00, -1.8392e+00, -1.3593e+00,
-1.0708e-01, -4.8953e-01, -8.0555e-01, 3.9991e-01,
-9.1285e-01, 8.6847e-01, 1.0606e+00, -4.9274e-01,
4.5981e-01, 1.6834e+00, -9.8807e-01, 2.6967e-01,
4.3776e-01, -4.8359e-01, 4.5406e-01, 4.0696e-01,
-1.0907e+00, 6.8932e-01, 1.0888e+00, 1.8094e+00,
-1.4869e-01, 4.1970e-01, -1.8624e+00, 1.0929e+00,
-9.3750e-01, -7.6250e-02, 1.0046e-01, 6.7658e-01,
-4.2577e-01, -1.0919e+00, 4.7995e-01, 1.0179e-02,
4.8813e-01, -8.3874e-01, -4.7199e-01, 1.6579e+00,
9.9249e-01, 3.8174e-01, -7.1392e-01, 1.0164e+00,
-1.5815e+00, 1.1763e+00, -1.3188e+00, 1.3245e-01,
-2.9018e-01, 8.3135e-01, 1.0943e+00, 8.2248e-01,
4.5446e-01, -9.6561e-01, -5.8614e-02, -2.9006e-01,
5.2021e-01, -6.0390e-01, 3.0321e-01, 4.5954e-01,
7.2848e-01, 1.3785e+00, 2.1936e+00, -1.0706e+00,
-8.0374e-01, 5.1213e-01, -6.6987e-01, -9.4933e-01,
-5.9791e-01, 5.1065e-01, -7.0811e-01, 2.3308e-01,
8.5605e-01, -3.0330e-02, 1.5928e-01, -1.4606e+00,
7.3862e-01, 4.3125e-01, -1.3411e+00, 1.8326e+00,
1.8568e+00, 2.8493e-01, -1.6747e+00, -3.2247e-01,
9.3073e-01, -1.7680e-01, -2.2606e-01, -6.5106e-02,
-5.0144e-01, -2.0640e+00, -1.0589e+00, -2.2046e+00,
-1.9990e-01, -1.1351e+00, 6.2866e-01, 1.4880e+00,
-3.3857e-01, 1.5668e+00, 2.1422e+00, -3.4383e-01,
-2.4342e+00, -1.2439e+00, -8.0512e-01, -3.2559e+00,
-6.3807e-01, 1.1760e-01, 3.4460e-02, -5.3086e-01,
2.8369e-01, 6.5826e-01, 3.7609e-01, 2.8028e+00,
-3.8766e-01, 1.6431e+00, -2.0741e+00, 1.0597e+00,
6.6386e-01, -5.2328e-01, 3.3672e-01, 2.7480e-01,
-1.8608e+00, -6.9663e-01, -1.6192e+00, 7.2563e-01,
-1.1524e+00, 1.5168e+00, -2.8304e-02, 9.8365e-01,
-9.4490e-01, 5.6756e-01, 2.0809e-01, -9.7573e-02,
-2.1137e+00, -6.4979e-01, 8.2734e-01, 3.8077e-01,
2.5612e-01, -7.6662e-01, -5.5603e-01, -4.0522e-01,
5.1989e-01, 1.5444e-01, 1.8704e+00, 7.1280e-01,
1.1467e+00, 5.9627e-01, -7.3921e-01, -3.4968e-01,
1.8599e+00, -1.1041e+00, -2.3853e-01, -4.3193e-01,
2.3111e-01, 5.9705e-01, -9.5238e-01, 7.3269e-01,
1.3038e-01, 1.7601e-01, -2.9576e-01, -6.4016e-01,
-6.7409e-01, -1.8346e+00, 8.3341e-01, -9.7870e-01,
-1.4306e+00, -9.4199e-01, 1.2516e+00, 1.0657e+00,
-1.2150e-01, -6.2417e-01, -6.4439e-01, -9.3574e-01,
-9.6785e-02, -1.6706e+00, -2.2785e-01, -1.4566e+00,
-2.5150e-01, -3.6753e-01, 9.4414e-01, 1.6085e+00,
1.0900e-01, -1.2137e-01, -4.1640e-01, 1.3739e-01,
-2.2474e+00, 1.6401e+00, -1.5782e+00, 1.0885e+00,
-9.6149e-01, -1.9913e+00, -4.5521e-01, 5.0405e-01,
-9.6981e-01, 1.5174e+00, -3.0409e-01, 1.7048e-02,
1.6476e+00, -6.6048e-01, -1.3568e+00, 3.5049e-01,
-6.4467e-01, -1.3988e+00, -4.2626e-01, 7.0339e-01,
1.4927e+00, 1.2195e+00, 6.9660e-01, -8.4541e-01,
-1.5827e+00, 3.4529e-01, -7.5112e-02, -1.1637e-01,
1.5122e+00, -2.6272e+00, 1.2840e-02, 7.6793e-01,
-8.6641e-02, 1.1255e+00, -2.4018e+00, 9.8756e-01,
-2.9810e-01, 2.9796e-01, 2.7212e-01, -9.7553e-01,
5.0748e-02, 7.6478e-01, -9.1160e-01, -9.9771e-01,
2.4188e-01, 8.9348e-01, 1.3442e+00, -7.2805e-01,
1.1476e+00, 1.8307e+00, -7.6564e-01, -6.6653e-01,
-7.6550e-01, 1.0272e+00, -3.7147e-01, 6.7420e-01,
6.8468e-01, 3.2785e-01, 2.0223e-02, -4.4411e-01,
-5.1191e-01, 1.8423e-01, 2.8611e-01, -3.4461e-01,
-4.4585e-01, 1.9783e+00, -1.0879e+00, -5.6692e-01,
-7.2405e-01, 1.5254e+00, 1.2653e+00, -9.3838e-01,
1.0988e+00, 3.0421e+00, -1.1624e-01, 1.8377e+00,
-9.4519e-01, -7.4331e-01, -1.0865e+00, 6.6719e-01,
3.1974e-01, -5.3476e-01, -2.1053e-02, -2.4812e-01,
-3.9560e-01, 1.1778e+00, 3.7759e-01, -8.1591e-02,
1.5402e+00, -1.9674e+00, -1.0467e+00, 7.5852e-01,
-1.0396e-01, -6.1871e-01, 2.4630e-01, 1.0236e-02,
-4.5390e-01, 7.4653e-01, -1.3116e+00, 3.6403e-01,
-2.5249e+00, 9.0032e-01, 1.5520e+00, -8.0951e-01,
9.2712e-01, -1.6440e+00, 5.2229e-01, 7.2525e-01,
9.6999e-01, -9.3643e-01, 1.4146e-01, 2.9098e-01,
5.3350e-01, 2.5990e-01, -1.7270e+00, -7.0666e-02,
9.0025e-01, 1.3825e+00, 1.3886e-01, 2.7108e-02,
-7.0772e-01, -1.1485e+00, 2.7389e-01, 3.9894e-01,
-1.0489e+00, 6.7750e-01, -8.9123e-01, -2.7301e+00,
-1.1917e+00, -9.5536e-02, 1.6926e+00, 4.8504e-02,
-1.9987e+00, 5.2862e-01, 9.8830e-01, 3.9250e-01,
-3.7450e-01, -2.3399e-02, -8.5238e-02, 4.3510e-01,
1.0814e-01, -1.2999e-01, 5.1889e-01, -9.0872e-01,
-9.3853e-01, 2.6634e-01, 4.3744e-01, 6.1139e-01,
2.2371e+00, -3.6886e-01, 5.9552e-01, 6.1600e-01,
-1.5865e+00, -3.8211e-01, 2.4326e-01, 8.8552e-01,
-1.0340e+00, 1.8068e-02, -1.9139e+00, -3.9426e-01,
-9.7648e-01, -6.0806e-01, 2.1936e+00, -4.4606e-01,
1.5602e+00, 2.9392e-01, -3.0292e+00, 1.1875e+00,
-2.5792e-02, 1.2425e+00, -7.6052e-01, -6.8854e-01,
1.3156e+00, 6.1663e-01, 1.6821e+00, -9.1207e-01,
1.9710e-01, -5.6679e-01, -1.0562e+00, 2.4669e-01,
1.2825e+00, -4.3459e-01, -2.7151e-01, 6.6350e-01,
-4.3323e-01, 5.5758e-02, 1.5185e-01, 1.2906e-01,
5.5407e-01, -1.4261e+00, 2.2622e-01, 3.7091e-01,
4.1306e-01, -2.3203e+00, -8.3359e-01, 1.7636e+00,
-1.5495e+00, 6.6051e-01, 5.6354e-01, -1.9527e-01,
8.5107e-01, -2.3931e+00, -2.3421e-01, 1.1888e+00,
-1.0594e+00, 6.2851e-01, -1.3775e+00, -6.5201e-02,
3.0458e-01, 1.1636e+00, -6.0922e-01, 2.0854e+00,
1.5355e-01, -7.4531e-01, -1.8329e+00, 6.0452e-01,
-2.5297e+00, -8.2483e-01, -8.8189e-01, 6.8193e-01,
1.6612e+00, 2.2647e+00, -5.7919e-01, -1.1492e+00,
1.5298e-01, 3.8167e-01, 6.5361e-02, -8.1700e-01,
-1.5235e-01, -3.7881e-01, -1.0060e+00, -2.1513e-01,
-1.1122e+00, 8.0422e-01, 5.7555e-01, 6.6217e-01,
5.0129e-01, 8.7807e-01, 9.5846e-01, -5.0699e-01,
3.7605e-01, 1.6987e+00, 5.3857e-03, 2.5597e-01,
-1.3209e+00, 8.0491e-01, -2.8251e+00, -1.0559e+00,
1.3500e-01, 5.3869e-01, 1.0596e+00, -5.7831e-02,
-9.5305e-01, -3.1547e-01, 7.6320e-01, 1.1009e-01,
-6.0412e-01, 9.0778e-01, -1.3630e+00, 3.0187e-01,
-1.1741e+00, -2.1583e+00, -6.8394e-02, -9.4370e-01,
1.3844e+00, 3.1456e-01, 4.0980e-01, 1.9587e+00,
-6.9097e-01, -9.3789e-01, 1.4375e+00, 7.0058e-01,
2.8419e+00, 1.1875e+00, -8.4268e-01, -2.4446e-01,
-1.1808e+00, -1.0691e-01, 2.4419e+00, 4.1647e-02,
-4.5697e-01, -9.8667e-02, -3.0512e-01, -6.7193e-01,
-9.5333e-01, 2.7987e-01, 6.4956e-01, 1.0922e+00,
8.1288e-01, 1.8410e+00, -8.8393e-01, -5.2649e-01,
1.0470e+00, 1.1594e+00, -1.2706e-01, 3.7100e-01,
-6.6160e-01, 3.3514e-01, 8.6161e-01, -1.1329e+00,
8.7563e-01, 6.3428e-01, -1.2725e+00, 2.2057e-01,
5.4102e-01, -9.3567e-01, -5.0721e-01, -1.1695e+00,
-2.9957e-01, 2.3472e+00, 1.9165e+00, 2.1309e-01,
2.2340e-01, -8.6680e-01, -1.5443e+00, 5.7678e-01,
2.5506e-01, 7.4844e-01, 1.3575e+00, 9.3532e-01,
6.0210e-01, -1.1870e-01, 6.2202e-01, 4.3340e-01,
-1.3384e+00, 2.7305e+00, 1.7295e+00, 9.4027e-01,
-1.0183e+00, 1.9346e+00, 1.8115e-01, 3.2015e-03,
1.6801e-01, 7.5525e-02, -8.2961e-01, -1.5448e+00,
1.2715e+00, 1.5713e+00, 1.3393e+00, -4.3901e-01,
3.9984e-01, -5.8021e-01, 4.6471e-01, 1.2497e-01,
1.9665e+00, -1.5795e+00, 1.9053e-01, 1.0708e+00,
1.0197e+00, -7.8543e-02, 4.5320e-01, -1.2027e+00,
1.2633e+00, 1.6214e+00, 2.4617e+00, 5.4668e-01,
-9.8773e-02, 3.1133e-01, 1.3334e+00, 7.7131e-01,
6.3079e-01, 8.5653e-01, 2.4144e-03, -9.3452e-01,
-4.2948e-01, -2.9245e-01, -1.0416e-01, 7.8610e-01,
-5.2402e-01, 3.7801e-01, 1.8903e+00, 8.0435e-01,
9.1606e-01, 5.9666e-01, -1.8585e-01, 8.5730e-01,
6.1173e-01, 8.2100e-01, -2.2582e-01, 1.4613e+00,
3.0393e-01, -4.9605e-01, -1.0132e+00, -8.1873e-01,
8.5322e-01, 6.3965e-01, 4.5412e-01, 4.7523e-01,
-7.7331e-01, -3.6932e-01, -2.1503e-01, -6.0607e-01,
-6.1219e-01, -4.9750e-01, 1.2487e+00, 2.3770e-01,
1.4365e-01, -9.1987e-01, 4.0770e-01, -1.7777e+00,
1.9846e-01, 4.1188e-01, -1.2347e-02, 3.3137e-01,
5.3932e-02, 6.4529e-02, 8.3614e-01, 5.3003e-01,
5.0183e-01, 1.0101e+00, 4.0212e-01, 1.0008e+00,
7.1749e-01, 1.6729e+00, 1.9496e-01, 7.4977e-01,
4.8851e-01, -8.5023e-02, -2.0530e-01, -3.4973e-01,
-4.7546e-01, -1.1011e-01, 8.0335e-02, -1.1475e+00,
-1.6759e-01, -1.1628e+00, 1.5434e+00, 1.4420e-01,
-7.7785e-01, -9.7057e-01, 1.3429e-01, 9.6382e-01,
3.6604e-01, -1.0984e+00, -1.8589e+00, -1.4024e+00]),
size=(10000, 10000), nnz=1000, layout=torch.sparse_csr)
tensor([0.2347, 0.9244, 0.7007, ..., 0.7546, 0.8955, 0.5807])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 1000
Density: 1e-05
Time: 10.643239498138428 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_10000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 162988, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [10000, 10000], "MATRIX_SIZE": 100000000, "MATRIX_NNZ": 4999, "MATRIX_DENSITY": 4.999e-05, "TIME_S": 10.489216327667236, "TIME_S_1KI": 0.06435575826237046, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 532.5132316040992, "W": 50.29, "J_1KI": 3.2671928706659337, "W_1KI": 0.30855032272314525, "W_D": 40.7775, "J_D": 431.7868025797606, "W_D_1KI": 0.250187130340884, "J_D_1KI": 0.0015350033765730241}

16
pytorch/output_1core/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, 4998, 4999]),
col_indices=tensor([ 135, 4852, 7675, ..., 8242, 9510, 5080]),
values=tensor([-0.5492, 1.2472, 0.2842, ..., 0.5096, 1.1862,
-0.3033]), size=(10000, 10000), nnz=4999,
layout=torch.sparse_csr)
tensor([0.1525, 0.9434, 0.8321, ..., 0.0657, 0.5857, 0.7418])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([10000, 10000])
Size: 100000000
NNZ: 4999
Density: 4.999e-05
Time: 10.489216327667236 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_20000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 44363, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 39998, "MATRIX_DENSITY": 9.9995e-05, "TIME_S": 10.488017559051514, "TIME_S_1KI": 0.23641362304288516, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 526.0746799468994, "W": 49.760000000000005, "J_1KI": 11.858410836663422, "W_1KI": 1.1216554335820392, "W_D": 40.27875, "J_D": 425.8366261035204, "W_D_1KI": 0.9079356671099791, "J_D_1KI": 0.020466056558618197}

16
pytorch/output_1core/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, 2, 2, ..., 39993, 39996, 39998]),
col_indices=tensor([12560, 17978, 12094, ..., 7541, 14037, 17712]),
values=tensor([ 0.0578, 0.0703, -0.7576, ..., 0.5104, -1.0726,
-0.6932]), size=(20000, 20000), nnz=39998,
layout=torch.sparse_csr)
tensor([0.2932, 0.3582, 0.9907, ..., 0.9728, 0.2484, 0.8066])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 39998
Density: 9.9995e-05
Time: 10.488017559051514 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_20000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 144303, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 4000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.480739116668701, "TIME_S_1KI": 0.0726300847291373, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 531.7523377656936, "W": 50.89999999999999, "J_1KI": 3.6849707751446164, "W_1KI": 0.3527300194729146, "W_D": 41.34374999999999, "J_D": 431.91818692535156, "W_D_1KI": 0.2865065175360179, "J_D_1KI": 0.001985450874451799}

16
pytorch/output_1core/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, 0, ..., 4000, 4000, 4000]),
col_indices=tensor([ 9066, 345, 11126, ..., 9719, 6530, 16548]),
values=tensor([-1.5343, 0.4773, -0.7801, ..., 2.3084, 1.9114,
0.7099]), size=(20000, 20000), nnz=4000,
layout=torch.sparse_csr)
tensor([0.8886, 0.9459, 0.9102, ..., 0.1910, 0.7609, 0.2804])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 4000
Density: 1e-05
Time: 10.480739116668701 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_20000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 58112, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [20000, 20000], "MATRIX_SIZE": 400000000, "MATRIX_NNZ": 20000, "MATRIX_DENSITY": 5e-05, "TIME_S": 10.547230005264282, "TIME_S_1KI": 0.18149831369191013, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 535.8703639006615, "W": 50.89, "J_1KI": 9.221337484524048, "W_1KI": 0.875722742290749, "W_D": 41.3525, "J_D": 435.44073930442335, "W_D_1KI": 0.7116000137665198, "J_D_1KI": 0.012245319620156247}

16
pytorch/output_1core/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, 0, 2, ..., 19997, 19999, 20000]),
col_indices=tensor([ 4689, 12751, 5485, ..., 4694, 6467, 17055]),
values=tensor([-1.2647, 1.0456, -0.0576, ..., -1.0665, 0.0821,
-2.2501]), size=(20000, 20000), nnz=20000,
layout=torch.sparse_csr)
tensor([0.6014, 0.6440, 0.5127, ..., 0.6380, 0.5224, 0.6478])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([20000, 20000])
Size: 400000000
NNZ: 20000
Density: 5e-05
Time: 10.547230005264282 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_50000_0.0001.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 11345, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 249975, "MATRIX_DENSITY": 9.999e-05, "TIME_S": 10.375968217849731, "TIME_S_1KI": 0.9145851227721227, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 541.7785520744324, "W": 52.04, "J_1KI": 47.754830504577555, "W_1KI": 4.587042750110181, "W_D": 42.4325, "J_D": 441.75669505953783, "W_D_1KI": 3.7401939180255614, "J_D_1KI": 0.3296777362737383}

17
pytorch/output_1core/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, 6, 11, ..., 249969, 249974,
249975]),
col_indices=tensor([ 2591, 11540, 17691, ..., 44394, 47863, 9441]),
values=tensor([ 0.8221, 0.8133, 1.6367, ..., -0.4455, -0.0184,
-0.1335]), size=(50000, 50000), nnz=249975,
layout=torch.sparse_csr)
tensor([0.6937, 0.1230, 0.4662, ..., 0.1698, 0.0658, 0.6889])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 249975
Density: 9.999e-05
Time: 10.375968217849731 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_50000_1e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 26674, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 25000, "MATRIX_DENSITY": 1e-05, "TIME_S": 10.05628752708435, "TIME_S_1KI": 0.37700710531170245, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 509.4300830769539, "W": 51.03, "J_1KI": 19.09837606196873, "W_1KI": 1.9130988978031043, "W_D": 41.565, "J_D": 414.94143451094624, "W_D_1KI": 1.5582589787808352, "J_D_1KI": 0.05841864657647279}

16
pytorch/output_1core/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([18289, 816, 22022, ..., 17006, 7802, 41272]),
values=tensor([-1.3405, 1.6567, -0.4947, ..., 0.7138, 1.3141,
-1.5291]), size=(50000, 50000), nnz=25000,
layout=torch.sparse_csr)
tensor([0.1595, 0.4181, 0.5988, ..., 0.7340, 0.2989, 0.4476])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 25000
Density: 1e-05
Time: 10.05628752708435 seconds

1
pytorch/output_1core/xeon_4216_10_2_10_50000_5e-05.json Normal file
View File

@ -0,0 +1 @@
{"CPU": "Xeon 4216", "ITERATIONS": 14264, "MATRIX_TYPE": "synthetic", "MATRIX_FORMAT": "csr", "MATRIX_SHAPE": [50000, 50000], "MATRIX_SIZE": 2500000000, "MATRIX_NNZ": 124996, "MATRIX_DENSITY": 4.99984e-05, "TIME_S": 10.336601495742798, "TIME_S_1KI": 0.724663593363909, "BASELINE_TIME_S": 2, "BASELINE_DELAY_S": 10, "J": 533.7157925939559, "W": 51.46999999999999, "J_1KI": 37.41697929009786, "W_1KI": 3.608384744812114, "W_D": 41.99499999999999, "J_D": 435.46521682500827, "W_D_1KI": 2.944125070106561, "J_D_1KI": 0.20640248668722386}

17
pytorch/output_1core/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, 3, 9, ..., 124991, 124994,
124996]),
col_indices=tensor([ 2395, 14429, 23023, ..., 39947, 32304, 46209]),
values=tensor([-0.0601, -1.0010, -0.0844, ..., 0.6539, -0.7041,
2.0265]), size=(50000, 50000), nnz=124996,
layout=torch.sparse_csr)
tensor([0.0956, 0.8159, 0.2718, ..., 0.1492, 0.0329, 0.4158])
Matrix: synthetic
Matrix: csr
Shape: torch.Size([50000, 50000])
Size: 2500000000
NNZ: 124996
Density: 4.99984e-05
Time: 10.336601495742798 seconds