/dports/misc/mxnet/incubator-mxnet-1.9.0/3rdparty/tvm/tests/python/frontend/pytorch/ |
H A D | test_forward.py | 24 import torch 510 end = torch.add(torch.tensor(4), 1) 511 return torch.arange(end) + torch.ones((5,), dtype=torch.int64) 515 end = torch.add(torch.tensor(4.0), torch.tensor(1.0)) 516 return torch.arange(end) + torch.ones((5,), dtype=torch.float) 521 end = torch.add(torch.tensor(4), 1) 530 step = torch.add(torch.tensor(2.5), torch.tensor(4.1)) 562 y = torch.add(torch.tensor(5, dtype=torch.float32), 1) 1130 d1 = torch.tensor(3) * torch.tensor(10) * torch.tensor(10) 2959 for dt in [torch.int32, torch.int64, torch.double]: [all …]
|
/dports/misc/glow/glow-f24d960e3cc80db95ac0bc17b1900dbf60ca044a/torch_glow/tests/nodes/ |
H A D | quantized_conv2d_test.py | 5 import torch 14 qu = torch.nn.quantized.Quantize(1 / 16, 0, torch.quint8) 15 qi = torch.nn.quantized.Quantize(1 / 16, 0, torch.qint8) 30 b_zero = torch.zeros(1) 31 b = torch.randn(1) 60 x = torch.tensor(range(5), dtype=torch.float) 64 q = torch.nn.quantized.Quantize(0.1, 2, torch.quint8) 93 x = torch.tensor(range(5), dtype=torch.float) 97 q = torch.nn.quantized.Quantize(0.1, 2, torch.quint8) 128 with torch.no_grad(): [all …]
|
H A D | quantized_conv3d_test.py | 5 import torch 19 qu = torch.nn.quantized.Quantize(1 / 16, 0, torch.quint8) 20 qi = torch.nn.quantized.Quantize(1 / 16, 0, torch.qint8) 35 b_zero = torch.zeros(1) 36 b = torch.randn(1) 65 x = torch.tensor(range(5), dtype=torch.float) 67 x = torch.cat((x, x, x)) 70 q = torch.nn.quantized.Quantize(0.1, 2, torch.quint8) 99 x = torch.tensor(range(5), dtype=torch.float) 104 q = torch.nn.quantized.Quantize(0.1, 2, torch.quint8) [all …]
|
H A D | quantized_linear_test.py | 3 import torch 13 q = torch.nn.quantized.Quantize( 17 linear = torch.nn.Linear(5, 5) 27 x = torch.tensor(range(5), dtype=torch.float) 28 x = torch.cat((x, x, x, x, x)) 29 x = torch.reshape(x, [5, 5]) 48 linear = torch.nn.Linear(5, 5) 58 x = torch.tensor(range(5), dtype=torch.float) 60 x = torch.reshape(x, [5, 5]) 97 bias = torch.tensor([1, 1, 1, 1, 1, 1, 1], dtype=torch.float) * 0.1 [all …]
|
H A D | quantized_conv3d_relu_test.py | 6 import torch 13 with torch.no_grad(): 14 x = torch.tensor(range(5), dtype=torch.float) 16 x = torch.cat((x, x, x)) 19 q = torch.nn.quantized.Quantize(1, 2, torch.quint8) 22 relu = torch.nn.ReLU() 28 torch.arange(72 / groups, dtype=torch.float).reshape( 75 with torch.no_grad(): 76 x = torch.tensor(range(5), dtype=torch.float) 81 q = torch.nn.quantized.Quantize(1, 2, torch.quint8) [all …]
|
H A D | quantized_conv2d_relu_test.py | 3 import torch 13 with torch.no_grad(): 14 x = torch.tensor(range(5), dtype=torch.float) / 3 16 x = torch.cat((x, x, x)) 18 q = torch.nn.quantized.Quantize(1, 2, torch.quint8) 20 relu = torch.nn.ReLU() 25 conv.weight.set_(torch.arange(36/groups, dtype=torch.float).reshape([3, 67 with torch.no_grad(): 68 x = torch.tensor(range(5), dtype=torch.float) / 3 72 q = torch.nn.quantized.Quantize(1, 2, torch.quint8) [all …]
|
H A D | quantized_add_relu_test.py | 3 import torch 14 q = torch.nn.quantized.Quantize( 18 torch.ops.quantized.add_relu( 22 x = torch.tensor([1, 2, 3, 4], dtype=torch.float32) 23 y = torch.tensor([5, 6, 7, 8], dtype=torch.float32) 40 q1 = torch.nn.quantized.Quantize( 43 q2 = torch.nn.quantized.Quantize( 53 x = torch.randn([5, 5]) 54 y = torch.randn([5, 5]) 84 x = torch.randn([5, 5]) [all …]
|
H A D | quantized_add_test.py | 3 import torch 14 q = torch.nn.quantized.Quantize( 19 x = torch.tensor([1, 2, 3, 4], dtype=torch.float32) 20 y = torch.tensor([5, 6, 7, 8], dtype=torch.float32) 37 q1 = torch.nn.quantized.Quantize( 45 torch.ops.quantized.add( 49 x = torch.randn([5, 5]) 50 y = torch.randn([5, 5]) 75 torch.ops.quantized.add( 79 x = torch.randn([5, 5]) [all …]
|
H A D | rsub_test.py | 3 import torch 14 c = torch.rsub(a, b) 15 return torch.rsub(c, c) 17 x = torch.randn(4) 18 y = torch.randn(4) 26 c = torch.rsub(a, b) 30 y = torch.randn(4, 2) 42 y = torch.randn(1, 2) 53 x = torch.randn(4, 2) 64 x = torch.randn(4) [all …]
|
/dports/misc/glow/glow-f24d960e3cc80db95ac0bc17b1900dbf60ca044a/torch_glow/src/ |
H A D | PyTorchModelLoader.h | 109 const at::ArrayRef<torch::jit::IValue> inputs_; 115 std::unordered_map<const torch::jit::Value *, torch::jit::IValue> qparamsMap_; 410 Error loadMul(const torch::jit::Node *ptNode); 414 Error loadDiv(const torch::jit::Node *ptNode); 418 Error loadAdd(const torch::jit::Node *ptNode); 422 Error loadSub(const torch::jit::Node *ptNode); 430 Error loadMax(const torch::jit::Node *ptNode); 446 Error loadExp(const torch::jit::Node *ptNode); 553 Error loadT(const torch::jit::Node *ptNode); 613 Error loadMM(const torch::jit::Node *ptNode); [all …]
|
H A D | GlowFuser.cpp | 37 torch::jit::value_list 40 torch::jit::value_list result; in sortReverseTopological() 48 [&](torch::jit::Value *a, torch::jit::Value *b) { in sortReverseTopological() 98 bool aliasChecks(torch::jit::Node *consumer, torch::jit::Node *producer, in aliasChecks() 118 torch::jit::Node *tryMerge(torch::jit::Node *consumer, in tryMerge() 149 producer, [](torch::jit::Value *) -> torch::jit::Value * { in tryMerge() 168 std::shared_ptr<torch::jit::Graph> getSubgraph(torch::jit::Node *n) { in getSubgraph() 172 const std::shared_ptr<torch::jit::Graph> 173 getSubgraph(const torch::jit::Node *n) { in getSubgraph() 178 getNewNode(torch::jit::Node *node, torch::jit::AliasDb &aliasDb, in getNewNode() [all …]
|
H A D | FuseKnownPatterns.cpp | 35 torch::jit::RegisterOperators op({torch::jit::Operator( in registerDummyOperator() 37 [](const torch::jit::Node *node) -> torch::jit::Operation { in registerDummyOperator() 113 torch::jit::SubgraphRewriter rewriter; in fuseLinearPrepack() 129 torch::jit::SubgraphRewriter rewriter; in fuseNumToTensorToNum() 163 fusedNode->i_(torch::jit::attr::dim, dim); in fuseConcat() 207 torch::jit::Value *inputValue = nullptr; in fuseBranchedLinearPattern() 208 torch::jit::Value *cValue = nullptr; in fuseBranchedLinearPattern() 209 torch::jit::Value *weightValue = nullptr; in fuseBranchedLinearPattern() 210 torch::jit::Value *biasValue = nullptr; in fuseBranchedLinearPattern() 211 torch::jit::Value *dValue = nullptr; in fuseBranchedLinearPattern() [all …]
|
/dports/math/faiss/faiss-1.7.1/faiss/gpu/test/ |
H A D | torch_test_contrib_gpu.py | 7 import torch 32 xb_torch_gpu = torch.rand(10000, 128, device=torch.device('cuda', 0), dtype=torch.float32) 57 new_i_torch_gpu = torch.zeros(10, 10, device=torch.device('cuda', 0), dtype=torch.int64) 80 xb = torch.rand(1000, d, device=torch.device('cuda', 0), dtype=torch.float32) 117 xb = torch.rand(100, d, device=torch.device('cuda', 0), dtype=torch.float32) 131 y = torch.empty(d, dtype=torch.float32) 136 y = torch.empty(d, device=torch.device('cuda', 0), dtype=torch.float32) 151 y = torch.empty(10, d, dtype=torch.float32) 156 y = torch.empty(10, d, device=torch.device('cuda', 0), dtype=torch.float32) 167 xb = torch.rand(10000, d, device=torch.device('cuda', 0), dtype=torch.float32) [all …]
|
/dports/math/faiss/faiss-1.7.1/tests/ |
H A D | torch_test_contrib.py | 7 import torch 36 xq_torch = torch.rand(10, d, dtype=torch.float32) 69 xb = torch.rand(1000, d, dtype=torch.float32) 90 xb = torch.rand(100, d, dtype=torch.float32) 103 y = torch.empty(d, dtype=torch.float32) 117 y = torch.empty(10, d, dtype=torch.float32) 125 xb = torch.rand(1000, d, dtype=torch.float32) 132 xq = torch.rand(10, d, dtype=torch.float32) 212 xq = torch.rand(10, d, dtype=torch.float32) 230 xq = torch.rand(10, d, dtype=torch.float32) [all …]
|
/dports/math/py-pytorchvideo/pytorchvideo-0.1.3/pytorchvideo/transforms/ |
H A D | transforms.py | 6 import torch 29 def __call__(self, x: Dict[str, torch.Tensor]) -> Dict[str, torch.Tensor]: 62 def forward(self, x: torch.Tensor) -> torch.Tensor: 101 def forward(self, x: torch.Tensor) -> torch.Tensor: 120 def forward(self, x: torch.Tensor) -> torch.Tensor: 163 def forward(self, x: torch.Tensor) -> torch.Tensor: 179 def forward(self, x: torch.Tensor) -> torch.Tensor: 215 def __call__(self, x: torch.Tensor) -> torch.Tensor: 250 def forward(self, x: torch.Tensor) -> torch.Tensor: 311 def forward(self, x: torch.Tensor) -> torch.Tensor: [all …]
|
H A D | functional.py | 8 import torch 21 ) -> torch.Tensor: 44 @torch.jit.ignore 47 ) -> torch.Tensor: 97 ) -> torch.Tensor: 240 ) -> Tuple[torch.Tensor, torch.Tensor]: 268 images: torch.Tensor, size: int, boxes: torch.Tensor 269 ) -> Tuple[torch.Tensor, torch.Tensor]: 381 prob: float, images: torch.Tensor, boxes: torch.Tensor 382 ) -> Tuple[torch.Tensor, torch.Tensor]: [all …]
|
/dports/math/py-fvcore/fvcore-0.1.5.post20210924/fvcore/nn/ |
H A D | distributed.py | 3 import torch 4 import torch.distributed as dist 5 from torch.autograd.function import Function 13 def forward(ctx, input: torch.Tensor) -> torch.Tensor: 18 return torch.sum(inputs, dim=0) 21 def backward(ctx, grad_output: torch.Tensor) -> torch.Tensor: 26 def differentiable_all_reduce(input: torch.Tensor) -> torch.Tensor: 41 def forward(ctx, x: torch.Tensor) -> Tuple[torch.Tensor, ...]: 47 def backward(ctx, *grads: torch.Tensor) -> torch.Tensor: 48 all_gradients = torch.stack(grads) [all …]
|
H A D | precise_bn.py | 8 import torch 10 from torch import nn 14 torch.nn.BatchNorm1d, 15 torch.nn.BatchNorm2d, 16 torch.nn.BatchNorm3d, 17 torch.nn.SyncBatchNorm, 39 self.pop_mean: torch.Tensor = torch.zeros_like(mean_buffer) 40 self.pop_var: torch.Tensor = torch.zeros_like(var_buffer) 44 self, batch_mean: torch.Tensor, batch_var: torch.Tensor, batch_size: int 66 self.pop_mean: torch.Tensor = torch.zeros_like(mean_buffer) [all …]
|
/dports/misc/ncnn/ncnn-20211208/tools/pnnx/tests/ |
H A D | test_F_batch_norm.py | 15 import torch 16 import torch.nn as nn 46 m1 = torch.rand(2) 47 v1 = torch.rand(2) 48 w1 = torch.rand(2) 49 b1 = torch.rand(2) 50 m2 = torch.rand(3) 51 v2 = torch.rand(3) 52 w2 = torch.rand(3) 53 b2 = torch.rand(3) [all …]
|
H A D | test_F_instance_norm.py | 15 import torch 16 import torch.nn as nn 42 m0 = torch.rand(12) 43 v0 = torch.rand(12) 44 w0 = torch.rand(12) 45 b0 = torch.rand(12) 46 m1 = torch.rand(3) 47 v1 = torch.rand(3) 48 w1 = torch.rand(3) 49 b1 = torch.rand(3) [all …]
|
/dports/games/supertux2/SuperTux-v0.6.3-Source/data/levels/halloween2014/ |
H A D | halloween1.stl | 684 (torch 688 (torch 693 (torch 698 (torch 703 (torch 708 (torch 713 (torch 718 (torch 723 (torch 728 (torch [all …]
|
H A D | mound.stl | 634 (torch 638 (torch 643 (torch 648 (torch 653 (torch 658 (torch 662 (torch 667 (torch 671 (torch 676 (torch [all …]
|
/dports/math/faiss/faiss-1.7.1/contrib/ |
H A D | torch_utils.py | 25 import torch 166 labels = torch.empty(n, k, device=x.device, dtype=torch.int64) 215 D = torch.empty(n, k, device=x.device, dtype=torch.float32) 222 I = torch.empty(n, k, device=x.device, dtype=torch.int64) 251 D = torch.empty(n, k, device=x.device, dtype=torch.float32) 258 I = torch.empty(n, k, device=x.device, dtype=torch.int64) 304 x = torch.empty(self.d, device=device, dtype=torch.float32) 447 x = torch.empty(n, self.d, dtype=torch.float32) 536 D = torch.empty(nq, k, device=xb.device, dtype=torch.float32) 543 I = torch.empty(nq, k, device=xb.device, dtype=torch.int64) [all …]
|
/dports/misc/glow/glow-f24d960e3cc80db95ac0bc17b1900dbf60ca044a/torch_glow/tests/functionality/ |
H A D | conv_to_glow_test.py | 4 import torch 13 with torch.no_grad(): 18 conv_op = torch.nn.Conv2d(3, 10, 3) 20 conv_op = torch.nn.Conv3d(3, 10, 3) 30 model = torch.nn.Sequential( 49 traced_m = torch.jit.trace(m, (x)) 54 sim.set(x.size(), torch.float32) 64 x = torch.randn([1, 3, 30, 30]) 69 x = torch.randn([1, 3, 30, 30]) 74 x = torch.randn([1, 3, 30, 30, 30]) [all …]
|
/dports/science/py-kliff/kliff-0.3.0/kliff/models/ |
H A D | linear_regression.py | 1 import torch 2 import torch.nn as nn 6 from torch.utils.data import DataLoader 32 A = torch.inverse(torch.mm(X.t(), X)) 33 beta = torch.mv(torch.mm(A, X.t()), y) 50 self.layer.bias = torch.nn.Parameter(beta[0:1]) 77 x_ = torch.cat((zeta, torch.transpose(dzeta))) 78 y_ = torch.cat((e, f)) 83 x_ = torch.transpose(dzeta) 93 X = torch.cat(X) [all …]
|