/aosp_15_r20/external/tensorflow/tensorflow/core/grappler/costs/graph_properties_testdata/ |
H A D | large_graph.pbtxt.html | 123622 …name: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Expan… 123644 …name: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Expan… 123647 …input: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Expa… 123683 …name: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Expan… 123705 …name: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Expan… 123708 …input: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Expa… 123744 …name: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Conv2… 123746 …input: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Expa… 123747 …input: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Expa… 123817 …name: "seq2seq/seq2seq_2/convert_to_lin_specgram/dilated_conv1d_stack/1x1_residual_in/conv1d/Squee… [all …]
|
/aosp_15_r20/external/executorch/backends/xnnpack/test/ops/ |
H A D | conv1d.py | 22 class Conv1d(torch.nn.Module): class in TestConv1d 34 self.conv1d = torch.nn.Conv1d( 46 return self.conv1d(x) 59 self.conv1 = torch.nn.Conv1d( 70 self.conv2 = torch.nn.Conv1d( 109 .check_count({"torch.ops.aten.conv1d.default": conv_count}) 123 self.Conv1d(dtype=torch.float16), 132 self._test_conv1d(self.Conv1d(), inputs, 1, dynamic_shape=dynamic_shapes) 145 self.Conv1d(), inputs, 1, quantized=True, dynamic_shape=dynamic_shapes 163 self.Conv1d(),
|
/aosp_15_r20/external/executorch/backends/arm/test/ops/ |
H A D | test_conv1d.py | 19 class Conv1d(torch.nn.Module): class 88 torch.nn.Conv1d( 111 conv1d_2_3x2x40_nobias = Conv1d( 122 conv1d_3_1x3x256_st1 = Conv1d( 132 conv1d_3_1x3x12_st2_pd1 = Conv1d( 142 conv1d_1_1x2x128_st1 = Conv1d( 152 conv1d_2_1x2x14_st2 = Conv1d( 162 conv1d_5_3x2x128_st1 = Conv1d( 172 conv1d_3_1x3x224_st2_pd1 = Conv1d( 182 two_conv1d_nobias = Conv1d( [all …]
|
H A D | test_depthwise_conv.py | 15 from executorch.backends.arm.test.ops.test_conv1d import Conv1d 31 dw_conv1d_3_1x3x14_gp3_st1 = Conv1d( 42 dw_conv1d_2_1x6x4_gp6_st1 = Conv1d( 65 dw_conv1d_3_1x3x256_gp3_st1 = Conv1d( 124 two_dw_conv1d = Conv1d( 170 """Tests Conv1D and Conv2D where groups == in_channels and out_channels = K * in_channels. This 257 # Expected to fail as conv1d needs transpose which is not supported
|
/aosp_15_r20/external/pytorch/torch/ao/nn/intrinsic/qat/modules/ |
H A D | conv_fused.py | 437 class ConvBn1d(_ConvBnNd, nn.Conv1d): 439 A ConvBn1d module is a module fused from Conv1d and BatchNorm1d, 443 We combined the interface of :class:`torch.nn.Conv1d` and 446 Similar to :class:`torch.nn.Conv1d`, with FakeQuantize modules initialized 457 _FLOAT_CONV_MODULE = nn.Conv1d 461 # Conv1d args 508 A ConvBnReLU1d module is a module fused from Conv1d, BatchNorm1d and ReLU, 512 We combined the interface of :class:`torch.nn.Conv1d` and 515 Similar to `torch.nn.Conv1d`, with FakeQuantize modules initialized to 524 _FLOAT_CONV_MODULE = nn.Conv1d [all …]
|
/aosp_15_r20/external/pytorch/torch/ao/nn/quantized/dynamic/modules/ |
H A D | conv.py | 18 "Conv1d", 27 class Conv1d(nnq.Conv1d): class 31 :class:`~torch.nn.Conv1d` and :class:`~torch.ao.nn.quantized.dynamic.Conv1d` and 39 See :class:`~torch.nn.Conv1d` for other attributes. 44 >>> m = nn.quantized.dynamic.Conv1d(16, 33, 3, stride=2) 50 _FLOAT_MODULE = nn.Conv1d 100 # Padding in Conv1d is stored as (p, p), need to get (p,) 283 For special notes, please, see :class:`~torch.ao.nn.quantized.dynamic.Conv1d` 301 >>> downsample = nndq.Conv1d(16, 16, 3, stride=2, padding=1)
|
/aosp_15_r20/external/pytorch/test/nn/ |
H A D | test_convolution.py | 104 module = nn.Conv1d( 116 module = nn.Conv1d( 226 torch.nn.Conv1d(1, 1, kernel_size=3, dilation=2, stride=2, groups=0) 235 module = nn.Conv1d( 238 expect = F.conv1d(x, module.weight, module.bias, padding="same") 242 module = nn.Conv1d( 245 expect = F.conv1d(x, module.weight, module.bias, padding="same", dilation=2) 249 module = nn.Conv1d( 257 expect = F.conv1d(x_padded, module.weight, module.bias, padding="valid") 263 module = nn.Conv1d( [all …]
|
/aosp_15_r20/external/pytorch/benchmarks/functional_autograd_benchmark/ |
H A D | torchaudio_models.py | 34 nn.Conv1d( 42 nn.Conv1d( 46 nn.Conv1d( 50 nn.Conv1d( 54 nn.Conv1d( 58 nn.Conv1d( 62 nn.Conv1d( 66 nn.Conv1d( 70 nn.Conv1d( 74 nn.Conv1d( [all …]
|
/aosp_15_r20/external/libopus/dnn/torch/osce/utils/layers/ |
H A D | td_shaper.py | 54 self.feature_alpha1_f = norm(nn.Conv1d(self.feature_dim, frame_size, 2)) 55 self.feature_alpha1_t = norm(nn.Conv1d(self.env_dim, frame_size, 2)) 56 self.feature_alpha2 = norm(nn.Conv1d(frame_size, frame_size, 2)) 62 self.feature_alpha1b = norm(nn.Conv1d(self.feature_dim + self.env_dim, frame_size, 2)) 63 self.feature_alpha1c = norm(nn.Conv1d(self.feature_dim + self.env_dim, frame_size, 2)) 65 self.feature_alpha2b = norm(nn.Conv1d(frame_size, frame_size, 2)) 66 self.feature_alpha2c = norm(nn.Conv1d(frame_size, frame_size, 2))
|
/aosp_15_r20/external/pytorch/torch/ao/nn/quantized/reference/modules/ |
H A D | conv.py | 13 "Conv1d", 54 class Conv1d(_ConvNd, nn.Conv1d): class 70 nn.Conv1d.__init__( 90 x(float) ------------- F.conv1d --- 94 x -- quant --- *dequant -- *F.conv1d --- *quant - dequant 95 and the backend should be able to fuse the ops with `*` into a quantized conv1d 98 result = F.conv1d( 317 and the backend should be able to fuse the ops with `*` into a quantized conv1d
|
/aosp_15_r20/external/tensorflow/tensorflow/python/layers/ |
H A D | convolutional.py | 20 Conv1D = convolutional.Conv1D variable 21 conv1d = convolutional.conv1d variable 37 Convolution1D = Conv1D 43 convolution1d = conv1d
|
/aosp_15_r20/external/executorch/backends/cadence/reference/operators/ |
H A D | quantized_conv_out.cpp | 154 // quantized::conv1d or quantized::conv2d based on the dimensionality of 174 bool conv1d = input.dim() == 3; in quantized_conv_out() local 178 const int h = conv1d ? 1 : input.size(2); in quantized_conv_out() 179 const int w = conv1d ? input.size(2) : input.size(3); in quantized_conv_out() 183 const int wh = conv1d ? 1 : weight.size(2); in quantized_conv_out() 184 const int ww = conv1d ? weight.size(2) : weight.size(3); in quantized_conv_out() 186 const int oh = conv1d ? 1 : out.size(2); in quantized_conv_out() 187 const int ow = conv1d ? out.size(2) : out.size(3); in quantized_conv_out()
|
/aosp_15_r20/external/pytorch/torch/csrc/jit/passes/onnx/ |
H A D | unpack_quantized_weights.cpp | 247 // CONV1D needs a different unpacking from CONV, since it's 250 enum class QuantizedParamsType { CONV1D, CONV, LINEAR }; enumerator 373 params_type == QuantizedParamsType::CONV1D) && in unpackQuantizedWeightsHelper() 386 // kSpatialDim = 2 even it's for Conv1D from torch.op to adopt Conv2D, in unpackQuantizedWeightsHelper() 387 // so we need a special unpack for Conv1D which has Conv2D dim. in unpackQuantizedWeightsHelper() 390 if (params_type != QuantizedParamsType::CONV1D || i != 0) { in unpackQuantizedWeightsHelper() 396 if (params_type != QuantizedParamsType::CONV1D || i != 0) { in unpackQuantizedWeightsHelper() 402 if (params_type != QuantizedParamsType::CONV1D || i != 0) { in unpackQuantizedWeightsHelper() 408 if (params_type != QuantizedParamsType::CONV1D || i != 0) { in unpackQuantizedWeightsHelper() 428 if (params_type == QuantizedParamsType::CONV1D) { in unpackQuantizedWeightsHelper() [all …]
|
/aosp_15_r20/external/pytorch/torch/ao/nn/qat/modules/ |
H A D | conv.py | 11 __all__ = ["Conv1d", "Conv2d", "Conv3d"] 126 class Conv1d(_ConvNd, nn.Conv1d): class 128 A Conv1d module attached with FakeQuantize modules for weight, 131 We adopt the same interface as :class:`~torch.nn.Conv1d` 139 _FLOAT_MODULE = nn.Conv1d 140 _FLOAT_CONV_MODULE = nn.Conv1d
|
/aosp_15_r20/external/pytorch/torch/ao/ns/fx/ |
H A D | weight_utils.py | 52 if isinstance(mod, (nn.Conv1d, nn.Conv2d, nn.Conv3d)): 159 # Conv1d 160 nn.Conv1d: mod_weight_detach, 162 nnq.Conv1d: mod_weight_bias_0, 163 nnqat.Conv1d: mod_weight_detach, 202 F.conv1d: get_conv_fun_weight, 205 toq.conv1d: get_qconv_fun_weight,
|
H A D | mappings.py | 30 nn.Conv1d, 40 F.conv1d, 374 # example: nn.Conv1d, nn.ReLU fused into nni.ConvReLU1d 490 F.conv1d, 515 toq.conv1d, 582 nn.Conv1d, 585 nnqat.Conv1d, 643 nnq.Conv1d,
|
/aosp_15_r20/external/pytorch/torch/ao/quantization/ |
H A D | quantization_mappings.py | 61 nn.Conv1d: nnqr.Conv1d, 82 nn.Conv1d: nnq.Conv1d, 116 nniqat.ConvBn1d: nnq.Conv1d, 166 # nn.Conv1d: nnqd.Conv1d,
|
H A D | fuser_method_mappings.py | 42 nn.Conv1d: nni.ConvBn1d, 89 nn.Conv1d: nni.ConvBnReLU1d, 107 nn.Conv1d: nni.ConvReLU1d, 195 (nn.Conv1d, nn.BatchNorm1d): fuse_conv_bn, 196 (nn.Conv1d, nn.BatchNorm1d, nn.ReLU): fuse_conv_bn_relu, 201 (nn.Conv1d, nn.ReLU): _sequential_wrapper2(nni.ConvReLU1d),
|
/aosp_15_r20/external/pytorch/test/jit/ |
H A D | test_optimize_for_mobile_preserve_debug_info.py | 49 return F.conv1d(x, self.weight, self.bias) 59 "prim::ListUnpack": "aten::conv1d", 60 "prim::ListConstruct": "aten::conv1d", 61 "aten::unsqueeze": "aten::conv1d", 62 "aten::conv2d": "aten::conv1d", 63 "aten::squeeze": "aten::conv1d",
|
/aosp_15_r20/external/pytorch/test/xpu/ |
H A D | test_conv.py | 381 z = F.conv1d(x, y, padding="same", dilation=dilation, stride=stride) 386 expect = F.conv1d(x, y, padding=1) 387 actual = F.conv1d(x, y, padding="same") 392 expect = F.conv1d(x, y, padding=3, dilation=2) 393 actual = F.conv1d(x, y, padding="same", dilation=2) 396 expect = F.conv1d(x, y, padding=5, dilation=3)[..., 1:] 397 actual = F.conv1d(x, y, padding="same", dilation=3) 422 expect = F.conv1d(x, y) 423 actual = F.conv1d(x, y, padding="valid") 447 z = F.conv1d(x, y, padding=3, dilation=2) [all …]
|
/aosp_15_r20/external/libopus/dnn/torch/fwgan/models/ |
H A D | fwgan400.py | 20 …self.conv = which_norm(nn.Conv1d(in_ch,out_ch,kernel_size,dilation=dilation, groups=groups, bias= … 27 …if isinstance(m, nn.Conv1d) or isinstance(m, nn.ConvTranspose1d) or isinstance(m, nn.Linear) or is… 50 if isinstance(m, nn.Conv1d) or isinstance(m, nn.ConvTranspose1d)\ 82 …if isinstance(m, nn.Conv1d) or isinstance(m, nn.ConvTranspose1d) or isinstance(m, nn.Linear) or is… 138 … if isinstance(m, nn.Conv1d) or isinstance(m, nn.ConvTranspose1d) or isinstance(m, nn.Linear) or\ 174 if isinstance(m, nn.Conv1d) or isinstance(m, nn.ConvTranspose1d) or\ 228 … if isinstance(m, nn.Conv1d) or isinstance(m, nn.ConvTranspose1d) or isinstance(m, nn.Linear) or\
|
/aosp_15_r20/external/pytorch/torch/ao/nn/intrinsic/quantized/modules/ |
H A D | conv_relu.py | 21 class ConvReLU1d(nnq.Conv1d): 23 A ConvReLU1d module is a fused module of Conv1d and ReLU 25 We adopt the same interface as :class:`torch.ao.nn.quantized.Conv1d`. 28 Same as torch.ao.nn.quantized.Conv1d 67 # Padding in Conv1d is stored as (p, p), need to get (p,) 98 ), "BatchNorm1d should be fused into Conv1d before converting to reference module"
|
/aosp_15_r20/external/pytorch/torch/csrc/api/include/torch/nn/functional/ |
H A D | conv.h | 26 inline Tensor conv1d( in conv1d() function 36 return torch::conv1d( in conv1d() 45 /// https://pytorch.org/docs/main/nn.functional.html#torch.nn.functional.conv1d 54 /// F::conv1d(x, weight, F::Conv1dFuncOptions().stride(1)); 56 inline Tensor conv1d( 60 return detail::conv1d(
|
/aosp_15_r20/external/libopus/dnn/torch/dnntools/dnntools/sparsification/ |
H A D | conv1d_sparsifier.py | 43 task_list contains a list of tuples (conv1d, params), where conv1d is an instance 44 of torch.nn.Conv1d and params is a tuple (density, [m, n]), 66 >>> conv = torch.nn.Conv1d(8, 16, 8) 125 conv = torch.nn.Conv1d(8, 16, 8)
|
/aosp_15_r20/external/pytorch/benchmarks/operator_benchmark/pt/ |
H A D | conv_test.py | 10 Microbenchmarks for Conv1d and ConvTranspose1d operators. 19 self.conv1d = nn.Conv1d(IC, OC, kernel, stride=stride).to(device=device) 20 self.set_module_name("Conv1d") 23 return self.conv1d(input)
|