| /external/tensorflow/tensorflow/python/keras/applications/ |
| D | vgg19.py | 143 x = layers.Conv2D(
146 x = layers.Conv2D(
151 x = layers.Conv2D(
153 x = layers.Conv2D(
158 x = layers.Conv2D(
160 x = layers.Conv2D(
162 x = layers.Conv2D(
164 x = layers.Conv2D(
169 x = layers.Conv2D(
171 x = layers.Conv2D(
[all …]
|
| D | vgg16.py | 143 x = layers.Conv2D(
146 x = layers.Conv2D(
151 x = layers.Conv2D(
153 x = layers.Conv2D(
158 x = layers.Conv2D(
160 x = layers.Conv2D(
162 x = layers.Conv2D(
167 x = layers.Conv2D(
169 x = layers.Conv2D(
171 x = layers.Conv2D(
[all …]
|
| D | resnet.py | 168 x = layers.Conv2D(64, 7, strides=2, use_bias=use_bias, name='conv1_conv')(x)
244 shortcut = layers.Conv2D(
251 x = layers.Conv2D(filters, 1, strides=stride, name=name + '_1_conv')(x)
256 x = layers.Conv2D(
262 x = layers.Conv2D(4 * filters, 1, name=name + '_3_conv')(x)
312 shortcut = layers.Conv2D(
317 x = layers.Conv2D(
324 x = layers.Conv2D(
334 x = layers.Conv2D(4 * filters, 1, name=name + '_3_conv')(x)
384 shortcut = layers.Conv2D(
[all …]
|
| D | mobilenet_v3.py | 265 x = layers.Conv2D(
285 x = layers.Conv2D(
295 x = layers.Conv2D(
311 x = layers.Conv2D(classes, kernel_size=1, padding='same', name='Logits')(x)
471 x = layers.Conv2D(
478 x = layers.Conv2D(
498 x = layers.Conv2D(
536 x = layers.Conv2D(
|
| D | xception.py | 146 x = layers.Conv2D(
153 x = layers.Conv2D(64, (3, 3), use_bias=False, name='block1_conv2')(x)
157 residual = layers.Conv2D(
175 residual = layers.Conv2D(
194 residual = layers.Conv2D(
244 residual = layers.Conv2D(
|
| D | efficientnet.py | 322 x = layers.Conv2D(
358 x = layers.Conv2D(
447 x = layers.Conv2D(
487 se = layers.Conv2D(
495 se = layers.Conv2D(
505 x = layers.Conv2D(
|
| /external/tensorflow/tensorflow/lite/delegates/xnnpack/ |
| D | conv_2d_test.cc | 28 TEST(Conv2D, 1x1) {
54 TEST(Conv2D, 3x3) {
80 TEST(Conv2D, 3x3Stride2) {
108 TEST(Conv2D, SmallKernelWithSamePadding) { in TEST() argument
136 TEST(Conv2D, SmallKernelWithValidPadding) { in TEST() argument
164 TEST(Conv2D, StrideWithSamePadding) { in TEST() argument
196 TEST(Conv2D, StrideWithValidPadding) { in TEST() argument
228 TEST(Conv2D, DilationWithSamePadding) { in TEST() argument
260 TEST(Conv2D, DilationWithValidPadding) { in TEST() argument
292 TEST(Conv2D, FP16Weights) { in TEST() argument
[all …]
|
| /external/tensorflow/tensorflow/compiler/mlir/lite/tests/ |
| D | dilated-conv.mlir | 7 …%1 = "tf.Conv2D"(%0, %arg1) {padding = "VALID", strides = [1, 1, 1, 1]} : (tensor<4x68x68x3xf32>, …
13 …// CHECK-NEXT: [[RESULT:%.*]] = "tf.Conv2D"([[INPUT]], [[FILTER]]) {dilations = [1, 2, 2, 1], padd…
20 …%1 = "tf.Conv2D"(%0, %arg2) {padding = "VALID", strides = [1, 1, 1, 1]} : (tensor<4x68x68x3xf32>, …
26 …// CHECK-NEXT: [[RESULT:%.*]] = "tf.Conv2D"([[INPUT]], [[FILTER]]) {dilations = [1, 2, 2, 1], padd…
35 …%1 = "tf.Conv2D"(%0, %arg1) {padding = "VALID", strides = [1, 1, 1, 1]} : (tensor<4x68x68x3xf32>, …
41 …// CHECK-NEXT: [[RESULT:%.*]] = "tf.Conv2D"([[INPUT]], [[FILTER]]) {dilations = [1, 2, 2, 1], padd…
50 …%1 = "tf.Conv2D"(%0, %arg1) {padding = "VALID", dilations = [1, 2, 2, 1], strides = [1, 1, 1, 1]} …
56 // CHECK-NEXT: [[CONV:%.*]] = "tf.Conv2D"
81 …%1 = "tf.Conv2D"(%0, %arg2) {padding = "VALID", strides = [1, 1, 1, 1]} : (tensor<4x68x68x3xf32>, …
89 …// CHECK-NEXT: [[CONV:%.*]] = "tf.Conv2D"([[INPUT]], [[FILTER]]) {dilations = [1, 2, 2, 1], paddin…
[all …]
|
| /external/tensorflow/tensorflow/python/layers/ |
| D | convolutional.py | 26 Conv2D = convolutional.Conv2D variable
42 Convolution2D = Conv2D
|
| /external/tensorflow/tensorflow/python/keras/integration_test/ |
| D | gradient_checkpoint_test.py | 36 model.add(layers.Conv2D(10, 5, padding='same', activation=tf.nn.relu))
38 model.add(layers.Conv2D(40, 5, padding='same', activation=tf.nn.relu))
40 model.add(layers.Conv2D(20, 5, padding='same', activation=tf.nn.relu))
59 model.add(layers.Conv2D(10, 5, padding='same', activation=tf.nn.relu))
61 model.add(layers.Conv2D(40, 5, padding='same', activation=tf.nn.relu))
63 model.add(layers.Conv2D(20, 5, padding='same', activation=tf.nn.relu))
|
| /external/tensorflow/tensorflow/python/eager/benchmarks/resnet50/ |
| D | resnet50.py | 54 self.conv2a = layers.Conv2D(
59 self.conv2b = layers.Conv2D(
68 self.conv2c = layers.Conv2D(
118 self.conv2a = layers.Conv2D(
126 self.conv2b = layers.Conv2D(
135 self.conv2c = layers.Conv2D(
140 self.conv_shortcut = layers.Conv2D(
233 self.conv1 = layers.Conv2D(
|
| /external/tensorflow/tensorflow/compiler/mlir/tensorflow/tests/ |
| D | layout_optimization_layout_assignment_gpu_cc_70.mlir | 10 // CHECK: "tf.Conv2D"(%[[INPUT_TRANSPOSE:[0-9]*]], %arg1)
12 %0 = "tf.Conv2D"(%input, %filter)
26 // CHECK: "tf.Conv2D"(%[[INPUT_TRANSPOSE:[0-9]*]], %arg1)
28 %0 = "tf.Conv2D"(%input, %filter)
37 // CHECK: "tf.Conv2D"(%arg0, %arg1)
39 %1 = "tf.Conv2D"(%input, %filter)
53 // CHECK: "tf.Conv2D"(%[[INPUT_TRANSPOSE:[0-9]*]], %arg1)
55 %0 = "tf.Conv2D"(%input, %filter)
|
| D | layout_optimization_to_nchw.mlir | 11 %2 = "tf.Conv2D"(%1, %arg1)
23 // Check that Conv2D computed in NCHW format, and all redundant transpose
26 // CHECK: %[[CONV:[0-9]*]] = "tf.Conv2D"(%arg0, %arg1)
|
| D | layout_optimization_to_nhwc.mlir | 46 %5 = "tf.Conv2D"(%4, %arg3)
55 // CHECK: %[[CONV0:[0-9]*]] = "tf.Conv2D"
89 %9 = "tf.Conv2D"(%8, %arg4)
98 // CHECK: %[[CONV1:[0-9]*]] = "tf.Conv2D"(%[[MAX_POOL]], %arg4)
116 %11 = "tf.Conv2D"(%8, %arg4)
125 // CHECK: %[[CONV2:[0-9]*]] = "tf.Conv2D"(%[[MAX_POOL]], %arg4)
|
| D | optimize.mlir | 8 …%0 = "tf.Conv2D"(%arg, %filter) {T = "tfdtype$DT_FLOAT", data_format = "NHWC", dilations = [1, 2, …
15 // CHECK-NEXT: %[[conv:.*]] = "tf.Conv2D"(%arg0, %[[cst]])
25 …%0 = "tf.Conv2D"(%arg, %filter) {T = "tfdtype$DT_FLOAT", data_format = "NHWC", dilations = [1, 2, …
32 // CHECK-NEXT: %[[conv:.*]] = "tf.Conv2D"(%arg0, %[[cst]])
|
| D | fused_kernel_matcher.mlir | 4 // Conv2D + BiasAdd + <Activation> fusions.
12 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
23 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
35 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
47 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
57 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
72 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
83 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
92 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
103 …%0 = "tf.Conv2D"(%arg2, %arg1) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings …
|
| /external/tensorflow/tensorflow/core/profiler/g3doc/ |
| D | profile_model_architecture.md | 73 init/init_conv/Conv2D (113.25m/113.25m flops)
76 unit_1_0/sub1/conv1/Conv2D (603.98m/603.98m flops)
77 unit_1_0/sub2/conv2/Conv2D (603.98m/603.98m flops)
78 unit_1_1/sub1/conv1/Conv2D (603.98m/603.98m flops)
79 unit_1_1/sub2/conv2/Conv2D (603.98m/603.98m flops)
84 Conv2D 17.63b float_ops (100.00%, 100.00%)
|
| /external/tensorflow/tensorflow/core/kernels/ |
| D | conv_ops_benchmark_test.cc | 76 static Conv2DGraph Conv2D(int batch, int height, int width, int in_depth, in Conv2D() function
110 Conv2DGraph conv_graph = Conv2D<T>(batch, height, width, in_depth, filter_w, in Conv2DWithBias()
159 Conv2DGraph conv_graph = Conv2D<T>(batch, height, width, in_depth, filter_w, in Conv2DWithBatchNorm()
323 test::Benchmark(#type, Conv2D<float>(N, H, W, C, FW, FH, FC).graph, \
326 BM_SET_INFO(N, H, W, C, type, LABEL, Conv2D); \
338 BM_SET_INFO(N, H, W, C, type, LABEL, Conv2D); \
352 BM_SET_INFO(N, H, W, C, type, LABEL, Conv2D); \
365 BM_SET_INFO(N, H, W, C, type, LABEL, Conv2D); \
378 BM_SET_INFO(N, H, W, C, type, LABEL, Conv2D); \
391 BM_SET_INFO(N, H, W, C, type, LABEL, Conv2D); \
[all …]
|
| D | conv_ops_test.cc | 219 Output conv = Conv2D(root.WithOpName("conv"), mirror_pad, casted_filter, in CompareFusedAndSeparate()
273 Output conv = Conv2D(root.WithOpName("conv"), mirror_pad, casted_filter, in CompareFusedPadOnlyAndSeparate()
619 ops::Conv2D conv = ops::Conv2D( in RunConv2DWithBias()
624 ops::Conv2D::Attrs().ExplicitPaddings(explicit_paddings)); in RunConv2DWithBias()
640 ops::Conv2D conv = ops::Conv2D( in RunConv2DWithBiasAndActivation()
645 ops::Conv2D::Attrs().ExplicitPaddings(explicit_paddings)); in RunConv2DWithBiasAndActivation()
674 ops::Conv2D conv = ops::Conv2D( in RunConv2DWithBatchNorm()
679 ops::Conv2D::Attrs().ExplicitPaddings(explicit_paddings)); in RunConv2DWithBatchNorm()
704 ops::Conv2D conv = ops::Conv2D( in RunConv2DWithBatchNormAndActivation()
709 ops::Conv2D::Attrs().ExplicitPaddings(explicit_paddings)); in RunConv2DWithBatchNormAndActivation()
|
| /external/tensorflow/tensorflow/core/ops/compat/ops_history_v1/ |
| D | Conv2D.pbtxt | 2 name: "Conv2D"
61 name: "Conv2D"
133 name: "Conv2D"
206 name: "Conv2D"
288 name: "Conv2D"
|
| /external/tensorflow/tensorflow/core/ops/compat/ops_history_v2/ |
| D | Conv2D.pbtxt | 2 name: "Conv2D"
61 name: "Conv2D"
133 name: "Conv2D"
206 name: "Conv2D"
288 name: "Conv2D"
|
| /external/tensorflow/tensorflow/python/keras/benchmarks/keras_examples_benchmarks/ |
| D | cifar10_cnn_benchmark_test.py | 40 tf.keras.layers.Conv2D(
43 model.add(tf.keras.layers.Conv2D(32, (3, 3)))
48 model.add(tf.keras.layers.Conv2D(64, (3, 3), padding='same'))
50 model.add(tf.keras.layers.Conv2D(64, (3, 3)))
|
| /external/tensorflow/tensorflow/core/grappler/optimizers/ |
| D | generic_layout_optimizer_transposer_test.cc | 111 auto conv2d = ops::Conv2D( in SimpleConv2D()
113 {1, kStride1, kStride2, 1}, "SAME", ops::Conv2D::DataFormat(kSrcFormat)); in SimpleConv2D()
296 Output conv2d = ops::Conv2D( in CreateSimpleAddN()
298 {1, 2, 4, 1}, "SAME", ops::Conv2D::DataFormat(kSrcFormat)); in CreateSimpleAddN()
321 ops::Conv2D(scope.WithOpName("conv2d_1").WithDevice("/device:GPU:0"), in CreateSimpleIdentityN()
323 ops::Conv2D::DataFormat(kDstFormat)); in CreateSimpleIdentityN()
331 ops::Conv2D(scope.WithOpName("conv2d_2").WithDevice("/device:GPU:0"), in CreateSimpleIdentityN()
333 ops::Conv2D::DataFormat(kSrcFormat)); in CreateSimpleIdentityN()
387 Output conv2d = ops::Conv2D( in ReduceTransposerKeepDims()
389 {1, 2, 4, 1}, "SAME", ops::Conv2D::DataFormat(kSrcFormat)); in ReduceTransposerKeepDims()
[all …]
|
| /external/tensorflow/tensorflow/compiler/jit/tests/ |
| D | keras_imagenet_main.pbtxt | 18194 name: "res5a_branch1_1/Conv2D/ReadVariableOp"
18242 name: "res4a_branch1_1/Conv2D/ReadVariableOp"
18290 name: "res3a_branch1_1/Conv2D/ReadVariableOp"
18338 name: "res2a_branch1_1/Conv2D/ReadVariableOp"
18386 name: "conv1_1/Conv2D/ReadVariableOp"
18453 name: "res2a_branch2c_1/Conv2D/ReadVariableOp"
18501 name: "res2a_branch2b_1/Conv2D/ReadVariableOp"
18549 name: "res2a_branch2a_1/Conv2D/ReadVariableOp"
18597 name: "res2b_branch2c_1/Conv2D/ReadVariableOp"
18645 name: "res2b_branch2b_1/Conv2D/ReadVariableOp"
[all …]
|
| /external/tensorflow/tensorflow/lite/micro/testing/ |
| D | generate_test_models.py | 44 tf.keras.layers.Conv2D(16, 3, activation="relu", input_shape=input_shape))
45 model.add(tf.keras.layers.Conv2D(32, 3, activation="relu"))
|