- Sort Score
- Result 10 results
- Languages All
Results 1 - 7 of 7 for 16x16 (0.04 sec)
-
tensorflow/compiler/mlir/lite/python/graphdef_to_tfl_flatbuffer.cc
pass_config.legalize_custom_tensor_list_ops = toco_flags.legalize_custom_tensor_list_ops(); // Disable the unfolding of the 16x16 TF::BatchMatMulOp to avoid the // conversion to an unsupported 16x16 TFL::FullyConnectedOp. if (toco_flags.inference_type() == toco::IODataType::QUANTIZED_INT16) { pass_config.unfold_batch_matmul = false; }
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon Mar 11 19:29:56 UTC 2024 - 5.1K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/python/jax_to_tfl_flatbuffer.cc
pass_config.unfold_batch_matmul = toco_flags.unfold_batchmatmul(); pass_config.lower_tensor_list_ops = toco_flags.lower_tensor_list_ops(); // Disable the unfolding of the 16x16 TF::BatchMatMulOp to avoid the // conversion to an unsupported 16x16 TFL::FullyConnectedOp. if (toco_flags.inference_type() == toco::IODataType::QUANTIZED_INT16) { pass_config.unfold_batch_matmul = false; }
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon Mar 11 19:29:56 UTC 2024 - 8K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/tests/optimize_no_verify.mlir
%0 = "tfl.conv_2d"(%arg0, %arg1, %cst_0) {dilation_h_factor = 2 : i32, dilation_w_factor = 3 : i32, fused_activation_function = "NONE", padding = "SAME", stride_h = 4 : i32, stride_w = 5 : i32} : (tensor<256x32x32x3xf16>, tensor<16x3x3x3xf16>, tensor<16xf16>) -> tensor<256x8x7x16xf16>
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu May 02 09:41:17 UTC 2024 - 5.8K bytes - Viewed (0) -
src/net/netip/slow_test.go
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Mon Mar 11 19:54:31 UTC 2024 - 5.8K bytes - Viewed (0) -
src/crypto/aes/asm_s390x.s
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Mon Mar 04 17:29:44 UTC 2024 - 4.4K bytes - Viewed (0) -
.github/workflows/update-rbe.yml
map sigbuild-r2.14-clang-python3.9 2.14-python3.9 map sigbuild-r2.14-clang-python3.10 2.14-python3.10 map sigbuild-r2.14-clang-python3.11 2.14-python3.11 # TF 2.16 map sigbuild-r2.16 2.16-python3.11 map sigbuild-r2.16-python3.9 2.16-python3.9 map sigbuild-r2.16-python3.10 2.16-python3.10 map sigbuild-r2.16-python3.11 2.16-python3.11 map sigbuild-r2.16-python3.12 2.16-python3.12
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Apr 10 15:40:34 UTC 2024 - 7.2K bytes - Viewed (0) -
src/runtime/sys_windows_amd64.s
// In any case, even if function has 0,1,2,3,4 args, there is reserved // but uninitialized "shadow space" for the first 4 args. // The values are in registers. MOVQ CX, (16+0)(SP) MOVQ DX, (16+8)(SP) MOVQ R8, (16+16)(SP) MOVQ R9, (16+24)(SP) // R8 = address of args vector LEAQ (16+0)(SP), R8 // remove return address from stack, we are not returning to callbackasm, but to its caller. MOVQ 0(SP), AX ADDQ $8, SP
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Mon Feb 19 07:24:08 UTC 2024 - 8.4K bytes - Viewed (0)