- Sort Score
- Result 10 results
- Languages All
Results 1 - 6 of 6 for lowerings (0.14 sec)
-
tensorflow/compiler/mlir/tensorflow/transforms/tf_passes.td
This op is rewritten to generic ops that perform the scale and shift and can operate on non-quantized types. Currently, TF_DequantizeOp is the only op with a lowering that falls in this category. When more lowerings are added (e.g. QuantizeV2Op), they should be added to this pass. }]; let constructor = "TF::CreateLowerQuantizedPass()"; }
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Jun 12 21:18:05 UTC 2024 - 99.6K bytes - Viewed (0) -
src/cmd/compile/internal/ssa/_gen/ARM64.rules
(RoundToEven ...) => (FRINTND ...) (Trunc ...) => (FRINTZD ...) (FMA x y z) => (FMADDD z x y) (Sqrt32 ...) => (FSQRTS ...) (Min(64|32)F ...) => (FMIN(D|S) ...) (Max(64|32)F ...) => (FMAX(D|S) ...) // lowering rotates // we do rotate detection in generic rules, if the following rules need to be changed, check generic rules first.
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Thu May 23 15:49:20 UTC 2024 - 113.1K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/optimize.cc
if (is_output_shape_dynamic && is_broadcast_shape_dynamic) return rewriter.notifyMatchFailure( loc, "output_rank or broadcast_to shape not supported"); // Allow lowering when the input's elements type is F32, BFloat16, I32 or // I16. if (!(mlir::isa<BFloat16Type, Float32Type>(element_type) || element_type.isInteger(32) || element_type.isInteger(16)))
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Apr 30 00:40:15 UTC 2024 - 102.3K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/stablehlo/transforms/legalize_hlo.cc
} return expanded_input; } #include "tensorflow/compiler/mlir/lite/stablehlo/transforms/generated_legalize_hlo.inc" /// Performs the lowering to TF dialect. void LegalizeHloToTf::runOnOperation() { MLIRContext& context = getContext(); mlir::ModuleOp module = getOperation(); RewritePatternSet patterns(&getContext());
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 154.9K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/ir/tf_ops_n_z.cc
} bool StridedSliceOp::GetSlicedBoundRanges( SmallVectorImpl<int64_t> *slice_begin, SmallVectorImpl<int64_t> *slice_end, SmallVectorImpl<int64_t> *slice_stride) { // TODO(hinsu): Support lowering for ops with dynamic begin and end values // when it is possible to derive indices based on mask attributes. DenseIntElementsAttr sparse_begin_attr, sparse_end_attr, sparse_strides_attr;
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu May 09 22:07:10 UTC 2024 - 170.8K bytes - Viewed (0) -
src/cmd/compile/internal/ssa/_gen/generic.rules
// Putting [1]*byte and similar into direct interfaces. (IMake _typ (ArrayMake1 val)) => (IMake _typ val) (ArraySelect [0] (IData x)) => (IData x) // string ops // Decomposing StringMake and lowering of StringPtr and StringLen // happens in a later pass, dec, so that these operations are available // to other passes for optimizations. (StringPtr (StringMake (Addr <t> {s} base) _)) => (Addr <t> {s} base)
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Thu May 16 22:21:05 UTC 2024 - 135.3K bytes - Viewed (0)