- Sort Score
- Result 10 results
- Languages All
Results 21 - 30 of 260 for weights (0.18 sec)
-
src/internal/profile/graph.go
ret += edge.Weight } return ret } type edgeList []*Edge func (el edgeList) Len() int { return len(el) } func (el edgeList) Less(i, j int) bool { if el[i].Weight != el[j].Weight { return abs64(el[i].Weight) > abs64(el[j].Weight) } from1 := el[i].Src.Info.PrintableName() from2 := el[j].Src.Info.PrintableName() if from1 != from2 { return from1 < from2 }
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Mon Feb 05 20:59:15 UTC 2024 - 13.1K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/prepare_quantize_helper.h
input.getDefiningOp())) { // Tensors with derived scale are biases, and handled in propagation. if (tensor_property.use_derived_scale) continue; // For weights, use quantization scale inferred from the values. if (failed(processConstantOp(op, input.getDefiningOp(), index, tensor_property, rewriter))) { return failure();
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri May 03 18:01:23 UTC 2024 - 28K bytes - Viewed (0) -
pkg/scheduler/framework/plugins/interpodaffinity/scoring_test.go
}, // Consider Affinity, Anti Affinity and symmetry together. // for Affinity, the weights are: 8, 0, 0, 0 // for Anti Affinity, the weights are: 0, -5, 0, 0 // for Affinity symmetry, the weights are: 0, 0, 8, 0 // for Anti Affinity symmetry, the weights are: 0, 0, 0, -5 {
Registered: Sat Jun 15 01:39:40 UTC 2024 - Last Modified: Fri Dec 15 03:30:06 UTC 2023 - 44.8K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/quantization/lite/quantize_weights_test.cc
LOG(INFO) << quantized_tensor->name()->str() << " " << float_tensor->name()->str(); if (ExpectEqualTensor(quantized_tensor, float_tensor)) { if (quantized && quantized_tensor->name()->str().find("weights")) { // If tensor is quantized, data type and buffer contents can be // different between float and quantized tensors. So do those tests // separately in the test body without checking them here.
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Jun 12 23:15:24 UTC 2024 - 32.3K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/utils/const_tensor_utils.cc
} storage_type = mlir::cast<mlir::IntegerType>(raw_elem_type); } // TFlite uses narrow-range [u]int8 for constant buffers of quantized weights. // Since we don't know which ones are weights, we represent this optimization // as a change in the storage bounds for the type for all constants of this // type. const int bitwidth = storage_type.getIntOrFloatBitWidth();
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue May 07 23:04:40 UTC 2024 - 16.6K bytes - Viewed (0) -
pilot/pkg/networking/core/networkfilter_test.go
Registered: Fri Jun 14 15:00:06 UTC 2024 - Last Modified: Wed Apr 17 22:20:44 UTC 2024 - 25.8K bytes - Viewed (0) -
tensorflow/compiler/mlir/quantization/tensorflow/passes/prepare_quantize_drq.cc
OpSet op_set_; Option<bool> enable_per_channel_quantization_{ *this, "enable-per-channel-quantization", llvm::cl::init(false), llvm::cl::desc("Whether enable per-channel quantized weights.")}; }; // If the weight is applicable to dynamic range quantization, insert Quantize // and Dequantize ops with per-tensor scale. class PrepareDRQQuantizableOp : public OpRewritePattern<arith::ConstantOp> { public:
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 11.5K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/schema/schema_v3b.fbs
SPARSE = 1, DENSE = 2, } table LSHProjectionOptions { type: LSHProjectionType; } table SVDFOptions { rank:int; fused_activation_function:ActivationFunctionType; // For weights-only quantization, use asymmetric quantization for non // constant inputs at evaluation time. asymmetric_quantize_inputs:bool; } // An implementation of TensorFlow RNNCell. table RNNOptions {
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue May 28 14:28:27 UTC 2024 - 30K bytes - Viewed (0) -
pkg/test/loadbalancersim/lb_test.go
wg := sync.WaitGroup{} clientLatencies := make([]timeseries.Data, len(s.mesh.Clients())) for i, client := range s.mesh.Clients() { i := i client := client wg.Add(1) go func() { // Assign weights to the endpoints. var conns []*loadbalancer.WeightedConnection for _, n := range s.mesh.Nodes() { conns = append(conns, s.newWeightedConnection(client, n)) } // Create a load balancer
Registered: Fri Jun 14 15:00:06 UTC 2024 - Last Modified: Thu May 19 23:29:30 UTC 2022 - 11K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/tf_to_tfl_flatbuffer.cc
&q_builder, input_model, quantized_type, use_updated_hybrid_scheme, ::tflite::optimize::QuantizerType::OLD_QUANTIZER) != kTfLiteOk) { return absl::InvalidArgumentError( "Quantize weights transformation failed."); } const uint8_t* q_buffer = q_builder.GetBufferPointer(); *result = std::string(reinterpret_cast<const char*>(q_buffer), q_builder.GetSize()); return absl::OkStatus(); }
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri May 03 18:01:23 UTC 2024 - 23.8K bytes - Viewed (0)