- Sort Score
- Result 10 results
- Languages All
Results 1 - 2 of 2 for min_elements_for_weights_ (0.34 sec)
-
tensorflow/compiler/mlir/lite/transforms/prepare_quantize_dynamic_range.cc
!quant_specs_.disable_per_channel; enable_dynamic_range_per_channel_quantization_for_dense_layers_ = !quant_specs_.disable_per_channel_for_dense_layers; min_elements_for_weights_ = quant_specs_.minimum_elements_for_weights; } // The function might contain stats ops which are redundant for processing // dynamic range quantization. And stats ops may cause conflict while
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 20.8K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/passes.td
"enable-dynamic-range-per-channel-quantization-for-dense-layers", "bool", "true", "Whether enable per-channel quantized weights for Fully Connected layers (default is per tensor).">, Option<"min_elements_for_weights_", "min-elements-for-weights", "int64_t", "1024", "The minimum number of elements in a weights array required to apply quantization.">, Option<"enable_float16_quantization_",
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Apr 24 20:30:06 UTC 2024 - 22.6K bytes - Viewed (0)