Searched refs:dynamically_quantize_per_channel (Results 1 – 4 of 4) sorted by relevance
| /external/pytorch/benchmarks/gpt_fast/ |
| D | mixtral_moe_quantize.py | 12 def dynamically_quantize_per_channel(x, quant_min, quant_max, target_dtype): function
82 int8_weight, scales, _ = dynamically_quantize_per_channel(
97 bit8_weight, scales, _ = dynamically_quantize_per_channel(
|
| D | quantize.py | 10 def dynamically_quantize_per_channel(x, quant_min, quant_max, target_dtype): function
69 int8_weight, scales, _ = dynamically_quantize_per_channel(
|
| /external/executorch/examples/models/llama/tests/ |
| D | test_pre_quantization_transforms.py | 18 dynamically_quantize_per_channel,
104 weight_int8, scales, _ = dynamically_quantize_per_channel(
|
| /external/executorch/examples/models/llama/source_transformation/ |
| D | quantize.py | 176 def dynamically_quantize_per_channel( function
367 weight, scales, _ = dynamically_quantize_per_channel(
590 weight, scales, _ = dynamically_quantize_per_channel(
|