Searched refs:quant_model (Results 1 – 7 of 7) sorted by relevance
/external/libopus/training/ |
D | rnn_train.py | 34 def quant_model(model): function 108 quant_model(model) 114 quant_model(model) 124 quant_model(model) 130 quant_model(model) 136 quant_model(model) 142 quant_model(model) 148 quant_model(model) 154 quant_model(model) 160 quant_model(model) [all …]
|
/external/pytorch/test/quantization/core/ |
D | test_workflow_module.py | 985 quant_model = copy.deepcopy(model.module) 986 quant_model = torch.ao.quantization.convert(quant_model.eval().cpu(), inplace=False) 988 out = quant_model(torch.rand(1, 3, 28, 28)) 1298 quant_model = prepare_qat(model, 1302 for name, mod in quant_model.named_modules(): 1308 quant_model(indices) 1311 self.assertEqual(quant_model.emb1.weight_fake_quant.zero_point.dtype, torch.float32) 1312 self.assertEqual(quant_model.emb2.weight_fake_quant.zero_point.dtype, torch.float32) 1314 inference_gm = convert(quant_model.eval().cpu(), 1328 quant_model = prepare_qat(model, [all …]
|
/external/executorch/backends/arm/test/quantizer/ |
D | test_generic_annotater.py | 37 quant_model = tester.quantize().get_artifact() 38 partitions = get_source_partitions(quant_model.graph, [model.op])
|
/external/tensorflow/tensorflow/lite/tools/optimize/debugging/python/ |
D | debugger.py | 292 self.quant_model = convert.mlir_quantize( 300 model_content=self.quant_model)
|
/external/tensorflow/tensorflow/compiler/mlir/lite/quantization/lite/ |
D | quantize_weights_test.cc | 161 std::unique_ptr<ModelT> quant_model = in FindMatchingExpectedTensor() local 183 quant_model->buffers[quantized_tensor->buffer()].get()->data; in FindMatchingExpectedTensor()
|
D | quantize_model_test.cc | 180 const ModelT& quant_model, in FindMatchingExpectedTensor() argument 183 GetBuiltinCode(quant_model.operator_codes[quant_op.opcode_index].get()); in FindMatchingExpectedTensor()
|
/external/pytorch/test/quantization/fx/ |
D | test_quantize_fx.py | 9564 quant_model = convert_fx(prepared_fx_model, 9576 checkQuantized(quant_model) 9604 quant_model = convert_fx(prepared_fx_model, 9616 checkQuantized(quant_model)
|