olive-recipes/sam2.1-hiera-small/QNN/sam21_vision_encoder_qnn_ctx.json at main · microsoft/olive-recipes · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
{
    "input_model": {
        "type": "ONNXModel",
        "model_path": "sam21_vision_encoder.onnx"
    },
    "systems": {
        "local_system": {
            "type": "LocalSystem",
            "accelerators": [
                {
                    "device": "gpu",
                    "execution_providers": [
                        "CUDAExecutionProvider"
                    ]
                }
            ]
        },
        "target_system": {
            "type": "PythonEnvironment",
            "python_environment_path": "/path/to/qnn/env/bin",
            "accelerators": [
                {
                    "execution_providers": [
                        "QNNExecutionProvider"
                    ]
                }
            ]
        }
    },
    "data_configs": [
        {
            "name": "latency_data_config",
            "user_script": "user_script.py",
            "load_dataset_config": {
                "type": "local_dataset"
            },
            "dataloader_config": {
                "type": "ve_data_loader",
                "batch_size": 1
            }
        },
        {
            "name": "quantize_data_config",
            "user_script": "user_script.py",
            "load_dataset_config": {
                "type": "local_dataset"
            },
            "dataloader_config": {
                "type": "ve_quantize_data_loader",
                "data_num": 100
            }
        }
    ],
    "evaluators": {
        "common_evaluator": {
            "metrics": [
                {
                    "name": "latency",
                    "type": "latency",
                    "data_config": "latency_data_config",
                    "sub_types": [
                        {
                            "name": "avg",
                            "priority": 1
                        }
                    ]
                }
            ]
        }
    },
    "passes": {
        "surgeries": {
            "type": "GraphSurgeries",
            "surgeries": [
                {
                    "surgeon": "RenameInputs",
                    "old_names": [
                        "input.1"
                    ],
                    "new_names": [
                        "pixel_values"
                    ]
                }
            ]
        },
        "quantization": {
            "type": "OnnxStaticQuantization",
            "data_config": "quantize_data_config",
            "activation_type": "uint8",
            "precision": "uint8",
            "calibrate_method": "MinMax",
            "quant_preprocess": true,
            "calibration_providers": [
                "CUDAExecutionProvider"
            ]
        },
        "cb": {
            "type": "EPContextBinaryGenerator",
            "provider_options": {
                "htp_performance_mode": "burst",
                "htp_graph_finalization_optimization_mode": "3",
                "offload_graph_io_quantization": "0",
                "soc_model": "60"
            },
            "weight_sharing": false
        }
    },
    "log_severity_level": 1,
    "ort_log_severity_level": 1,
    "ort_py_log_severity_level": 1,
    "evaluate_input_model": false,
    "host": "local_system",
    "target": "target_system",
    "cache_dir": "cache_encoder",
    "output_dir": "model/encoder"
}