You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
|`model_path`| string | "" | The onnx file name for yolox model |
48
-
|`model_name`| string | "" | The yolox model name: <br /> "yolox-sPlus-T4-960x960-pseudo-finetune" for detection only, could reduce resource and processing_time <br /> "yolox-sPlus-opt-pseudoV2-T4-960x960-T4-seg16cls" for multi-task including semantic segmentation |
49
-
|`label_path`| string | "" | The label file with label names for detected objects written on it |
|`build_only`| bool | false | shutdown node after TensorRT engine file is built |
52
-
|`calibration_algorithm`| string | "MinMax" | Calibration algorithm to be used for quantization when precision==int8. Valid value is one of: Entropy",("Legacy" \| "Percentile"), "MinMax"]|
53
-
|`dla_core_id`| int | -1 | If positive ID value is specified, the node assign inference task to the DLA core |
54
-
|`quantize_first_layer`| bool | false | If true, set the operating precision for the first (input) layer to be fp16. This option is valid only when precision==int8 |
55
-
|`quantize_last_layer`| bool | false | If true, set the operating precision for the last (output) layer to be fp16. This option is valid only when precision==int8 |
56
-
|`profile_per_layer`| bool | false | If true, profiler function will be enabled. Since the profile function may affect execution speed, it is recommended to set this flag true only for development purpose. |
57
-
|`clip_value`| double | 0.0 | If positive value is specified, the value of each layer output will be clipped between [0.0, clip_value]. This option is valid only when precision==int8 and used to manually specify the dynamic range instead of using any calibration |
58
-
|`preprocess_on_gpu`| bool | true | If true, pre-processing is performed on GPU |
59
-
|`calibration_image_list_path`| string | "" | Path to a file which contains path to images. Those images will be used for int8 quantization. |
60
-
|`yolox_s_plus_opt_param_path`| string | "" | Path to parameter file |
61
-
|`is_publish_color_mask`| bool | false | If true, publish color mask for result visualization |
62
-
|`is_roi_overlap_segment`| bool | true | If true, overlay detected object roi onto semantic segmentation to avoid over-filtering pointcloud especially small size objects |
63
-
|`overlap_roi_score_threshold`| float | 0.3 | minimum existence_probability of detected roi considered to replace segmentation |
64
-
|`roi_overlay_segment_label.UNKNOWN`| bool | true | If true, unknown objects roi will be overlaid onto sematic segmentation mask. |
65
-
|`roi_overlay_segment_label.CAR`| bool | false | If true, car objects roi will be overlaid onto sematic segmentation mask. |
66
-
|`roi_overlay_segment_label.TRUCK`| bool | false | If true, truck objects roi will be overlaid onto sematic segmentation mask. |
67
-
|`roi_overlay_segment_label.BUS`| bool | false | If true, bus objects roi will be overlaid onto sematic segmentation mask. |
68
-
|`roi_overlay_segment_label.MOTORCYCLE`| bool | true | If true, motorcycle objects roi will be overlaid onto sematic segmentation mask. |
69
-
|`roi_overlay_segment_label.BICYCLE`| bool | true | If true, bicycle objects roi will be overlaid onto sematic segmentation mask. |
70
-
|`roi_overlay_segment_label.PEDESTRIAN`| bool | true | If true, pedestrian objects roi will be overlaid onto sematic segmentation mask. |
71
-
|`roi_overlay_segment_label.ANIMAL`| bool | true | If true, animal objects roi will be overlaid onto sematic segmentation mask. |
precision: "int8"# Operation precision to be used on inference. Valid value is one of: [fp32, fp16, int8].
30
32
calibration_algorithm: "Entropy"# Calibration algorithm to be used for quantization when precision==int8. Valid value is one of: [Entropy, (Legacy | Percentile), MinMax].
31
33
dla_core_id: -1# If positive ID value is specified, the node assign inference task to the DLA core.
model_path: "$(var data_path)/tensorrt_yolox/$(var model_name).onnx"# The onnx file name for YOLOX model.
4
+
label_path: "$(var data_path)/tensorrt_yolox/label.txt"# The label file path for YOLOX model.
5
+
score_threshold: 0.35# Objects with a score lower than this value will be ignored. This threshold will be ignored if specified model contains EfficientNMS_TRT module in it.
6
+
nms_threshold: 0.7# Detection results will be ignored if IoU over this value. This threshold will be ignored if specified model contains EfficientNMS_TRT module in it.
7
7
precision: "fp16"# Operation precision to be used on inference. Valid value is one of: [fp32, fp16, int8].
8
8
calibration_algorithm: "MinMax"# Calibration algorithm to be used for quantization when precision==int8. Valid value is one of: [Entropy, (Legacy | Percentile), MinMax].
9
9
dla_core_id: -1# If positive ID value is specified, the node assign inference task to the DLA core.
0 commit comments