Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix: update machine learning models download intructions #589

Merged
merged 7 commits into from
Jul 23, 2024
Merged

fix: update machine learning models download intructions #589

Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
a9ef280
fix: update machine learning models update intructions
badai-nguyen Jul 11, 2024
f864932
style(pre-commit): autofix
pre-commit-ci[bot] Jul 11, 2024
2a826eb
fix: add transfusion download
badai-nguyen Jul 13, 2024
de49241
fix: update trafic light fine detection model
badai-nguyen Jul 13, 2024
e22939b
fix: update model download instrution
badai-nguyen Jul 16, 2024
41f9e6e
style(pre-commit): autofix
pre-commit-ci[bot] Jul 16, 2024
749739c
typo: add console
badai-nguyen Jul 23, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
100 changes: 14 additions & 86 deletions docs/models/index.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,103 +1,31 @@
# Machine learning models

The Autoware perception stack uses models for inference. These models are automatically downloaded if using `ansible`, but they can also be downloaded manually.
The Autoware perception stack uses models for inference. These models are automatically downloaded as part of the `setup-dev-env.sh` script.

## ONNX model files
The models are hosted by Web.Auto.

### Download instructions
Default models directory (`data_dir`) is `~/autoware_data`.

The ONNX model files are stored in a common location, hosted by Web.Auto
## Download instructions

Any tool that can download files from the web (e.g. `wget` or `curl`) is the only requirement for downloading these files:

```console
# yabloc_pose_initializer

$ mkdir -p ~/autoware_data/yabloc_pose_initializer/
$ wget -P ~/autoware_data/yabloc_pose_initializer/ \
https://s3.ap-northeast-2.wasabisys.com/pinto-model-zoo/136_road-segmentation-adas-0001/resources.tar.gz


# image_projection_based_fusion

$ mkdir -p ~/autoware_data/image_projection_based_fusion/
$ wget -P ~/autoware_data/image_projection_based_fusion/ \
https://awf.ml.dev.web.auto/perception/models/pointpainting/v4/pts_voxel_encoder_pointpainting.onnx \
https://awf.ml.dev.web.auto/perception/models/pointpainting/v4/pts_backbone_neck_head_pointpainting.onnx
Please follow the download instruction in [autoware download instructions](https://github.com/autowarefoundation/autoware/blob/main/ansible/roles/artifacts/README.md#L15) for updated models downloading.

The models can be also downloaded manually using download tools such as `wget` or `curl`. The latest urls of weight files and param files for each model can be found at [autoware main.yaml file](https://github.com/autowarefoundation/autoware/blob/main/ansible/roles/artifacts/tasks/main.yaml)

# lidar_apollo_instance_segmentation

$ mkdir -p ~/autoware_data/lidar_apollo_instance_segmentation/
$ wget -P ~/autoware_data/lidar_apollo_instance_segmentation/ \
https://awf.ml.dev.web.auto/perception/models/lidar_apollo_instance_segmentation/vlp-16.onnx \
https://awf.ml.dev.web.auto/perception/models/lidar_apollo_instance_segmentation/hdl-64.onnx \
https://awf.ml.dev.web.auto/perception/models/lidar_apollo_instance_segmentation/vls-128.onnx

The example of downloading `lidar_centerpoint` model:

```console
# lidar_centerpoint

$ mkdir -p ~/autoware_data/lidar_centerpoint/
$ wget -P ~/autoware_data/lidar_centerpoint/ \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/pts_voxel_encoder_centerpoint.onnx \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/pts_backbone_neck_head_centerpoint.onnx \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/pts_voxel_encoder_centerpoint_tiny.onnx \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/pts_backbone_neck_head_centerpoint_tiny.onnx


# tensorrt_yolo

$ mkdir -p ~/autoware_data/tensorrt_yolo/
$ wget -P ~/autoware_data/tensorrt_yolo/ \
https://awf.ml.dev.web.auto/perception/models/yolov3.onnx \
https://awf.ml.dev.web.auto/perception/models/yolov4.onnx \
https://awf.ml.dev.web.auto/perception/models/yolov4-tiny.onnx \
https://awf.ml.dev.web.auto/perception/models/yolov5s.onnx \
https://awf.ml.dev.web.auto/perception/models/yolov5m.onnx \
https://awf.ml.dev.web.auto/perception/models/yolov5l.onnx \
https://awf.ml.dev.web.auto/perception/models/yolov5x.onnx \
https://awf.ml.dev.web.auto/perception/models/coco.names


# tensorrt_yolox

$ mkdir -p ~/autoware_data/tensorrt_yolox/
$ wget -P ~/autoware_data/tensorrt_yolox/ \
https://awf.ml.dev.web.auto/perception/models/yolox-tiny.onnx \
https://awf.ml.dev.web.auto/perception/models/yolox-sPlus-opt.onnx \
https://awf.ml.dev.web.auto/perception/models/yolox-sPlus-opt.EntropyV2-calibration.table \
https://awf.ml.dev.web.auto/perception/models/object_detection_yolox_s/v1/yolox-sPlus-T4-960x960-pseudo-finetune.onnx \
https://awf.ml.dev.web.auto/perception/models/object_detection_yolox_s/v1/yolox-sPlus-T4-960x960-pseudo-finetune.EntropyV2-calibration.table \
https://awf.ml.dev.web.auto/perception/models/label.txt


# traffic_light_classifier

$ mkdir -p ~/autoware_data/traffic_light_classifier/
$ wget -P ~/autoware_data/traffic_light_classifier/ \
https://awf.ml.dev.web.auto/perception/models/traffic_light_classifier/v2/traffic_light_classifier_mobilenetv2_batch_1.onnx \
https://awf.ml.dev.web.auto/perception/models/traffic_light_classifier/v2/traffic_light_classifier_mobilenetv2_batch_4.onnx \
https://awf.ml.dev.web.auto/perception/models/traffic_light_classifier/v2/traffic_light_classifier_mobilenetv2_batch_6.onnx \
https://awf.ml.dev.web.auto/perception/models/traffic_light_classifier/v2/traffic_light_classifier_efficientNet_b1_batch_1.onnx \
https://awf.ml.dev.web.auto/perception/models/traffic_light_classifier/v2/traffic_light_classifier_efficientNet_b1_batch_4.onnx \
https://awf.ml.dev.web.auto/perception/models/traffic_light_classifier/v2/traffic_light_classifier_efficientNet_b1_batch_6.onnx \
https://awf.ml.dev.web.auto/perception/models/traffic_light_classifier/v2/lamp_labels.txt


# traffic_light_fine_detector

$ mkdir -p ~/autoware_data/traffic_light_fine_detector/
$ wget -P ~/autoware_data/traffic_light_fine_detector/ \
https://awf.ml.dev.web.auto/perception/models/tlr_yolox_s/v2/tlr_yolox_s_batch_1.onnx \
https://awf.ml.dev.web.auto/perception/models/tlr_yolox_s/v2/tlr_yolox_s_batch_4.onnx \
https://awf.ml.dev.web.auto/perception/models/tlr_yolox_s/v2/tlr_yolox_s_batch_6.onnx \
https://awf.ml.dev.web.auto/perception/models/tlr_yolox_s/v2/tlr_labels.txt


# traffic_light_ssd_fine_detector

$ mkdir -p ~/autoware_data/traffic_light_ssd_fine_detector/
$ wget -P ~/autoware_data/traffic_light_ssd_fine_detector/ \
https://awf.ml.dev.web.auto/perception/models/mb2-ssd-lite-tlr.onnx \
https://awf.ml.dev.web.auto/perception/models/voc_labels_tl.txt
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/pts_backbone_neck_head_centerpoint_tiny.onnx \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/centerpoint_ml_package.param.yaml \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/centerpoint_tiny_ml_package.param.yaml \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/centerpoint_sigma_ml_package.param.yaml \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/detection_class_remapper.param.yaml \
https://awf.ml.dev.web.auto/perception/models/centerpoint/v2/deploy_metadata.yaml
```
Loading