Skip to main content

Apollo's model deployment and management tool

Project description

amodel

amodel is Apollo's model deployment and management tool.

Install

You can use below command to install amodel.

pip3 install -t /opt/apollo/python_tools amodel

Set environment

If you are running in Apollo docker, you can skip this step. If you are running outside of docker, the following environment needs to be set up.

export APOLLO_ROOT_DIR=your_apollo_dir

How to work

amodel provides the following commands:

  • list. Show models installed in Apollo.
  • info. Show details of a model.
  • install. Install the model to Apollo.
  • remove. Remove the model from Apollo.

List

You can get the installed models in Apollo through the list command.

$ amodel list
Name                |Task_type           |Sensor_type         |Framework           |Date
mask_pillars        |3d_detection        |lidar               |paddlepaddle        |2021-07-30
center_point        |3d_detection        |lidar               |paddlepaddle        |2022-07-22
point_pillars       |3d_detection        |lidar               |paddlepaddle        |2020-12-15
cnnseg16            |3d_segmentation     |lidar               |paddlepaddle        |2018-10-14
cnnseg128           |3d_segmentation     |lidar               |paddlepaddle        |2020-06-17
cnnseg64            |3d_segmentation     |lidar               |paddlepaddle        |2019-05-29
smoke               |3d_detection        |camera              |paddlepaddle        |2019-06-27
3d-yolo             |3d_detection        |camera              |paddlepaddle        |2019-12-08
denseline           |lane_detection      |camera              |paddlepaddle        |2019-05-29
darkSCNN            |lane_detection      |camera              |paddlepaddle        |2020-12-15
tl_detection        |tl_detection        |camera              |paddlepaddle        |2021-01-15
tl_recognition      |tl_recognition      |camera              |paddlepaddle        |2021-01-15

Info

Then you can use the info command to learn more about the details of the model.

$ amodel info point_pillars
name: point_pillars
date: 2020-12-15
task_type: 3d_detection
sensor_type: lidar
framework: paddlepaddle
model_files:
- name: pfe.onnx
  size: 4125
- name: pts_backbone.zip
  size: 16945051
- name: pts_bbox_head.zip
  size: 121150
- name: pts_middle_encoder.zip
  size: 3763
- name: pts_neck.zip
  size: 2420625
- name: pts_voxel_encoder.zip
  size: 17575
- name: rpn.onnx
  size: 18300546
dataset:
- waymo
- kitti
- nusense

Install

You can deploy the model using the install command.

# Install from local
amodel install xxx.zip
# Install from http
amodel install https://xxx.zip

Remove

You can delete models installed in Apollo with the remove command.

amodel remove point_pillars

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

amodel-0.1.1.tar.gz (6.7 kB view details)

Uploaded Source

Built Distribution

amodel-0.1.1-py3-none-any.whl (7.9 kB view details)

Uploaded Python 3

File details

Details for the file amodel-0.1.1.tar.gz.

File metadata

  • Download URL: amodel-0.1.1.tar.gz
  • Upload date:
  • Size: 6.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.13

File hashes

Hashes for amodel-0.1.1.tar.gz
Algorithm Hash digest
SHA256 11efa2b7c1f7c3e512137609d5c89a5a5e9a6c2f4510bc1b007a0389b2fdff4e
MD5 84c33c971a663fe4360a681bdec9aa59
BLAKE2b-256 12ee8d7c69496f501e07bf4b2e36b86af72b998cfd4d0916e85d970fdfdaffe1

See more details on using hashes here.

File details

Details for the file amodel-0.1.1-py3-none-any.whl.

File metadata

  • Download URL: amodel-0.1.1-py3-none-any.whl
  • Upload date:
  • Size: 7.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.13

File hashes

Hashes for amodel-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 e81246c9f8d564dbb2b392b1e4d157dabd308f24043e56cc6bb3a077a5878de0
MD5 0f619ee4220aa240c32403043d281a82
BLAKE2b-256 d43d93da1bf04b5f87f183139376ad0b55fc5182bd554a74ac01dc744e5caef3

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page