Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/imgcook/pipcook-plugin-detectron-maskrcnn-model
Pipcook plugin for detectron maskrcnn model definition.
https://github.com/imgcook/pipcook-plugin-detectron-maskrcnn-model
Last synced: 3 days ago
JSON representation
Pipcook plugin for detectron maskrcnn model definition.
- Host: GitHub
- URL: https://github.com/imgcook/pipcook-plugin-detectron-maskrcnn-model
- Owner: imgcook
- License: apache-2.0
- Created: 2020-08-12T12:18:04.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2020-10-20T09:38:26.000Z (about 4 years ago)
- Last Synced: 2024-08-02T19:01:36.045Z (4 months ago)
- Language: TypeScript
- Homepage:
- Size: 11.7 KB
- Stars: 0
- Watchers: 8
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-imgcook - imgcook/pipcook-plugin-detectron-maskrcnn-model
README
# @pipcook/plugins-detectron-maskrcnn-model-define
### Description
This plugin defines object detection / instance segmentation model. This model is [maskrcnn](https://arxiv.org/abs/1703.06870). The whole staff is built on top on [detectron2](https://github.com/facebookresearch/detectron2).
### Required Parameters
| Parameter | Type | Description |
|:----------|:-----|:------------|### Optional Parameters
| Parameter | Type | Default Value | Description |
|:----------|:-----|:------|:-----|
|baseLearningRate|number|0.00025|base learning rate|
|numWorkers|number|0|workers for data loader.|
|numClasses|number|0|Number of classes of dataset. Only specify this value when you provide your own model. Otherwise just leave as 0|
|recoverPath|string|null|path to model pretrained|### Example
```json
"modelDefine": {
"package": "@pipcook/plugins-detectron-maskrcnn-model-define"
},
```