-
In dataset config I set
`evaluation = dict(interval=1, metric='mAP', save_best='mAP')`
watched the model be trained for an epoch before the first evaluation, where I got an error
`KeyError: 'metric…
-
I was trying to train retinanet_swin-t-p4-w7_fpn_1x_coco with the following config:
```
classes = ['neck', 'rotten', 'sprouting', "tip", 'without_skin']
pretrained = 'https://github.com/SwinTransfo…
-
`2022-03-27 08:15:36,546 - mmdet - INFO - workflow: [('train', 1)], max: 36 epochs
2022-03-27 08:15:36,546 - mmdet - INFO - Checkpoints will be saved to /kaggle/working/mmdetection0224/work_dirs/fast…
MrWcy updated
2 years ago
-
https://github.com/open-mmlab/mmdeploy/blob/f2594c624b67910e55e24418832bd96685425b2f/mmdeploy/codebase/mmdet/deploy/object_detection_model.py#L586
This function only return the class name for open …
-
Here is my config file:
# model settings
model = dict(
type = 'YOLOV3',
pretrained = '/data0/steelai/mng/webapp/upload/preTrainedModel/2021/9/10/613b2624cc7ca91773574ffbZm72pD.pth',
b…
-
Thanks for your error report and we appreciate it a lot.
**Checklist**
1. I have searched related issues but cannot get the expected help.
2. I have read the [FAQ documentation](https://mmdetec…
-
Hi, first things first.. This lib is amazing and helped a lot to understand the errors related to the detections.
I was doing using this project for the initial evaluation but since there was no sup…
-
Issue description: I'm not able to understand how to prepare config file, checkpoint file for inferencing of Swin-Transformer object detection.
I used 'Swin-Transformer-Object-Detection/configs/sw…
-
Thanks for your error report and we appreciate it a lot.
**Checklist**
1. I have searched related issues but cannot get the expected help.
2. I have read the [FAQ documentation](https://mmdetec…
-
To train any model, it's necessary to build a dataset. In this matter, the dataset has to have images and text.
- [x] To precise the model for this project
- [x] To find a dataset to train the mod…