-
训练报错
KeyError: 'YOLOX: \'`cfg` or `default_args` must contain the key "type", but got {\\\'deepen_factor\\\': 1.0, \\\'widen_factor\\\': 1.0}\\nNone\''
我的配置文件如下：
`_base_ = ['../_base_/schedules/s…
-
Hi,
I wanted to fine tune a pre trained YOLACT model with my own custom data.
I have successfully done so already with a Mask-RCNN.
When loading the pre trained weights, the `torch.load_stat…
-
Hi! I've edited the coco.py file, inside dataset, with my custom dataset which works in coco format. I have not understand how I can train the network, in my case I will need the Mask R-CNN, with my c…
-
Dear all,
I have a general question about the shape of my validation loss. I use Foveabox on my costum dataset. The config file is the following:
```
dataset_type = 'CocoDataset'
data_root = …
-
I'm trying to understand the source code of mmdetection. When I define train_pipeline=[dict(type='Collect',keys=['img', 'gt_bboxes','gt_labels']', I'm trying to find out how 'gt_bboxes' and 'gt_label…
-
mmdet version 2.15.1 yolact_head.py function loss line176-192 filter out invalid anchors and GT just match the valid anchor；but line196-232 prediction for all anchors .There is an err…
-
When I want to use NAS FPN, how should I set the "stack_times"
-
**Describe the bug**
CocoDataset returns error no "ann_info" key when test_mode =True, It's ok when test_mode= False
**Code**
![image](https://user-images.githubusercontent.com/16681534/1218654…
-
I use the model of YOLOX, and the pipelines for train and val are different. But I saw in the Train.py line 168 it writes: val_dataset.pipeline = cfg.data.train.pipeline, I tried to fixed the train.pi…
-
Hello, it seems like I cant train YOLACT on a single GPU when using a custom dataset that isn't 550 x 550. I am going through the configs to understand why this may be. Is this an issue you encountere…