-
### System information
- **What is the top-level directory of the model you are using**: object_detection
- **Have I written custom code (as opposed to using a stock example script provided in Tenso…
yryun updated
4 years ago
-
### pytorch>=1.1
I notice that both the RGB and BGR input of `[n,c,h,w]` can get a good result(BGR is slightly higher).
```
model = fasterrcnn_resnet50_fpn(pretrained=True)
model.eval()
## …
-
Hello,
I found that the 'anchor_scale' is being used in faster_rcnn config files, and have tried to see what it does exactly.
Whenever I tried to change this from [8] to some other values or multi…
-
I tried two approaches for reproducing [the performance of Keypoint R-CNN with ResNet-50](https://pytorch.org/blog/torchvision03/#detection-models), box AP = 54.6, keypoint AP = 65.0:
a) use pretrain…
-
-
When you pass in bounding boxes for training that are beyond the dimensions of the images, no error is thrown.
-
## 🐛 Bug
Cannot export torchvision.models.detection.fasterrcnn_resnet50_fpn to `ONNX` format.
There already many reports about this error, such [issue1](https://github.com/pytorch/pytorch/issues…
-
I'm trying to fine tuning the _fasterrcnn_resnet50_fpn_ model using the scripts in _vision/references/detection/_.
As sanity check first of all i have trained the model with `train.py` script with …
-
Hi Yuxin, in multi-GPU setting, to compute one loss, I need to use the `nccl` ops to reduce-sum one variable. Some code pieces are as follows:
one_scalar_sum= tf.reduce_sum(one_scalar, name='on…
-
RT
i used 1mmdection 1.0 version
here is my config :
```
# model settings
model …