-
I use the libra_rcnn to detect two classes ,I changed the backbone based one senet101, when I test my model, I found that threre exists nothing, Here is my config:
```python
model = dict(
type=…
-
I have not changed anything other than anchor-scales in rpn-head. Because I want to increase the number and scale of the anchor
My modifications are as follows：
rpn_head=dict(
type='RPN…
-
## ❓ Questions and Help
I want to use Faster RCNN for training with ROI Pooling.
I use this command:
python tools/train_net.py --config-file "configs/e2e_faster_rcnn_R_50_FPN_1x.yaml" SOLVER.IMS_P…
-
Hi, thanks for the code! I tried to run the code with 8 V100 GPUs. However, all of the GPUs were just working alternatively during the training phase. The estimate time was more than 6 days. I checked…
-
## ❓ Questions and Help
Currently I am trying to install maskrcnn-benchmark on my space in a server where I want to do some tests. I was able to compile the repo on my machine using the CPU and eve…
-
I use custom dataset and cascade_rcnn_x101_64x4d_fpn_1x and the training run successf…
-
i want to use the fasterRCNN_FPN to do detection. But the num_class of pretrained model on coco dataset is 81, mine dataset just has one class, And it throw an exception.
File "/home/ed/anaconda…
-
I'm a freshman of caffe2. When I run the training code,
python2 tools/train_net.py \
--config_file configs/ava_r50_baseline.yaml \
CHECKPOINT.DIR /tmp/baseline-output
we meet the following…
-
I train data with one class, it's failure。
the config is modified on faster_rcnn_r50_fpn_1x_voc0712.py。 I change the num_classes to 2 ,it will failure, however when the num_classes is 21, it …
-
Hi,
Do you have the plan to implement a ROIAlign layer (instead of ROI Pooliing) for Faster R-CNN?
Thanks,