-
# code
## plateau
```
10634* python test/pspnet_test.py --backbone_pretrained=True --use_momentum=True --learning_rate=1e-2 --use_lr_mult=True --midnet_scale=10 --upsample_type=bilinear --upsampl…
-
Trained on the bloodcell dataset, multi label class loss plateau at epoch of 10 by the value of ~80 and is still ~80 at epoch of 50, but the other loss (xywh and objectiveness loss) are able to reach …
-
您好，我在阅读您的代码时，遇到了下面的问题：
```
def trans_mvsnet_loss(inputs, depth_gt_ms, mask_ms, **kwargs):
depth_loss_weights = kwargs.get("dlossw", None)
total_loss = torch.tensor(0.0, dtype=torch.float32…
-
# 1.将模型存储
model.save('my_model.h5')
# 2.加载模型报错
from nets.centernet_training import focal_loss, reg_l1_loss
from keras.models import load_model
model = load_model('my_model.h5', compile=False, cus…
-
File "/home/neousys/cjg/CenterPoint-KITTI/pcdet/models/dense_heads/centerpoint_head_single.py", line 687, in gaussian_focal_loss
pos_loss = (-(pred + eps).log() * (1 - pred).pow(alpha) * pos_weights)…
-
### What is the feature?
I want to know what is the difference between the following two cases when using the Multiple Losses?
1：
model = dict(
decode_head=dict(loss_decode=[dict(type='Cross…
-
### Prerequisite
- [X] I have searched [Issues](https://github.com/open-mmlab/mmdetection3d/issues) and [Discussions](https://github.com/open-mmlab/mmdetection3d/discussions) but cannot get the exp…
-
been training images on my model with masks for earrings and not earrings.
```
dice_loss = sm.losses.DiceLoss()
focal_loss = sm.losses.BinaryFocalLoss()
total_loss = dice_loss + (1 * focal_loss…
ghost updated
5 years ago
-
```
class categorical_focal_loss:
'''
Softmax version of focal loss.
m
FL = sum -alpha * (1 - p_o,c)^gamma * y_o,c * log(p_o,c)
…
-
Hi, thanks for your wonderful work! I have a question that whether your proposed KnnFocalLike loss is better than Focal loss. I do not find the comparison experiment in your paper.