-
**System information**
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04):16.04
- TensorFlow installed from (source or binary):pip3 install
- TensorFlow version (or github SHA if from source…
-
## Instructions To Reproduce the 🐛 Bug:
1. Full runnable code or full changes you made:
```
If making changes to the project itself, please use output of the following command:
git rev-parse HEAD;…
-
# Prerequisites
Please answer the following questions for yourself before submitting an issue.
- [ yes] I am using the latest TensorFlow Model Garden release and TensorFlow 2.
- [yes ] I am rep…
-
The input images the TensorBoard shows me are strangely augmented in terms of colour saturation and gamma. Any idea how can I fix it? Is it the wrong file format or compression?
My augmentation opt…
-
**System information**
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows 10
- Mobile device (e.g. iPhone 8, Pixel 2, Samsung Galaxy) if the issue happens on mobile device:
- Tensor…
-
I tried to test the model I trained in v1.x in v2.x. Then I found all models drop heavily. Some AP drop 10%, Some even drops more than 30%.
visualizing the detect image, I found the cls probability…
-
Hello.
I have a question regarding the number of feature elvels used in BiFPN. Am I missing something?
The original paper uses 5 feature levels from backbone as input to BiFPN (P3 - P7):
![image]…
-
I try to train EfficientDet-b0 model by this:
gtf.Train(num_epochs=50, model_output_dir="trained/")
but after first epoch I got this error
`Epoch: 1/50. Iteration: 166/166. Cls loss: 0.13480. Reg …
-
Hello @rakhimovv
I am training the model with s0_bv2_bifpn_f64_s3x.yaml config.
When I try to run apply_net.py on sample images, predictions seem to be very off.
Command I am using = "python appl…
-
I noticed your implementaiton is different from the offical's in two places:
```
if self.first_time:
p3, p4, p5 = inputs
p6_in = self.p5_to_p6(p5)
p7_in = s…