-
Hello,
I download the provided ckpt files and use `python test.py --img-size 736 --conf_thres 0.001` for reproducing results, but I got different AP and AP50 when compare with the table in readme.
…
-
the tutorial states that we can set input image size to any desired value .however the paper states as below
**d) Input image resolution: Since the feature levels 3–7 are used in BiFPN, the input res…
-
Hi @mingxingtan , I am curious about the influence of shortcut connection in the BiFPN structure.
Q1: Will these connections affect the performance a lot in BiFPN?
Q2: From the table5, I suppose t…
-
Hi I've done every step following your instruction and everything's fine up to training the model as follows and I stop it with Cntl_C...
I0905 18:37:49.263145 17124 model_lib_v2.py:652] Step 3900 pe…
-
**Versions:**
TF-GPU - 1.15
Keras - 2.2.5(as needed in requirements.txt)
Code - Latest pull of the repo and no changes have been made
I have been trying to train EfficientDet on 8*K80 GPUS. But…
-
The code for biFPN is bit complicated to understand, so some resource to learn it and also some re implement it will be helpful
-
Hi, thanks for sharing this project.
I want to add FPN for object detection, How can I export some specific layers?
for example, with backbone Resnet style, they defined layer name by : layer1, laye…
-
Traceback (most recent call last):
File "efficientdet_test_videos.py", line 39, in
model.load_state_dict(torch.load(f'logs/new/efficientdet-d1_280_123602.pth'))
File "/usr/local/lib/python…
-
[Info] loaded weights: efficientdet-d8.pth, resuming checkpoint from step: 0
[Error] Traceback (most recent call last):
File "train.py", line 228, in train
cls_loss, reg_loss = model(imgs, …
-