lucasjinreal / centernet_pro_max Goto Github PK
View Code? Open in Web Editor NEWExperiments based on CenterNet (more backbones, TensorRT deployment and mask head)
Experiments based on CenterNet (more backbones, TensorRT deployment and mask head)
需要的话可以加wx/qq 355152992
Although I had a lot of troubles running the code, I succeeded in the end! !! This is a great job!
the code for gaussian radius from CenterNet repo was brought from Cornernet and the gaussian_radius() function is wrong, it has been fixed by the author here is the code
def gaussian_radius(det_size, min_overlap):
height, width = det_size
a1 = 1
b1 = (height + width)
c1 = width * height * (1 - min_overlap) / (1 + min_overlap)
sq1 = np.sqrt(b1 ** 2 - 4 * a1 * c1)
r1 = (b1 - sq1) / (2 * a1)
a2 = 4
b2 = 2 * (height + width)
c2 = (1 - min_overlap) * width * height
sq2 = np.sqrt(b2 ** 2 - 4 * a2 * c2)
r2 = (b2 - sq2) / (2 * a2)
a3 = 4 * min_overlap
b3 = -2 * min_overlap * (height + width)
c3 = (min_overlap - 1) * width * height
sq3 = np.sqrt(b3 ** 2 - 4 * a3 * c3)
r3 = (b3 + sq3) / (2 * a3)
return min(r1, r2, r3)
Please refer to this for formula derivation
princeton-vl/CornerNet#110
简单看了一下想训练coco试试,但是在代码里没找到定位数据集path的地方。。。以及怎么指定使用第几个GPU训练,如果单机多卡训练的话应该改哪些设置,谢谢!
Thanks for your amazing work! I have tried to train the code on my own dataset with multi gpus, but it still works on the first one! And I find that the parameter '--num-gpus' not be used! How did you train on 8 gpus?
Thanks for the great job.
So far I m trying to train with "resnet18 + without DCN". But it occurs error as bellow.
File "/home/administrator/anaconda3/envs/CenterNet/lib/python3.6/site-packages/torch/nn/modules/conv.py", line 778, in forward output_padding, self.groups, self.dilation) RuntimeError: Given transposed=1, weight of size 2048 256 4 4, expected input[8, 512, 16, 16] to have 2048 channels, but got 512 channels instead
RuntimeError: Only tuples, lists and Variables supported as JIT inputs, but got dict
我需要改下那个文件呢
Good job @jinfagang
BTW I find a recent implementation for effdet. Try this repo, https://github.com/rwightman/efficientdet-pytorch/blob/master/train.py
Good luck to you,
Dear Sir:
when i run the demo code with CPU mode, nothing detected, the output is :
try load weights from: ./weights/model_0509999.pth
INFO 03.21 16:31:38 checkpoint.py:121: Loading checkpoint from ./weights/model_0509999.pth
tensor([ 1, 80, 112, 168])
tensor([[[18815, 16863, 16887, ..., 2773, 2781, 2757],
[18207, 18703, 18231, ..., 9472, 15063, 15039],
[18728, 18720, 18752, ..., 18243, 18227, 18251],
...,
[18648, 11093, 11760, ..., 7292, 7316, 7324],
[ 167, 18815, 18648, ..., 18670, 12431, 9408],
[18815, 18648, 167, ..., 3, 18773, 16743]]])
tensor(112)
tensor(168)
tensor(18816)
b.pred_boxes: []
b.scores: []
b.pred_classes: []
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.