-
Буду тут рассказывать что пробовал и какой вышел результат.
* Пока лучший результат дала модель FPN из репы qubvel. 0.722 Local. 0.708 LB
Хочу понять какой способ мерджить предикты в конце лучше…
-
Hi, I hope everyone is doing well. I want to make an object detector with yoloV4. my input images are of size from 200(width)x60(Height) to 400x150. my question is what shall be my height and width in…
-
## ❓ Questions and Help
Hello :)
I found some different configs in [data2vec_vision README](https://github.com/facebookresearch/data2vec_vision/tree/main/beit) and paper.
#### Q1: weight_decay
…
-
great work. thank you for sharing this source.
But i have a little question about the parameter of large feature map and boxaccv2 score in CUB dataset.
1. I found that `large_feature_map` paramete…
-
## 🐛 Bugs / Unexpected behaviors
Hi everyone I am experiencing an error when importing these libraries for pytorchvideo with python 3.12.1. I have tried using python 3.11.7 version but also received …
-
Hi, I'm trying to train YOLO v3 darknet with a custom dataset, but whenever I start the training, it shows that trained :0 and no training happens
I'm working on **Google Colab**
This is the traini…
-
Resnet은 2015년에 나온 후로, 많은 논문들에서 새로운 방법을 제시할 때, baseline으로서 사용이 되어 왔습니다.
이는 Resnet이 지닌 심플함과 안정적인 성능 때문일 겁니다.
시간이 지남에 따라서 많은 모델들이 나왔고, 많은 방법론들이 생겼습니다만,
이러한 논문들이 Resnet과 비교해서 얼마나 성능이 뛰어난지에 대해서는 공정하게 비교…
-
Cut mix를 어떻게 하느냐에 따라서 성능 비교를 해볼까요?
현재 cut mix 구조는 아래 코드와 같습니다.
```
def cutmix(image_x, image_y, size_x, size_y):
w, h = image_x.shape[0], image_x.shape[1]
# 복사할 이미지 랜덤 선택
cxs, …
-
Alexey, I'd like to flesh out some of the entries on this page: https://www.ccoderun.ca/darkmark/Configuration.html#ConfigNew
Specifically, some of the ones from the "new" table without any commen…
-
Hello,
I haven't made any modifications to the code. I cloned it, installed the requirements, and ran the script for training. No other options, no custom data.
As you can see, it ran train.py tw…