-
The author wrote following words in paper:
Additionally, we found that it was important to put very little or no weight decay (l2 regularization) on the depthwise filters since their are so few param…
-
同样的数据集，为什么我用resnet一下就收敛了，但是moblienet却收敛不了，是网络特性的原因吗
-
Hello AlexeyAB!
We all know that MoblieNet and StuffNet are more and more popular in industrial products. But there are no depthwise_convolutional_layer in DarkNet frame，so It is difficult to buildin…
-
`tflite_flutter: ^0.10.1`
I trained model using ssd_moblienet_v1(or ssd_mobilenet_v2) and made tflite model.
```
static Future _loadModel() async {
dev.log('Loading interpreter options.…
-
我用老版本或者是新版本的代码跑过有预训练权重和无预训练权重的m2和m3，跑的是kitti数据集，为什么得到的结果那么低，只有20%多，（正常应该是70多才对），是关于数据集的问题吗？
:
ubuntu 16.04
Python version:
2.7
Source framework with version (like Tensorflow 1.4.1 with GPU):
Tensorflow 1.9
Destination framework with version (like CNTK …
-
Hello,
I changed the v to 1 but it doesn't detect persons
-
### 深層学習day3
**Section1：再帰型ニューラルネットワークの概念**
【要点】
再帰型ニューラルネットワーク(RNN)とは時間的順序を追って一定間隔ごとに観測され、相互に統計的依存関係が認められるデータ系列(時系列データ)に対応可能なニューラルネットワークである。
RNNの構成例を下記に示す。
![image](https://user-images.githubu…
-
你好 我想问一下有尝试过把ufldv2的backbone换成其他的模型嘛
-
Hi, I try to replace the backbone of mask rcnn to mobilenetv2. And I firstly get the error like that.
```
RuntimeError: Expected to have finished reduction in the prior iteration before starting a…