-
你好，unsky。我在做小目标检测的实验，之前用你的旧代码（VGG16模型）召回率可以达到90%左右，用新的代码（RESNET50 sharedrcnn）召回率只有70%左右。尝试改了scales为[4,8]和FPN/lib/fast_rcnn/config.py里的一些参数，但是效果都不理想，能不能给一些建议，谢谢！
-
@XingangPan Could you tell me what's the source of weights in vgg.t7? The net of part (44) as follows is not the same as the basic vgg16, so I wonder if these weights are obtained from pretraining or…
-
Hi I have been trying to reproduce the results by training an ANN from scratch with the same network model of VGG16 as described in your code. Then I tried to use the main.py to convert my saved model…
-
Hi Eder,
Thanks for the Keras implementation! I'm excited about this idea as well. (Hopefully) quick question: I'm interested in learning an autoencoder for the output features of some other CNN …
-
**Model:**
```
import keras
import openml
import openml_keras
from keras.layers import GlobalAveragePooling2D, Conv2D, BatchNormalization, Dense, MaxPooling2D, Dropout, Flatten, Reshape
from ker…
-
While we handle most of the dependency installation automatically via the install script ( https://github.com/verivital/nnv/blob/master/nnv/install.m ), we need to make it clear what other dependenci…
-
I have tried convert_vgg.py and make a test,found that the result of keras model is different with the result of caffemodel, can you tell me why?
-
How to solve this problem?
-
请问您在使用t-sne进行数据可视化的时候，输入的是经过全连接层得到的特征吗，我直接数据原始数据发现结果异常。
-
Hi,
I followed the process of setup, downloaded checkpoints for vgg16.pth, latest.pth for global attr.
Anytime I run the demo I get the same output no matter the image.
[ Top3 Prediction ]
p…