-
Hello, thank you for your model first of all. I tried training from scratch the lightened CNN on CASIA dataset but I got training loss was 0. Moreover, I tried finetuning the model that was given on a…
-
Hi, I have tested your pretrained model with MS_Celeb_1M dataset on LFW, the results are pretty good:
average precsion(ap): 0.999541130076
best_threshold :0.328682, accuracy :0.992833
auc 0.9994…
-
Hi, @davidsandberg
Thanks for your excellent work. I am training the "CASIA_maxpy_clean" dataset according to your instruction of "Classifier training of inception resnet v1" step by step. But I fin…
-
hi,
i feel so excited when i saw your code. you really did a great job.
but i get 2 questions when i came to the fifth part Start classifier training in https://github.com/davidsandberg/fac…
-
I notice that you have trained a center-face model on MS-Celeb-1M dataset and get 99.3% on LFW. However, I do not train such a good results. In fact, my model just can not converge. The loss is around…
-
Hi Alfred,
I was testing your pretrained model on CASIA-Webface itself. While evaluating the images under `CASIA-WebFace/0005282/` I get the following result:
016.jpg predicted class: 8315
009.jpg pr…
-
Hi,
I see below table with size's picked for CASIA-WebFace training and lfw testing data set
Dataset size ec_mc_y ec_y
Training set (CASIA) 144x144 48 48
Te…
-
I am tasked with transfering deepid using caffe to that using mxnet. All parameters come from a caffe prototxt. But how I train the network,the maximum Train-accuracy is only up to like 0.40,while ca…
-
### Context of the issue.
Train with 1-million MS celebrity dataset
### Expected behavior
i want the new DNN model which gives greater accuracy
### Actual behavior.
i trained my dataset w…
-
The dataset is available [here](https://www.microsoft.com/en-us/research/project/ms-celeb-1m-challenge-recognizing-one-million-celebrities-real-world/).
A python program to decode the face thumbnails …