clcarwin sphereface_pytorch issues

clcarwin / sphereface_pytorch

A PyTorch Implementation of SphereFace.

MIT License

714 stars 172 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Improved logic for iterating over 'pairs.txt'

#64 sayandipdutta opened 2 years ago
0
Unnecessarily loading file contents of `pairs.txt` into memory `lfw_eval.py`

#63 sayandipdutta opened 2 years ago
0
Same index counter in lfw_eval.py at line 90 and line 106

#62 bashir-sadeghi opened 2 years ago
2
about lfw file

#61 sh1n00 opened 3 years ago
0
Can't load the pretrained model

#60 lychenyoko opened 3 years ago
1
Is the task in lfw_eval.py exactly the face recognition?

#59 rrryan2016 opened 3 years ago
1
Why the classnum is 10574 instead of 10575

#58 njuaplusplus opened 3 years ago
2
About test result

#57 Piaoyyt opened 3 years ago
0
failed to save model with lambda used

#56 iTomxy opened 4 years ago
0
why not use torch.cos when calculating the value of phi ?

#55 niyunsheng closed 4 years ago
0
When the shape of input is [1,3,112,96], the shape of output is [2,10574]?

#54 yBeOne opened 4 years ago
4
IndexError: too many indices for array

#53 tjulitianyi1997 opened 4 years ago
1
RE: Training Time on Casia Webface dataset

#52 Phillip1982 opened 4 years ago
0
RE: Facial landmark text file

#51 Phillip1982 closed 4 years ago
1
alignment(src_img, src_pts) function ？

#50 Ontheway361 opened 4 years ago
2
About renorm()

#49 neverUseThisName opened 4 years ago
0
The approximation of phi

#48 neverUseThisName opened 4 years ago
0
B phf 01

#47 stevenphf closed 5 years ago
0
CUDA error: out of memory

#46 YogaLYJ opened 5 years ago
3
Can we run the pytorch code on YTF dataset

#45 nayaneesh opened 5 years ago
0
Loss isn't decreasing

#44 ljtruong opened 5 years ago
3
About training data

#43 zorrocai opened 5 years ago
0
Update lfw_eval.py

#42 amznero opened 5 years ago
0
The purpose of imagesize

#41 fishman2008 opened 5 years ago
0
What your version of pyTorch?

#40 312shan opened 6 years ago
1
'float' object cannot be interpreted as an integer

#39 leadcain84 closed 6 years ago
1
Is it normal to have train accuracy reaching 99% already at epoch 10 (out of 20)?

#38 yxchng closed 6 years ago
0
Detailed steps to achieve the reported performance on LFW

#37 zeakey closed 4 years ago
0
training set

#36 bkj opened 6 years ago
1
read data errors

#35 zhao181 opened 6 years ago
1
About CASIA Washed List

#34 timho102003 opened 6 years ago
1
Is this implementation different from the original one since the original does not use angle loss?

#33 yxchng closed 6 years ago
0
Test lfw score

#32 Gerkam closed 6 years ago
1
Accuracy problem

#31 wyq0227 closed 6 years ago
2
stuck during training

#30 fluctlux closed 6 years ago
2
Change the threshold values, and speed it up

#29 johnnysclai closed 6 years ago
0
Sphereface64 just with softmax

#28 zhangxinyu-xyz opened 6 years ago
0
Classifier or Embedding

#27 justusschock opened 6 years ago
2
What's means of AngleLoss ?

#26 foocker closed 6 years ago
1
unicode error

#25 sevenHsu opened 6 years ago
5
when CrossEntropyLoss loss=nan

#24 JonesonZheng closed 6 years ago
2
Myphi is not monotonically decreasing ?

#23 ywu36 closed 6 years ago
1
Is it the correct way to calculate cross_entropy

#22 xf1994 closed 6 years ago
0
The annealing optimization strategy for A-Softmax loss

#21 taey16 closed 6 years ago
2
About your implementation

#20 HuangZiliAndy closed 6 years ago
3
thread start failed

#19 frischzenger opened 6 years ago
0
Model not found

#18 radonys closed 6 years ago
0
matlab_cp2tform.py warning

#17 lucaskyle closed 6 years ago
0
How to init net parameters for sphere net when train from scratch

#16 wangzhenhua2015 closed 6 years ago
1
gamma

#15 zhly0 closed 6 years ago
2