auroua/InsightFace_TF

Issues

• from models import base_server NOT IMPORTING

  #113 opened 2 years ago by shahzain110
  0

• About the accuracy computation(关于计算准确率的问题)

  #100 opened 5 years ago by shiyanpei
  1

• Performance issues in your project (by P3)

  #111 opened 3 years ago by DLPerf
  1

• Performance issues in train_nets.py(P2)

  #112 opened 3 years ago by DLPerf
  0

• compared with facenet

  #98 opened 5 years ago by shenriver
  2

• 训练代码 train_nets.py 内存泄漏

  #86 opened 4 years ago by Dawning23
  10

• Why training accuracy is 0.000000?

  #76 opened 6 years ago by alanMachineLeraning
  3

• Thank you for your code，could you tell why I feed input with shape（n，112，112，3），but just get the embedding with shape（1，512）

  #110 opened 4 years ago by ACheun9
  1

• about ckpt test

  #68 opened 6 years ago by zjz5250
  1

• How to download pretrain ?

  #109 opened 4 years ago by docongminh
  0

• In what format does the labels should be

  #108 opened 4 years ago by adarsh0906
  0

• Embeddings model d

  #107 opened 4 years ago by sonfire186
  0

• tfrecord

  #82 opened 6 years ago by huyz1117
  0

• How can I directly test the LFW data set with an existing model to get accuracy？Which folder is the model in?

  #106 opened 5 years ago by fjh0725
  0

• pre trained models

  #105 opened 5 years ago by rlrahulkanojia
  2

• could I get the 512-D face embedding features ?

  #104 opened 5 years ago by ucasiggcas
  0

• stom

  #103 opened 5 years ago by YeahWCAN
  0

• resnet_v1_50/block1/unit_1/bottleneck_v1/shortcut_bn/BatchNorm/beta

  #102 opened 5 years ago by Python-37
  0

• What is the face alignment method?

  #72 opened 6 years ago by LiuQiusuo
  4

• The Num_output parameter ? We will use it for what ?

  #91 opened 5 years ago by NguyenThaiHoc1
  2

• the number of samples is not 5.8M

  #101 opened 5 years ago by jetsmith
  0

• TypeError: must be real number, not Tensor

  #74 opened 6 years ago by alanMachineLeraning
  2

• Loss become nan

  #67 opened 6 years ago
  4

• How to solve the problem of "summary" errors during training？

  #99 opened 5 years ago by sunruina2
  0

• Did you try a lighter base such as MobileFaceNet?

  #97 opened 5 years ago by siyiding1216
  0

• Share model please,,..

  #96 opened 5 years ago by kyeongeunn
  0

• any instruction how to start?

  #95 opened 5 years ago by iheo
  0

• OutOfRangeError (see above for traceback): End of sequence

  #66 opened 6 years ago by Vincent630
  1

• center parallel strategy

  #93 opened 5 years ago by TengliEd
  0

• how to use the model

  #92 opened 5 years ago by shenriver
  1

• isn't mm = m?

  #90 opened 5 years ago by xuzhengxiao
  0

• why not call train_op directly?

  #84 opened 6 years ago by zhaowwenzhong
  1

• Why custom BatchNormLayer?

  #89 opened 5 years ago by sziraqui
  0

• What's the extraction code for model A - model D with Baidu Cloud disk link?

  #88 opened 5 years ago by BruceDai003
  1

• Is SE unit work

  #87 opened 5 years ago by PotentialPie
  0

• test the acc of model c on the lfw?

  #85 opened 6 years ago by flyduck
  2

• There is something wrong about lfw data preprocessing

  #83 opened 6 years ago by darchonyzx
  0

• loss stoped at 3, hard to decrease

  #81 opened 6 years ago by fengsky401
  0

• Why training accuracy is small after total_step 46680?

  #78 opened 6 years ago by alanMachineLeraning
  1

• Dataset not globally shuffled

  #80 opened 6 years ago by forseeking
  0

• may I ask your training time,please?

  #77 opened 6 years ago by friedhelm739
  1

• three questions

  #79 opened 6 years ago by zh7117
  0

• You must feed a value for placeholder tensor 'tower_3/resnet_v1_100/Placeholder' with dtype float

  #75 opened 6 years ago by alanMachineLeraning
  0

• How to finetune with new dataset ?

  #70 opened 6 years ago by bruinxiong
  1

• model without BN

  #73 opened 6 years ago by HsuTzuJen
  0

• Which dataset did you use for training model D?

  #71 opened 6 years ago by souschefistry
  0

• about Batch Norm, are there some tricks?

  #69 opened 6 years ago by darchonyzx
  0

• ModuleNotFoundError: No module named 'tl_layers_modify'

  #65 opened 6 years ago by zhanglaplace
  3

• why the process was killed when the error 'Filling up shuffle buffer (this may take a while): 577 of 10000'

  #64 opened 6 years ago by usake
  1

• What should I 都

  #63 opened 6 years ago by usake
  0