Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / bubbliiiing/facenet-pytorch issues and pull requests
#40 - 请问图片数据集在哪下载
Issue -
State: open - Opened by 141win 7 days ago
#39 - 为什么验证集val_accuracy的准确率只有百分之几,但是lfw的准确率却能到80多,训练集准确率也能到99多
Issue -
State: open - Opened by D6582 about 2 months ago
- 5 comments
#38 - model_data/lfw_pair.txt为什么不完整?
Issue -
State: open - Opened by TheTurtle2020 7 months ago
#37 - 训练epoch数
Issue -
State: open - Opened by AndrewDing616 7 months ago
#36 - ValueError:
Issue -
State: open - Opened by Changshengli0 8 months ago
- 1 comment
#35 - 为什么训练时,在LFW数据集的得分一直是0.500000+-0.0000
Issue -
State: open - Opened by Zhangwenyao1 about 1 year ago
#34 - predict能否改为realtime的即时识别?
Issue -
State: open - Opened by YenYunn about 1 year ago
#33 - 为什么训练时,在LFW数据集的得分一直是0.500000+-0.0000
Issue -
State: open - Opened by Tyler-Chenn about 1 year ago
- 1 comment
#32 - 关于使用百度飞桨怎么生成这个预训练权重的参数问题
Issue -
State: open - Opened by heshengjun811 about 1 year ago
#31 - 帅博, 对图片的预处理的代码可以发我吗? [email protected] 感谢~
Issue -
State: open - Opened by xingyuezhiyun over 1 year ago
#30 - b导,你这个用mobilenet怎么训练的呢
Issue -
State: open - Opened by aizsa2022 over 1 year ago
- 2 comments
#29 - 请问从哪可以获得输入图片分辨率大一点的训练权重
Issue -
State: open - Opened by tqinger over 1 year ago
#28 - eval_LFW验证精度问题
Issue -
State: closed - Opened by BaseJiang over 1 year ago
- 3 comments
#27 - 训练特征提取层的问题
Issue -
State: open - Opened by GHT0320 over 1 year ago
- 3 comments
#26 - utils中的dataloader.py第十一行调用时报错
Issue -
State: open - Opened by corazon2333 over 1 year ago
- 1 comment
#25 - 想问问如何输出loss曲线还有一些中间训练的参数(梯度啥的)
Issue -
State: open - Opened by mksasx over 1 year ago
- 1 comment
#24 - 训练时报错
Issue -
State: open - Opened by mksasx over 1 year ago
- 1 comment
#23 - loss绘制和打开
Issue -
State: open - Opened by samba1518 over 1 year ago
- 2 comments
#22 - 训练和测试时会调过一部分数据
Issue -
State: open - Opened by samba1518 over 1 year ago
#21 - ValueError: Expect x to not have duplicates
Issue -
State: open - Opened by gwegwger over 1 year ago
- 2 comments
#20 - colab 的代码
Issue -
State: open - Opened by xlflove over 1 year ago
- 1 comment
#19 - 验证时报错:ValueError: Cannot have number of splits n_splits=10 greater than the number of samples: n_samples=0.
Issue -
State: open - Opened by huangchuanfeng over 1 year ago
- 3 comments
#18 - 如果做迁移学习 图片应该怎么对齐呢,能问问初始训练模型时应用哪种图片对齐方式呢
Issue -
State: open - Opened by bopker over 1 year ago
- 6 comments
#17 - facenet_mobilenet.pth怎麼來的
Issue -
State: open - Opened by aa389393 over 1 year ago
- 2 comments
#17 - facenet_mobilenet.pth怎麼來的
Issue -
State: open - Opened by aa389393 over 1 year ago
- 2 comments
#16 - 关于其他数据集验证测试
Issue -
State: closed - Opened by z915287285 almost 2 years ago
- 2 comments
#15 - 能否只通过推理一张图片得到其分类。
Issue -
State: open - Opened by FZUChenhaoLin almost 2 years ago
- 3 comments
#14 - 请教下关于dataloader的代码有一点不明白
Issue -
State: closed - Opened by bigcat26 about 2 years ago
- 1 comment
#13 - 转ncnn部署的时候,推理结果一直都是1或者-1,请问有用这个模型部署过ncnn的吗
Issue -
State: open - Opened by bo-bobo about 2 years ago
- 1 comment
#12 - 使用Inception resenet v1从0开始训练的时候精度持续下降
Issue -
State: closed - Opened by Rayn-Wu about 2 years ago
#11 - 训练过程经常遇到BrokenPipeError: [Errno 32] Broken pipe
Issue -
State: closed - Opened by jia0511 about 2 years ago
- 1 comment
#10 - 使用NNI对模型剪枝遇到了算子问题
Issue -
State: open - Opened by jia0511 over 2 years ago
- 1 comment
#9 - 请问是用128个类的图片训练的吗?可以告知下是哪128类吗?
Issue -
State: open - Opened by Kitzzaaa over 2 years ago
- 1 comment
#8 - 请问这个是基于什么论文的?
Issue -
State: open - Opened by Kitzzaaa over 2 years ago
- 1 comment
#7 - 冻结训练
Issue -
State: open - Opened by hust-wuhan over 2 years ago
- 2 comments
#6 - 数据集在哪下载
Issue -
State: open - Opened by kundezui over 2 years ago
- 2 comments
#5 - CE_loss in validation data
Issue -
State: open - Opened by jack-gits over 2 years ago
- 17 comments
#4 - confuse about _triplet_loss
Issue -
State: closed - Opened by jack-gits over 2 years ago
- 1 comment
#2 - lfw数据集处理有什么区别? 精度目前97%
Issue -
State: open - Opened by xiaomujiang almost 3 years ago
- 4 comments
#1 - 谢谢楼主
Issue -
State: closed - Opened by huangluyao over 3 years ago