dface icon indicating copy to clipboard operation
dface copied to clipboard

TypeError: 'float' object is not subscriptable

Open Dawnlnz opened this issue 5 years ago • 4 comments

File "E:\DeepLearning\demo\DFace\dface\train_net\train.py", line 87, in train_pnet show1 = accuracy.data.tolist()[0] TypeError: 'float' object is not subscriptable

Dawnlnz avatar Mar 01 '19 10:03 Dawnlnz

try show1=str(accuracy.data.tolist())

RickyZ92 avatar Mar 11 '19 10:03 RickyZ92

Thank you very much.But I have one more question about training Onet. Do you have the traingImageList.txt of CeleBA or do you know how to get it? The label file I downloaded from the Internet does not meet the requirements,it just has 11 elements each line.But we need 15 elements each line. Thanks again!

| | 18281575602 | | [email protected] | 签名由网易邮箱大师定制 On 3/11/2019 18:03,RickyZ92[email protected] wrote:

try show1=str(accuracy.data.tolist())

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

Dawnlnz avatar Mar 12 '19 02:03 Dawnlnz

你的11行的是只有特征的文件 还有一个带4个参数的文件 把两个文件合合并在一起就行了。 我觉得作者在landmark部分实际用的是这个 (因为我之前训练过这个模型,从用法上来看比较像 但我自己没有试 不过我也比较蒙 因为这个数据集是lfw的。。并不是celeba的。。。) 如果你需要合成后的celeba文件 我可以直接给你

RickyZ92 avatar Mar 12 '19 05:03 RickyZ92

那麻烦你把合成的celeba的也给我一份吧,我想LFW和celeba两个都试试。

| | 18281575602 | | [email protected] | 签名由网易邮箱大师定制 On 3/12/2019 13:51,RickyZ92[email protected] wrote:

你的11行的是只有特征的文件 还有一个带4个参数的文件 把两个文件合合并在一起就行了。 我觉得作者在landmark部分实际用的是这个 (因为我之前训练过这个模型,从用法上来看比较像 但我自己没有试 不过我也比较蒙 因为这个数据集是lfw的。。并不是celeba的。。。) 如果你需要合成后的celeba文件 我可以直接给你

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

Dawnlnz avatar Mar 12 '19 05:03 Dawnlnz