Sorry, I did not see the HItNet model implementent? Only StereoNet and AdaptiveNet are provided in code? #35
Comments
|
Yes, the current hitnet code performance of the pytorch version is 2.48 on kitti.. I am experimenting with the first author of the hitnet paper to check the potential problems of current pytorch version. We plan to release the hitnet code after all the work is done（Maximize the performance of the pytorch version code). This process may take few weeks. |
|
ok, i got it. Thank you very much for your patience for open source the Hinet. I am looking forward to your upcoming version. |
|
@Magicboomliu you can watch this code repository to get timely code updates. |
|
As my personal computing resources are relatively tight at the moment, people with adequate gpu computing resources are also welcome to contact me. |
|
I am not sure, Are 2 x GTX2080ti( 11G each) avaliable for Hinet training at a rather small batch size ? Since there is no 3D convolution, the memory cost should not be so large I guess. |
|
For an experiment, generally 2 to 4 GPUs are needed. (2 is ok) |
|
model zoo 的模型文件请补充进去 |
|
您好，邮件已收到。感谢来信 ，我会尽快回复您。 Hello, I have already received the letter. Thanks for your mailing, I will reply as soon as possible. こんにちは！確かご電子メールをもらいました。私はできるだけ早くあなたに返事を出します。
|
|
感谢！祝你工作顺利
…------------------ 原始邮件 ------------------
发件人: "meteorshowers/X-StereoLab" ***@***.***>;
发送时间: 2023年2月12日(星期天) 中午1:23
***@***.***>;
***@***.******@***.***>;
主题: Re: [meteorshowers/X-StereoLab] Sorry, I did not see the HItNet model implementent? Only StereoNet and AdaptiveNet are provided in code? (#35)
您好，邮件已收到。感谢来信 ，我会尽快回复您。 Hello, I have already received the letter. Thanks for your mailing, I will reply as soon as possible. こんにちは！確かご電子メールをもらいました。私はできるだけ早くあなたに返事を出します。
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you commented.Message ID: ***@***.***>
|
No description provided.
The text was updated successfully, but these errors were encountered: