hikvision-research / opera

A Unified Toolbox for Object Perception & Application
Apache License 2.0
149 stars 13 forks source link

Ask for checkpoints #5

Closed lutianhao closed 2 years ago

lutianhao commented 2 years ago

Hi, I'm going to learn you paper and code, could you please share infos like checkpoints, etc for training and testing? I can not find any address to download checkpoints. Thanks!

shidahu commented 2 years ago

We have provided the checkpoints for each algorithm in Opera. The details can be found in configs/xxx/README.md. Thanks.

lutianhao commented 2 years ago
font{
    line-height: 1.6;
}
ul,ol{
    padding-left: 20px;
    list-style-position: inside;
}

    Thanks for your kind response, I've found that. Another question, what is the minimum amount of GPU space required for each model to be used?  

                ***@***.***

---- Replied Message ----

     From 

        Dahu ***@***.***>

     Date 

    6/29/2022 19:54

     To 

        ***@***.***>

     Cc 

        ***@***.***>
        ,

        ***@***.***>

     Subject 

          Re: [hikvision-research/opera] Ask for checkpoints (Issue #5)

We have provided the checkpoints for each algorithm in Opera. The details can be found in configs/xxx/README.md. Thanks.

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: @.***>

shidahu commented 2 years ago

Models were trained on NVIDIA Tesla V100 (32G) GPUs by default and a few models (e.g., Swin-L backbone) were trained on NVIDIA Tesla A100 (40G) GPUs. Thanks.