Question Regarding Exp.10 in Table 2 - Training Method for Planning-Only Experiment

[ChristinaTan0704]

Hi authors,

Thanks for the great work. It is truly impressive! Regarding Exp.10 in Table 2, I have a few questions about the 'planning-only' experiment. Could you please provide some further details on how exactly you trained the model for this experiment?

I've formulated a couple of assumptions based on my understanding:

(1) You might have trained the model using all modules, but only employed the planning loss to update the entire network. (2) Or, based on the slides from your talk, you might have removed or randomized Q and achieve results without occ post-processing.

Did you use approach (1) or (2), or was it a different method? Could you please elaborate a bit more on this experiment? Your assistance is greatly appreciated.

Thank you so much!

Best Regards, Christina

[yudefu]

Hi authors,

Thanks for the great work. It is truly impressive! Regarding Exp.10 in Table 2, I have a few questions about the 'planning-only' experiment. Could you please provide some further details on how exactly you trained the model for this experiment?

I've formulated a couple of assumptions based on my understanding:

(1) You might have trained the model using all modules, but only employed the planning loss to update the entire network. (2) Or, based on the slides from your talk, you might have removed or randomized Q and achieve results without occ post-processing.

Did you use approach (1) or (2), or was it a different method? Could you please elaborate a bit more on this experiment? Your assistance is greatly appreciated.

Thank you so much!

Best Regards, Christina

hello，Hello, may I add your QQ to facilitate the discussion of this paper and code（qq：3055355954）

[yudefu]

------------------ 原始邮件 ------------------ 发件人: "OpenDriveLab/UniAD" @.>; 发送时间: 2023年7月31日(星期一) 下午2:47 @.>; @.@.>; 主题: Re: [OpenDriveLab/UniAD] Question Regarding Exp.10 in Table 2 - Training Method for Planning-Only Experiment (Issue #93)

你好，可以加我微信吗？y1193547749

Could we possibly use WeChat or another app? I lost access to my QQ account years ago. Could you please email your contact information to @.***? Thanks your so much for your help!

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

[ChristinaTan0704]

Sorry for mistakenly closed the issue, still unsolved, thanks # Question Regarding Exp.10 in Table 2 - Training Method for Planning-Only Experiment

[Yihanhu]

Hi @ChristinaTan0704 ,

Yes, you are right, we trained the all modules but use a learned query for planning and without postprocessing.