You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First of all, thank you very much for open-sourcing your work!
I want to reproduce the ablation study for Figure 3(a). I didn't change the network structure,I used the parameters you trained for SCRIMP,then,I only modified the input to the communication part, specifically the value of c1 in the forward function of the SCRIMPNet class.。 Below is the code after my modifications.
After making these changes and running eval_model.py, the results for SCRIMP_WITHOUT_COM and SCRIMP_WITHOUT_ENCODER are poor. Does this mean there's an issue with my approach? Do I need to change the network architecture and retrain it, and then perform the ablation study for evaluation?or Could you tell me how to conduct the experiments for SCRIMP_WITHOUT_COM and SCRIMP_WITHOUT_ENCODER specifically?
I am looking forward to your response.Thank you!
The text was updated successfully, but these errors were encountered:
First of all, thank you very much for open-sourcing your work!
I want to reproduce the ablation study for Figure 3(a). I didn't change the network structure,I used the parameters you trained for SCRIMP,then,I only modified the input to the communication part, specifically the value of c1 in the forward function of the SCRIMPNet class.。 Below is the code after my modifications.
After making these changes and running eval_model.py, the results for SCRIMP_WITHOUT_COM and SCRIMP_WITHOUT_ENCODER are poor. Does this mean there's an issue with my approach? Do I need to change the network architecture and retrain it, and then perform the ablation study for evaluation?or Could you tell me how to conduct the experiments for SCRIMP_WITHOUT_COM and SCRIMP_WITHOUT_ENCODER specifically?
I am looking forward to your response.Thank you!
The text was updated successfully, but these errors were encountered: