Inquiry regarding the definition and configuration of new data parameters

[YijuGuo]

Hi,

I am currently working with the latest version of the project and noticed several new parameters in the bash configuration. To ensure a correct setup, could you please clarify the definitions and setting requirements for the following parameters?

• data.dynamic_filtering
• data.dynamic_filtering_strategy
• data.p_easy
• data.p_hard
• data.target_zero_variance
• data.sampling_batch_size
• data.real_train_batch_size

Specifically, it would be very helpful to know:

1. What is the specific purpose of each parameter?
2. Are there any constraints or recommended ranges for their values? (Is there a required ratio between sampling_batch_size and real_train_batch_size?)

Thank you for your help and for maintaining this repository!

Best regards,

Yiju