Hi, thank you for sharing your code. I am curious about how the final result in Table 2 was calculated. In section 3.1 of the paper, it is mentioned that we report the average results over five different initiations in Table 2. Do you run experiments with five seeds or average results for five different $[\lambda_1, \lambda_2]$? Which epoch do you report?
Hi, thank you for sharing your code. I am curious about how the final result in Table 2 was calculated. In section 3.1 of the paper, it is mentioned that we report the average results over five different initiations in Table 2. Do you run experiments with five seeds or average results for five different $[\lambda_1, \lambda_2]$? Which epoch do you report?