Range of Test Split

#5
by GraceKWhite - opened

Amazing MultiDialog!
Thanks to the author for the great work:)

I want to focus on the TTS task and use MultiDialog as the benchmark. At present, I don't know how to select the range of the test set.

Do we need to combine Test Freq and Test Rare into one test set to infer a result? That is 9,811+8,331.

We note that Table 4 shows the results of 300 randomly selected samples from the test set. I'm confused about which samples we should test during the inference to obtain results.

Integrated Vision Language Lab org

Hi,

Thank you for your interest in our work!

For the test set, you need to combine the test_freq and test_rare splits as you said.

We arbitrarily selected 300 samples from the combined test set for Table 4.

Sign up or log in to comment