Hello,
Thank you for the great work. Although the paper gives great detail in how to create the data as well as the author's response to this issue, I wanted to see if the author will provide at least some pseudo/high-level code for constructing the ECG-Reasoning benchmark.
As one cannot expect model's to do well without data, I would imagine people would want to augment other datasets to have a similar structure as the ECG-Reasoning benchmark, to be able to perform well on the benchmark.
Please let me know and thank you again.
Hello,
Thank you for the great work. Although the paper gives great detail in how to create the data as well as the author's response to this issue, I wanted to see if the author will provide at least some pseudo/high-level code for constructing the ECG-Reasoning benchmark.
As one cannot expect model's to do well without data, I would imagine people would want to augment other datasets to have a similar structure as the ECG-Reasoning benchmark, to be able to perform well on the benchmark.
Please let me know and thank you again.