Thanks for your great job. I have a question about how you split evaluation datasets (i.e. MIMIC-CXR and IU-Xray). In table 1 of your paper, you mentioned the sizes of MIMIC-CXR and IU-Xray are 1732 and 3301. The original sizes of these two datasets are different from yours. I think your are using subsets of these two datasets, right? Can you share how you choose subsets from the original datasets?
Hi authors,
Thanks for your great job. I have a question about how you split evaluation datasets (i.e. MIMIC-CXR and IU-Xray). In table 1 of your paper, you mentioned the sizes of MIMIC-CXR and IU-Xray are 1732 and 3301. The original sizes of these two datasets are different from yours. I think your are using subsets of these two datasets, right? Can you share how you choose subsets from the original datasets?
Thank you