zzxslp / wcl Goto Github PK

the problem is as following：
File "/Users/shihaozhe/WCL-main/modules/datasets.py", line 57, in getitem
label = example['label']
KeyError: 'label'

when I check the annotation.json, I don't find the label too , so how do I solve this problem, what does the label mean? any suggestions?
hoping for your reply, thx.

Hi Authors:

Thank you for the great work. However, I have not found the ChestBert and K-Means implementation from your code. Most of the part are similar from the R2GEN expect the contrastive loss. Could you please point me where are the ChestBert and K-Means implementation,

Thank you.

Hello, I follow your training setting in run_mimic_cxr.sh, but I got the BLEU_1 score is 36. so what I shoud do to achieve you score in your paper. (I run your code on 1 V100)

I checked at the site for MIMIC datasets mentioned. The site mentions total uncompressed size: 557.6 GB.
Can you let me know the resized image size. Is it 512x512.

Can you share the script for preprocessing the data from downloaded data set.

请问作者可以提供一下 带有label信息的cluster_annotation.json 吗?

Maybe I don't have access permissions. Since my PhysioNet account is **@163.com, not Gmail. How should I obtain the MIMIC-ABN dataset in this situation? Thank you!

I checked at the site for MIMIC datasets mentioned.
The site mentions it as size below.
Total uncompressed size: 557.6 GB.

Can you please let me know the size after resizing to 512 x 512.
Is it possible to share the resized data set.

WCL/modules/datasets.py 57行
label = example['label']
我看到数据源annotation.json也没有label这一属性，请问如何解决？