I wanted to play around with the pretrained model in the downstream task of report generation. Given that the pretrained weights of MedViLL are available, I thought that it should be straightforward to implement a Google Colab notebook to use a pretrained MedViLL in inference mode to generate reports from Chest X-ray images. However, I'm having a hard time figuring out how to do this. I've been reading the code inside https://github.com/SuperSupermoon/MedViLL/tree/master/downstream_task/report_generation_and_vqa but I don't see an easy to way to do it.
Basically I want to create a Google Colab notebook that does the following things:
Downloads a single Chest X-ray image
Downloads the pretrained weights of MedViLL
Has the code of MedViLL model
Creates an instance of MedViLL and loads the pretrained weights
Uses MedViLL in inference mode to generate a report from a single Chest X-ray image
Visualizes this image next to the generated report
Is something like that easy to implement? I would appreciate some guidance on how to achieve this.
Hi! First of all, congrats for the awesome work.
I wanted to play around with the pretrained model in the downstream task of report generation. Given that the pretrained weights of MedViLL are available, I thought that it should be straightforward to implement a Google Colab notebook to use a pretrained MedViLL in inference mode to generate reports from Chest X-ray images. However, I'm having a hard time figuring out how to do this. I've been reading the code inside https://github.com/SuperSupermoon/MedViLL/tree/master/downstream_task/report_generation_and_vqa but I don't see an easy to way to do it.
Basically I want to create a Google Colab notebook that does the following things:
Is something like that easy to implement? I would appreciate some guidance on how to achieve this.
Thanks in advance.