Hi ,
I dont know if this is not related to the forum. I am not an expert in Pytorch or DL. I just wanted to do a prediction whether an embryo is healthy. I need a time lapse image dataset of embryos. Where can I find the dataset? How do I arrive at a correct network model which can be used? Right now I have an idea of using Resnet (how do I arrive at a correct network model which suits my problem statement in general)! Suggestions are welcome
Thank you!
I don’t know, if there is an open dataset for this particular task, but would recommend to read the latest papers in this area and see, if the authors have mentioned particular datasets they have been using. If not, you might want to contact some of them.
This also applies for the second question: if possible, I would try to find a good baseline and start with this model in order to try to reproduce the results. If there is no baseline, you could try to use “standard” model such as the mentioned ResNet. Generally, I would also recommend to check out some courses, which explain how to build a custom model, such as FastAI.
Thank you so much sir, Currently found this https://www.kaggle.com/ditzsins/grades-embryo , I think it would be out of scope for me to ask if the above dataset is a time lapse image (Apologize if it is!). Yes sir FastAI is a great idea as it makes training process and other stuff more easier . Thank you so much again and sure to update you sir!