mcspooder/camull-net

A 3D convolutional neural network for the classification of Alzheimer's disease and its precursor Mild Cognitive Impairement.

camull-net's Issues

The input data size in the code is different from that in the paper

Dear author, your data is different from the data in the original paper. Have you done anything else?

evaluation throws a pytorch cpu error

Your cross validation does not initialize each fold model, but uses the previously trained model as the fold initialization model. Is this wrong?

data management

Hi,
Nice work and I highly appreciate it. Can you please guide about how to place data in directory if someone doesn't have access to ADNI and wants to use their own dataset, and also is it possible to just use MRIs, no clinical data.
I am waiting for your kind reply.
Thanks

ValueError: num_samples should be a positive integer value, but got num_samples=0

thank you for sharing such a detailed code!!! Can you please tell me how to solve this error ?
"ValueError: num_samples should be a positive integer value, but got num_samples=0"

Missing folders during evaluation

When evaluating the model, errors are thrown if there are is no corresponding log file in the log folder or the correspondings graphs folder is missing.

In otherwords if someone uses the code to evaluate a model that they haven't trained they will get an error because the training code wouldn't have made the needed folders. However it should be possible to evaluate an external model downloaded from elsewhere.

No way to dynamically set batch size when running code on differnt machines.

Batch size essentially dicates how many data samples can be loaded in parallel. If this number is higher than what the users GPU/s support then a cuda error is thrown.

Dear author,when I Run "python camull_cli.py",I can't find this fold

Dear author,when I Run "python camull_cli.py",I can't find this fold

conn = sqlite3.connect("..\weights\neural-network.db")
sqlite3.OperationalError: unable to open database file

can you help me?

Originally posted by @yujianzhong3257298 in #8 (comment)