gregwchase/nih-chest-xray

Identifying diseases in chest X-rays using convolutional neural networks

Technology Stack

This my current technology stack, which is what I use in my professional and personal development.

	Package	Use Cases	Notes
	AWS	Cloud computing services	EC2 S3 ECS
	Dask	Distributed computing	Lighter than Spark Follows Pandas API GPU compatible
	FastAPI	REST API's	Asynchronous Faster than Flask Easier to test
	Poetry	Project package management and publishing	Resolves package dependencies
	RAPIDS	GPU data science libraries	cuDF cuML Dask cuDF

Contact Information

Website	Username
LinkedIn	gregwchase
Medium	gregorywchase
Twitter	@gregwchase

nih-chest-xray's Issues

Any progress on this dataset?

I have seen some Kaggle competition dataset.
Are they different dataset?

Cnn_keras converting to float 32

Hi! First of all, thank you very much for this great contribution and sharing of your deep learning model with us. I would like to ask the question with regards to converting X_train and X_test to 'float32' in cnn_keras.py. The point is, I get the 'Memory error' when running your code with images scaled to 512x512, how do you go about solving this one?

Thank you very much once again!

Cannot reshape (85696,512,512,1)

nih-chest-xray/src/cnn_keras.py

Line 192 in 045b439

X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, channels)

Hi thanks for the great contribution but I was getting this error - cannot reshape 85696 into shape (85696,512,512,1)
Number of images that got downloaded from dataset were 107,121