We hope to achieve an chest Xray detecting system

The top-1/5 accuracy rates by using single center crop (crop size: 224x224, image size: 256xN)

First, download the above pretrained weights to the imagenet_models folder.

Run test_inference.py for an example of how to use the pretrained model to make inference.

python test_inference.py

for image viewer, write code:

d1 = dict[b'data'][1] print(d1) NewImage = Image.new('RGB', (32, 32)) print(d1[0]) pixelList = []

    for i in range(0, 1024):
        pixel = (d1[i], d1[i + 1024], d1[i + 2048])
        pixelList.append(pixel)
    print(pixelList)
    NewImage.putdata(pixelList)
    NewImage.save('test5.jpg')

Feel free to modify the code to your need