Use Ganomaly for image anomaly detection.

[Instructions]

1. 1_visdom.ipynb

Start the visdom server.

2. 2_train.ipynb

Perform training.

Set the parameters parameter:

• isize: input image size

• niter: number of training iterations

• batchsize: the size of the batch

• nz: the dimension of the z vector

• lr: learning rate

After the setting is completed, it can be executed.

3. 3_test_to_find_threshold.ipynb

Test the model and find anomaly thresholds.

Set the parameters:

• isize: input image size

• batchsize: the size of the batch

After the setting is completed, it can be executed.

After execution, the anomaly threshold value can be obtained.

4. 4_inference.ipynb

Perform inference.

Set the parameters:

• isize: input image size

• batchsize: the size of the batch

• image_file: inferenced image

• an_threshold: anomaly threshold

After the setting is completed, it can be executed.

After execution, you can know whether the inferenced image is anomaly.

Jupyter-Image-Ganomaly-inference.png

This SDK is built in AppForAI - AI Dev Tools.

Purchase license separately: USD 600, permanent authorization, single APP authorization, single machine authorization, one-year activation, one-year download, one-year update, one-year email technical support.