AutoAiLib.general_tester(model path or object, labels, preprocessor)

Class Video Demo

AutoAiLib.general_tester.predict_single(example)

• example- If you have defined a preprocessor for your tester, this should comply with the preprocessor's argument. If you have not defined a preprocessor, example must be in a form that your model will accept.

AutoAiLib.general_tester.predict_many(container=None, testing_folder = None, csv_dir)

• container - This can be a container of test objects (any iterable). If preprocessor is defined, these objects must comply with the preprocessors parameter. Otherwise they must be in a form that your model will accept.
• testing_dir - This can be a path to a testing folder which has sub folders of all classes. Again, must be preprocessed or have preprocessor defined.
• csv_dir - This function compiles data into a csv folder to allow users to easily extract data from it, if not defined it will return a pandas data frame.

AutoAi.convnet_tester(model path or object, labels)

Class Video Demo

AutoAi.image_predict(model_path, image_path, labels)

This function takes 3 arguments: a path to a keras model, a path to an image, and a list of labels.

Demo:

Given a the correct arguments, we get the following output, as well as this image saved to our model directory.

AutoAi.manual_test(model, testing_dir, labels)

This function tests a model given labels and testing data. It then compiles the results in a CSV file, and groups the results by class, and by correct and incorrect.

• Model - Path of model that you want to test or model object.
• Testing_dir - Path to the directory with your testing data.
• Labels - Dictionary of the classes, in form (index:class_name)

Example csv:

Update! This has now been packaged in the AutoAI.data_compiler class. AutoAi.data_compiler(self,src, dest, **kwargs)
AutoAi.data_compiler.run() will compile the data based on the constructor parameters.

This function takes 2 required arguments, an original data source file, and a path to the desired data directory. Given just these two arguments, this function will create a new testing data folder at dest with training, validation, and testing folders, containing folders for each class. You can alter the ratio with the ratio arguments, as well as provide a number of img transforms to do if you are using images.

• Src - Path to a folder that contains a folder for each class and then data examples in those class folders.
• Dest - Path to a folder where you want the data to end up.
• Num_imgs_per_class - This number of images will be added to the original set for each class through transforms. The theoretical limit for this would be 3! * original images per class

Demo:

Given the a path to the following folder: If augmentation used the following results will be yielded: Then these images will be copied to the dest folder with copied file structure, but an added upper layer: Example showing the images made it:

Homeless Methods:

model_to_img(model_path)

• Returns an image form of your model.

plot(history=None, file=None, min_=0, max_=1)

• history- numpy file (Keras callback)
• file - path to a .npy file.
• min_ - the minimum of accuracy/loss in the graph
• max_ - the maximum of accuracy/loss in the graph, the closer the min and max, the more zoomed your graph will be