from cortex_class import cortex
# import test data
path = r"training_sets\AND.txt"
# set up a cortex
cort = cortex('cortex_1')
cort.import_training(path)
cort.populate(0) # 0 hidden layer neurons
cort.min_acc = 1.0
cort.t_function("Threshold", cort.output_neurons)
#train the cortex
cort.train(100, 50, 10000)
cort.print_training_accuracy_report()
cort.interogate()
from cortex_class import cortex
# set up a cortex
input_path = r"training_sets\XNOR_XOR_AND_OR_NOR_NAND.txt"
cort = cortex('cortex_1')
num_hidden = 4
min_acc = 1.0
cort.import_training(input_path)
cort.normalize_training(0, 1)
cort.populate(num_hidden)
cort.min_acc = min_acc
cort.t_function("Threshold", cort.output_neurons)
# make sure even after export and re-import, the cortex can still train
cort.export_state('initial_state.txt')
cort.import_state('initial_state.txt')
#train the cortex
incriment = 100
stabilize = 500
maximum = 10000
cort.train(incriment, stabilize, maximum)
cort.print_training_accuracy_report()
cort.export_state('final_state.txt')
from cortex_class import cortex
cortex_path = "cortexes\All_logic_gates_cortex9.txt"
cort = cortex("cort1")
cort.import_state(cortex_path)
cort.interogate()