blusson/PT21-22-Reseau-Neurones
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

privatisation

Browse Source
This commit is contained in:
eynard 2021-12-09 18:50:57 +01:00
parent b844d10347
commit 75bc43d48f
2 changed files with 17 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
sobek/network.py
View File

@ -6,33 +6,36 @@ class network:
if type(inputLayerSize) != int: if type(inputLayerSize) != int:
raise TypeError("The input layer size must be an int!") raise TypeError("The input layer size must be an int!")
self.weights = [] self.__weights = []
self.inputLayerSize = inputLayerSize self.__inputLayerSize = inputLayerSize
self.oldLayerSize = inputLayerSize oldLayerSize = inputLayerSize
for layerSize in layerSizes: for layerSize in layerSizes:
self.weights.append( np.random.default_rng(42).random((self.oldLayerSize, layerSize)) ) self.__weights.append( np.random.default_rng(42).random((oldLayerSize, layerSize)) )
self.oldLayerSize = layerSize oldLayerSize = layerSize
self.biases = [[0]*layerSize for layerSize in layerSizes] self.__biases = [[0]*layerSize for layerSize in layerSizes]
self.weights = np.array(self.weights, dtype=object) self.__weights = np.array(self.__weights, dtype=object)
self.biases = np.array(self.biases, dtype=object) self.__biases = np.array(self.__biases, dtype=object)
def reLu(value): def __reLu(value):
return max(0, value) return max(0, value)
def process(self, input): def process(self, input):
if type(input) != np.ndarray: if type(input) != np.ndarray:
raise TypeError("The input must be a vector!") raise TypeError("The input must be a vector!")
if input.size != self.inputLayerSize: if input.size != self.__inputLayerSize:
raise ValueError("The input vector has the wrong size!") raise ValueError("The input vector has the wrong size!")
if input.dtype != np.float64: if input.dtype != np.float64:
raise TypeError("The input vector must contain floats!") raise TypeError("The input vector must contain floats!")
for layerWeights, bias in zip(self.weights, self.biases): for layerWeights, bias in zip(self.__weights, self.__biases):
input = np.matmul(input, layerWeights) input = np.matmul(input, layerWeights)
input = np.add(input, bias) input = np.add(input, bias)
#reLu application #reLu application
with np.nditer(input, op_flags=['readwrite']) as layer: with np.nditer(input, op_flags=['readwrite']) as layer:
for neuron in layer: for neuron in layer:
neuron = network.reLu(neuron) neuron = network.__reLu(neuron)
return input return input
def train(self, inputs, results):

2
test.py
View File

@ -9,6 +9,6 @@ for y in test.weights:
for y in test.biases: for y in test.biases:
print(y, end="\n\n")""" print(y, end="\n\n")"""
print(network.reLu(8)) #print(network.__reLu(8))
print(test.process(np.random.default_rng(42).random((16)))) print(test.process(np.random.default_rng(42).random((16))))