BackpropTrainer Model

Untitled.ipynb

@@ -31,11 +31,33 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 6,
    "id": "04bbec40",
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "outputs": [
+    {
+     "ename": "ModuleNotFoundError",
+     "evalue": "No module named 'pybrain3'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
+      "Input \u001b[0;32mIn [6]\u001b[0m, in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mpybrain3\u001b[39;00m\n\u001b[1;32m      2\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mpickle\u001b[39;00m\n\u001b[1;32m      3\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mmatplotlib\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpylab\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mplt\u001b[39;00m \n",
+      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'pybrain3'"
+     ]
+    }
+   ],
+   "source": [
+    "import pybrain3\n",
+    "import pickle\n",
+    "import matplotlib.pylab as plt \n",
+    "from numpy import ravel\n",
+    "from pybrain3.tools.shortcuts import buildNetwork\n",
+    "from pybrain3.datasets import SupervisedDataSet\n",
+    "from pybrain3.supervised.trainers import BackpropTrainer \n",
+    "from pybrain3.tools.xml.networkwriter import NetworkWriter \n",
+    "from pybrain3.tools.xml.networkreader import NetworkReader"
+   ]
   },
   {
    "cell_type": "code",

neurone/main.py

@@ -0,0 +1,72 @@
+import pybrain3
+import pickle
+import matplotlib.pylab as plt
+from numpy import ravel
+from pybrain3.tools.shortcuts import buildNetwork
+from pybrain3.datasets import SupervisedDataSet
+from pybrain3.supervised.trainers import BackpropTrainer 
+from pybrain3.tools.xml.networkwriter import NetworkWriter
+from pybrain3.tools.xml.networkreader import NetworkReader
+
+# https://www.machinelearningmastery.ru/how-to-configure-the-number-of-layers-and-nodes-in-a-neural-network/
+class BackPropagationModel():
+    def __init__(self, metrics:int = 4, 
+                 predictions:int = 1, 
+                 input_layer:int = 4, 
+                 hidden_layer:int = 3,
+                 output_layer:int = 1):
+        # 4 метрики, 1 предикшн
+        self.metrics = metrics
+        self.predictions = predictions
+        self.input_layer = input_layer
+        self.hidden_layer = hidden_layer
+        self.output_layer = output_layer
+        self.ds = SupervisedDataSet(metrics, predictions)
+        
+    
+    def activateModel(self):
+        self.net = buildNetwork(self.input_layer, self.hidden_layer, self.output_layer, bias=True)
+        self.trainer = BackpropTrainer(self.net, dataset=self.ds, momentum=0.1, learningrate=0.01, verbose=True, weightdecay=0.01)
+
+        self.trnerr, self.valerr = self.trainer.trainUntilConvergence()
+        
+        plt.plot(self.trnerr, 'b', self.valerr, 'r')
+        plt.show()
+
+    def addDataToModel(self, input:list, target:list):
+        self.ds.addSample(inp=input, target=target)
+    
+    def predict(self, data:list):
+        y = self.net.activate(data)
+        print(y)
+        return y
+    
+    def saveModel(self):
+        fileObject = open('model.txt', 'wb') 
+        pickle.dump(self.net, fileObject) 
+        fileObject.close()
+        
+        
+    @staticmethod
+    def getModel():
+        fileObject = open('model.txt', 'rb') 
+        net2 = pickle.load(fileObject) 
+        fileObject.close()
+        return net2
+        
+model = BackPropagationModel()
+
+model.addDataToModel([2, 3, 80, 1], [5])
+model.addDataToModel([5, 5, 50, 2], [4])
+model.addDataToModel([10, 7, 40, 3], [3])
+model.addDataToModel([15, 9, 20, 4], [2])
+model.addDataToModel([20, 11, 10, 5], [1])
+
+model.activateModel()
+
+model.saveModel()
+
+# USE MODEL - >
+# model = BackPropagationModel.getModel()
+# print(model.activate([2, 3, 80, 1]))
+