From dfd71eaf079061b7c0252e019832accfbdbd1c0e Mon Sep 17 00:00:00 2001
From: Robert Sunbury <sunrobbie00@gmail.com>
Date: Thu, 22 Sep 2022 13:31:09 -0400
Subject: [PATCH] Initial commit to add a simple network file

---
 SimpleNet.py | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 48 insertions(+)
 create mode 100644 SimpleNet.py

diff --git a/SimpleNet.py b/SimpleNet.py
new file mode 100644
index 0000000..d2cabc4
--- /dev/null
+++ b/SimpleNet.py
@@ -0,0 +1,48 @@
+# -*- coding: utf-8 -*-
+"""
+Created on Thu Sep 22 13:07:34 2022
+
+@author: sunro
+"""
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as f
+import matplotlib as plt
+
+class simpleNet(nn.Module):
+    
+    def __init__(self,inFeature=2,outFeature=2,hiddenLayer1=8,hiddenLayer2=9):
+        super().__init__()
+        self.fc1 = nn.Linear(inFeature,hiddenLayer1)
+        self.fc2 = nn.Linear(hiddenLayer1,hiddenLayer2)
+        self.out = nn.Linear(hiddenLayer2,outFeature)
+        self.criterion = nn.CrossEntropyLoss()
+        self.optimizer = torch.optim.Adam(self.parameters(), lr=0.01)
+    
+    def predict(self, x):
+        x = f.relu(self.fc1(x))
+        x = f.relu(self.fc2(x))
+        x = self.out(x)
+        return x
+    
+    def train(self, x, y, epochs=100):
+        losses = []
+        i=0
+        for i in range(epochs):
+            i=i+1
+            predict = self.predict(x)
+            loss = self.criterion(predict, y)
+            losses.append(loss)
+            if i%10 == 1:
+                print(f'Epoch number: {i:2} Corresponding loss: {loss.item():10.2f}')
+            
+            self.optimizer.zero_grad()
+            loss.backward()
+            self.optimizer.step()
+        self.drawGD(losses, epochs)
+        
+    def drawGD(self,losses,epoches):
+        plt.plot(range(epoches),losses)
+        plt.xlabel("Epoches")
+        plt.ylabel("Losses")
\ No newline at end of file