FINISH

algorithms/neural_network/neural_network_interface.py

@@ -14,7 +14,9 @@ CNN = NeuralNetwork().to(device)
 def train(model):
     model.train()
     trainset = WaterSandTreeGrass('./data/train_csv_file.csv', transform=setup_photos)
+    testset = WaterSandTreeGrass('./data/test_csv_file.csv', transform=setup_photos)
     train_loader = DataLoader(trainset, batch_size=batch_size, shuffle=True)
+    test_loader = DataLoader(testset, batch_size=batch_size, shuffle=True)
 
     criterion = nn.CrossEntropyLoss()
     optimizer = Adam(model.parameters(), lr=learning_rate)
@@ -33,11 +35,13 @@ def train(model):
             optimizer.step()
 
             if epoch % 2 == 0:
-                print("epoch: %3d loss: %.4f" % (epoch, loss.item()))
+                print("epoch: %d loss: %.4f" % (epoch, loss.item()))
 
-    print("FINISHED!")
-    print("Checking accuracy.")
+    print("FINISHED TRAINING!")
+    print("Checking accuracy for the train set.")
     check_accuracy(train_loader)
+    print("Checking accuracy for the test set.")
+    check_accuracy(test_loader)
 
     torch.save(model.state_dict(), "./learnedNetwork.pt")
 
@@ -62,7 +66,7 @@ def check_accuracy(loader):
             num_correct += (predictions == y).sum()
             num_samples += predictions.size(0)
 
-        print(f"Got {num_correct}/{num_samples} with accuracy {float(num_correct)/float(num_samples)*100:.2f}")
+        print(f"Got {num_correct}/{num_samples} with accuracy {float(num_correct)/float(num_samples)*100:.2f}%")
 
 
 def what_is_it(img_path, show_img=False):
@@ -83,4 +87,4 @@ def what_is_it(img_path, show_img=False):
         return id_to_class[idx]
 
 
-print(what_is_it('./data/test/sand/sand.png', True))
+train(CNN)

common/constants.py

@@ -74,7 +74,7 @@ BAR_HEIGHT_MULTIPLIER = 0.1
 #NEURAL_NETWORK
 learning_rate = 0.001
 batch_size = 7
-num_epochs = 10
+num_epochs = 100
 
 device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
 classes = ['grass', 'sand', 'tree', 'water']