Changed method for accuracy calculation:

AI/NN_accuracy.py

@@ -1,63 +1,47 @@
-import PIL
+import torch
+import torchvision
 import torchvision.transforms as transforms
-
-from AI import neural_network
+import torch.nn as nn
+import torch.nn.functional as F
+import torch.optim as optim
+import numpy as np
+from matplotlib.pyplot import imshow
+import os
+import PIL
+import numpy as np
+from matplotlib.pyplot import imshow
+import neural_network
+from matplotlib.pyplot import imshow
 
 
-# wcześniej grinder.py
+# wcześniej grader.py
 # Get accuracy for neural_network model 'network_model.pth'
 def NN_accuracy():
     # Create the model
-    model = neural_network.Net()
+    net = neural_network.Net()
 
     # Load state_dict
-    neural_network.load_network_from_structure(model)
-
-    # Create the preprocessing transformation here
-    transform = transforms.Compose([neural_network.Negative(), transforms.ToTensor()])
-
-    # load your image(s)
-    img = PIL.Image.open('../src/test/0_100.jpg')
-    img2 = PIL.Image.open('../src/test/1_100.jpg')
-    img3 = PIL.Image.open('../src/test/4_100.jpg')
-    img4 = PIL.Image.open('../src/test/5_100.jpg')
-
-    # Transform
-    input = transform(img)
-    input2 = transform(img2)
-    input3 = transform(img3)
-    input4 = transform(img4)
-
-    # unsqueeze batch dimension, in case you are dealing with a single image
-    input = input.unsqueeze(0)
-    input2 = input2.unsqueeze(0)
-    input3 = input3.unsqueeze(0)
-    input4 = input4.unsqueeze(0)
+    neural_network.load_network_from_structure(net)
 
     # Set model to eval
-    model.eval()
+    net.eval()
 
-    # Get prediction
-    output = model(input)
-    output2 = model(input2)
-    output3 = model(input3)
-    output4 = model(input4)
+    device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
-    print(output)
-    index = output.cpu().data.numpy().argmax()
-    print(index)
+    folderlist = os.listdir(os.path.dirname(__file__) + "\\test")
 
-    print(output2)
-    index = output2.cpu().data.numpy().argmax()
-    print(index)
+    tested = 0
+    correct = 0
 
-    print(output3)
-    index = output3.cpu().data.numpy().argmax()
-    print(index)
+    for folder in folderlist:
+        for file in os.listdir(os.path.dirname(__file__) + "\\test\\" + folder):
+            if neural_network.result_from_network(net, os.path.dirname(__file__) + "\\test\\" + folder + "\\" + file) == folder:
+                correct += 1
+                tested += 1
+            else:
+                tested += 1
 
-    print(output4)
-    index = output4.cpu().data.numpy().argmax()
-    print(index)
+    print(correct/tested)
 
 
 if __name__ == "__main__":