--- a
+++ b/utils.py
@@ -0,0 +1,47 @@
+import torch
+from torch.autograd import Variable
+import matplotlib.pyplot as plt
+from torchnet import meter
+
+def plot_training(costs, accs):
+    '''
+    Plots curve of Cost vs epochs and Accuracy vs epochs for 'train' and 'valid' sets during training
+    '''
+    train_acc = accs['train']
+    valid_acc = accs['valid']
+    train_cost = costs['train']
+    valid_cost = costs['valid']
+    epochs = range(len(train_acc))
+
+    plt.figure(figsize=(10, 5))
+    
+    plt.subplot(1, 2, 1,)
+    plt.plot(epochs, train_acc)
+    plt.plot(epochs, valid_acc)
+    plt.legend(['train', 'valid'], loc='upper left')
+    plt.title('Accuracy')
+    
+    plt.subplot(1, 2, 2)
+    plt.plot(epochs, train_cost)
+    plt.plot(epochs, valid_cost)
+    plt.legend(['train', 'valid'], loc='upper left')
+    plt.title('Cost')
+    
+    plt.show()
+
+def n_p(x):
+    '''convert numpy float to Variable tensor float'''    
+    return Variable(torch.cuda.FloatTensor([x]), requires_grad=False)
+
+def get_count(df, cat):
+    '''
+    Returns number of images in a study type dataframe which are of abnormal or normal
+    Args:
+    df -- dataframe
+    cat -- category, "positive" for abnormal and "negative" for normal
+    '''
+    return df[df['Path'].str.contains(cat)]['Count'].sum()
+
+
+if __name__=='main':
+    pass
\ No newline at end of file