diff --git a/stopsigncnn.py b/stopsigncnn.py
new file mode 100644
index 0000000..6512b94
--- /dev/null
+++ b/stopsigncnn.py
@@ -0,0 +1,197 @@
+# -*- coding: utf-8 -*-
+"""stopSignCNN.ipynb
+
+Automatically generated by Colaboratory.
+
+Original file is located at
+    https://colab.research.google.com/drive/1BLR4DHO2qn7DOEyZ-ysSMOTFPWsSZWf7
+"""
+
+import numpy as np 
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+import torch.optim as optim
+import torchvision
+from torch.utils.data.sampler import SubsetRandomSampler 
+import torchvision.transforms as transforms 
+import matplotlib.pyplot as plt
+import os
+import time
+
+#mount googledrive
+'''from google.colab import drive
+drive.mount('/content/gdrive',force_remount=True)'''
+
+#More helper functions
+#Transforms:
+''' compose , torchvision.transforms.RandomChoice(transforms)
+color jitter
+figure out what random affine means.
+random perspective
+random rotation
+LinearTransformation Chandra was talking about?
+RandomErasing
+'''
+np.random.seed(None)
+U1 = np.random.random()
+U2 = np.random.random()
+U3 = np.random.random()
+U4 = np.random.random()
+
+transformations = [];
+
+#color jitters 
+
+t1 = torchvision.transforms.ColorJitter(brightness=U1, contrast=U2, saturation=U3, hue=0.1) 
+t2 = torchvision.transforms.ColorJitter(brightness=U2, contrast=U3, saturation=U4, hue=0.2) 
+t3 = torchvision.transforms.ColorJitter(brightness=U3, contrast=U4, saturation=U1, hue=0.3) 
+t4 = torchvision.transforms.ColorJitter(brightness=U4, contrast=U1, saturation=U2, hue=0.4) 
+
+jitterList = [t1,t2,t3,t4];
+
+#No Transformation
+tn = torchvision.transforms.Resize((200,200),interpolation=2)
+
+#Rotations
+degrees = (-30,30)
+t5 = torchvision.transforms.RandomRotation(degrees, resample=False, expand=False, center=None, fill=0)
+
+#Random Perspectives
+t6 = torchvision.transforms.RandomPerspective(distortion_scale=0.5, p=1, interpolation=3, fill=0)
+
+#Affine (basically a shear)
+t7 = torchvision.transforms.RandomAffine(degrees, translate=None, scale=None, shear=degrees, resample=False, fillcolor=0)
+
+#compose
+transformationsList = [t5,t6,t7,tn]; # now adding t8 to test
+
+#Creating two sets of transformation for images
+randomJitter = torchvision.transforms.RandomChoice(jitterList);
+randomTransformation = torchvision.transforms.RandomChoice(transformationsList); 
+
+transformations.append(randomJitter)
+transformations.append(randomTransformation)
+
+#Path to Dataset
+master_path = '--------- INSERT LOCAL PATH OF DATASET HERE --------'
+
+#Helper Functions
+def get_relevant_indices(dataset, classes, target_classes):
+    indices = []
+    for i in range(len(dataset)):
+        # Check if the label is in the target classes
+        label_index = dataset[i][1] # ex: 3
+        label_class = classes[label_index] # ex: 'cat'
+        if label_class in target_classes:
+            indices.append(i)
+    return indices
+
+
+def get_data_loader(target_classes, batch_size): #TRANSFORMATIONS APPLIED HERE
+  transform = transforms.Compose([transforms.Resize((200,200),interpolation=2),randomJitter,randomTransformation,
+                                       transforms.ToTensor()])
+
+
+  # classes are folders in each directory with these names
+  classes = ["stopSign","notStopSign"]
+
+  #Creating the entire Training Dataset 
+  trainset = torchvision.datasets.ImageFolder(master_path, transform=transform)
+
+  #Getting the indices for training set inorder to split to validation and training
+  relevant_indices = get_relevant_indices(trainset,classes,target_classes)
+
+  #Split into train and validation
+  np.random.seed(1000)
+  np.random.shuffle(relevant_indices)
+  split = int(len(relevant_indices) * 0.70) #split at 70%
+  
+  '''Currently do not have sufficient data for a test set'''
+
+  #split into train and validation indices 
+  relevant_train_indices, relevant_val_indices = relevant_indices[:split], relevant_indices[split:]
+  train_sampler = SubsetRandomSampler(relevant_train_indices)
+  train_loader = torch.utils.data.DataLoader(trainset, batch_size=batch_size,
+                                               num_workers=1, sampler=train_sampler)
+  val_sampler = SubsetRandomSampler(relevant_val_indices)
+  val_loader = torch.utils.data.DataLoader(trainset, batch_size=batch_size,
+                                              num_workers=1, sampler=val_sampler)
+
+  return train_loader, val_loader, classes
+train_loader, val_loader, classes = get_data_loader(target_classes=["stopSign","notStopSign"],batch_size=8)
+
+def get_accuracy(model, data_loader):
+    correct = 0
+    total = 0
+    for imgs, labels in data_loader:
+        
+        if use_cuda and torch.cuda.is_available():
+          imgs = imgs.cuda()
+          labels = labels.cuda()
+
+        output = model(imgs)
+        #select index with maximum prediction score
+        pred = output.max(1, keepdim=True)[1]
+        correct += pred.eq(labels.view_as(pred)).sum().item()
+        total += imgs.shape[0]
+    return correct / total
+
+# Training Curve
+def plot_training_curve(path):
+    """ Plots the training curve for a model run, given the csv files
+    containing the train/validation error/loss.
+    Args:
+        path: The base path of the csv files produced during training
+    """
+    
+    train_acc = np.loadtxt("{}_train_acc.csv".format(path))
+    val_acc = np.loadtxt("{}_val_acc.csv".format(path))
+    plt.title("Train vs Validation Accuracy")
+    n = len(train_acc) # number of epochs
+    plt.plot(range(1,n+1), train_acc, label="Train")
+    plt.plot(range(1,n+1), val_acc, label="Validation")
+    plt.xlabel("Epoch")
+    plt.ylabel("Accuracy")
+    plt.legend(loc='best')
+    plt.show()
+
+#use_cuda = True
+#train(CNNClassifier(), train_loader, val_loader, batch_size=8, num_epochs=10, learn_rate=0.00025)
+
+k = 0
+for images, labels in train_loader:
+    # since batch_size = 1, there is only 1 image in `images`
+    image = images[0]
+    # place the colour channel at the end, instead of at the beginning
+    img = np.transpose(image, [1,2,0])
+    # normalize pixel intensity values to [0, 1]
+    img = img / 2 + 0.5
+    plt.subplot(10, 10, k+1)
+    plt.axis('off')
+    plt.imshow(img)
+
+    k += 1
+    if k > 99:
+        break
+
+class stopSignCNN(nn.Module):
+    def __init__(self):
+        super(stopSignCNN, self).__init__()
+        self.name = "stopSignCNN"
+        self.conv1 = nn.Conv2d(3, 5, 5)
+        self.pool = nn.MaxPool2d(2, 2)
+        self.conv2 = nn.Conv2d(5, 10, 5)
+        self.fc1 = nn.Linear(10 * 47 * 47, 220)
+        self.fc2 = nn.Linear(220, 2) #number of things to classify
+
+    def forward(self, img):
+        x = self.pool(F.relu(self.conv1(img)))
+        x = self.pool(F.relu(self.conv2(x)))
+        x = x.view(-1, 10 * 47 * 47)
+        x = F.relu(self.fc1(x))
+        x = self.fc2(x)
+        return x
+
+use_cuda = True
+train(stopSignCNN(), train_loader, val_loader, batch_size=8, num_epochs=10, learn_rate=0.00025)
\ No newline at end of file