New stuff

MobileNetV1.py

 import torch
+
 import torch.nn as nn
+import torch.nn.functional as F
 from torchsummary import summary
 import requests
 from PIL import Image
 from io import BytesIO
 
+from torchvision import transforms
+
+import numpy as np
+
+import pandas as pd 
+import os
+#from pycocotools.coco import COCO
+import skimage.io as io
+import matplotlib.pyplot as plt
+from pathlib import Path
+
 from coco_utils import get_image
 
+
 model = torch.hub.load('pytorch/vision:v0.10.0', 'mobilenet_v2', pretrained=True)
+#Enable following to only get CNN part:
+model = torch.nn.Sequential(*(list(model.children())[:-1]))
 model.eval()
 
-from PIL import Image
-from torchvision import transforms
-input_image = get_image("http://images.cocodataset.org/val2017/000000148783.jpg")
 preprocess = transforms.Compose([
-    transforms.Resize(256),
-    transforms.CenterCrop(224),
-    transforms.ToTensor(),
-    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
+  transforms.Resize(256),
+  transforms.CenterCrop(224),
+  transforms.ToTensor(),
+  transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
 ])
-input_tensor = preprocess(input_image)
-input_batch = input_tensor.unsqueeze(0) # create a mini-batch as expected by the model
 
-# move the input and model to GPU for speed if available
-if torch.cuda.is_available():
+#Outputs mini-batch that can be input to model
+def preprocessImage(image):
+  input_tensor = preprocess(input_image)
+  return input_tensor.unsqueeze(0) # create a mini-batch as expected by the model
+
+#Runs the CNN on input
+def runCNN(image):
+  input_batch = preprocessImage(image)
+  if torch.cuda.is_available():
     input_batch = input_batch.to('cuda')
     model.to('cuda')
+  with torch.no_grad():
+    return model(input_batch)
 
-with torch.no_grad():
-    output = model(input_batch)
-# Tensor of shape 1000, with confidence scores over Imagenet's 1000 classes
-# print(output[0])
-import numpy as np
-# # The output has unnormalized scores. To get probabilities, you can run a softmax on it.
-probabilities = torch.nn.functional.softmax(output[0], dim=0)
-# print(probabilities)
-a = np.argmax(probabilities)
-print(a)
-print(probabilities[a])
\ No newline at end of file
+
+if __name__=="__main__":
+    #Run entire thing
+    input_image = get_image("http://images.cocodataset.org/val2017/000000148783.jpg")
+    output = runCNN(input_image)
+    output.shape
+    #Shape of output:
+    #torch.Size([1, 1280, 7, 7])
\ No newline at end of file