#!/usr/bin/env python
# coding: utf-8

# In[1]:


import numpy as np
from keras.preprocessing import image
from keras.applications import resnet50


# In[2]:


from keras import backend as K
print(K.backend())


# In[3]:


# load pre-trained model
model = resnet50.ResNet50()


# In[4]:


# load and resixe image to match model nodes
img = image.load_img("Exercise Files/05/bay.jpg", target_size=(224, 224))


# <img src="bay.jpg" width="224">

# In[5]:


# convert image to np array
x = image.img_to_array(img)


# In[6]:


# the model expects multiple images, a list
x = np.expand_dims(x, axis=0)


# In[7]:


# scale input image to range used in trained NN
x = resnet50.preprocess_input(x)


# In[8]:


# run image through NN to make a predict
predictions = model.predict(x)


# In[11]:


# look up predicted class names
# returns top 5
# add ", top=n" for less/more results in `decode_predictions` arguments
predicted_classes = resnet50.decode_predictions(predictions)


# In[12]:


print("This is an image of:")

for imagenet_id, name, likelihood in predicted_classes[0]:
    print(" - {}: {:2f} likelihood".format(name, likelihood))


# In[ ]: