notebook.community

Edit and run

Using deep features to build an image classifier

Fire up GraphLab Create





In [20]:



    


import graphlab

Load a common image analysis dataset

We will use a popular benchmark dataset in computer vision called CIFAR-10.

(We've reduced the data to just 4 categories = {'cat','bird','automobile','dog'}.)

This dataset is already split into a training set and test set.





In [3]:



    


image_train = graphlab.SFrame('image_train_data/')
image_test = graphlab.SFrame('image_test_data/')



    


















    






[INFO] 1448853893 : INFO:     (initialize_globals_from_environment:282): Setting configuration variable GRAPHLAB_FILEIO_ALTERNATIVE_SSL_CERT_FILE to /Users/jrrembert/venvs/dato-env/lib/python2.7/site-packages/certifi/cacert.pem
1448853893 : INFO:     (initialize_globals_from_environment:282): Setting configuration variable GRAPHLAB_FILEIO_ALTERNATIVE_SSL_CERT_DIR to 
This non-commercial license of GraphLab Create is assigned to j.ryan.rembert@gmail.com and will expire on October 13, 2016. For commercial licensing options, visit https://dato.com/buy/.

[INFO] Start server at: ipc:///tmp/graphlab_server-61007 - Server binary: /Users/jrrembert/venvs/dato-env/lib/python2.7/site-packages/graphlab/unity_server - Server log: /tmp/graphlab_server_1448853893.log
[INFO] GraphLab Server Version: 1.7.1

Exploring the image data





In [4]:



    


graphlab.canvas.set_target('ipynb')



    











In [30]:



    


image_train['image'].show()

Train a classifier on the raw image pixels

We first start by training a classifier on just the raw pixels of the image.





In [6]:



    


raw_pixel_model = graphlab.logistic_classifier.create(image_train,target='label',
                                              features=['image_array'])



    


















    






PROGRESS: Creating a validation set from 5 percent of training data. This may take a while.
          You can set ``validation_set=None`` to disable validation tracking.

PROGRESS: Logistic regression:
PROGRESS: --------------------------------------------------------
PROGRESS: Number of examples          : 1910
PROGRESS: Number of classes           : 4
PROGRESS: Number of feature columns   : 1
PROGRESS: Number of unpacked features : 3072
PROGRESS: Number of coefficients    : 9219
PROGRESS: Starting L-BFGS
PROGRESS: --------------------------------------------------------
PROGRESS: +-----------+----------+-----------+--------------+-------------------+---------------------+
PROGRESS: | Iteration | Passes   | Step size | Elapsed Time | Training-accuracy | Validation-accuracy |
PROGRESS: +-----------+----------+-----------+--------------+-------------------+---------------------+
PROGRESS: | 1         | 6        | 0.000010  | 2.225952     | 0.291099          | 0.284211            |
PROGRESS: | 2         | 9        | 5.000000  | 2.865258     | 0.418325          | 0.442105            |
PROGRESS: | 3         | 10       | 5.000000  | 3.120809     | 0.342932          | 0.389474            |
PROGRESS: | 4         | 12       | 1.000000  | 3.582159     | 0.427749          | 0.452632            |
PROGRESS: | 5         | 13       | 1.000000  | 3.852789     | 0.429843          | 0.463158            |
PROGRESS: | 6         | 14       | 1.000000  | 4.167679     | 0.424607          | 0.452632            |
PROGRESS: | 10        | 19       | 1.000000  | 5.430332     | 0.517801          | 0.536842            |
PROGRESS: +-----------+----------+-----------+--------------+-------------------+---------------------+
PROGRESS: TERMINATED: Iteration limit reached.
PROGRESS: This model may not be optimal. To improve it, consider increasing `max_iterations`.

Make a prediction with the simple model based on raw pixels





In [7]:



    


image_test[0:3]['image'].show()



    


















    



























In [8]:



    


image_test[0:3]['label']



    


















    
Out[8]:








dtype: str
Rows: 3
['cat', 'automobile', 'cat']

















In [9]:



    


raw_pixel_model.predict(image_test[0:3])



    


















    
Out[9]:








dtype: str
Rows: 3
['bird', 'cat', 'bird']

The model makes wrong predictions for all three images.

Evaluating raw pixel model on test data





In [10]:



    


raw_pixel_model.evaluate(image_test)



    


















    
Out[10]:








{'accuracy': 0.47225, 'auc': 0.7179540833333328, 'confusion_matrix': Columns:
 	target_label	str
 	predicted_label	str
 	count	int
 
 Rows: 16
 
 Data:
 +--------------+-----------------+-------+
 | target_label | predicted_label | count |
 +--------------+-----------------+-------+
 |  automobile  |       bird      |   92  |
 |     bird     |       bird      |  493  |
 |     bird     |       dog       |  174  |
 |     dog      |       dog       |  414  |
 |     cat      |       dog       |  301  |
 |     dog      |    automobile   |  136  |
 |     bird     |       cat       |  153  |
 |     cat      |       cat       |  302  |
 |     cat      |    automobile   |  197  |
 |     dog      |       bird      |  227  |
 +--------------+-----------------+-------+
 [16 rows x 3 columns]
 Note: Only the head of the SFrame is printed.
 You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns., 'f1_score': 0.4652947170586577, 'log_loss': 1.218737142291628, 'precision': 0.46253672369976395, 'recall': 0.47224999999999995, 'roc_curve': Columns:
 	threshold	float
 	fpr	float
 	tpr	float
 	p	int
 	n	int
 	class	int
 
 Rows: 400004
 
 Data:
 +-----------+-----+-----+------+------+-------+
 | threshold | fpr | tpr |  p   |  n   | class |
 +-----------+-----+-----+------+------+-------+
 |    0.0    | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   1e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   2e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   3e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   4e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   5e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   6e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   7e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   8e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 |   9e-05   | 1.0 | 1.0 | 1000 | 3000 |   0   |
 +-----------+-----+-----+------+------+-------+
 [400004 rows x 6 columns]
 Note: Only the head of the SFrame is printed.
 You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.}

The accuracy of this model is poor, getting only about 46% accuracy.

Can we improve the model using deep features

We only have 2005 data points, so it is not possible to train a deep neural network effectively with so little data. Instead, we will use transfer learning: using deep features trained on the full ImageNet dataset, we will train a simple model on this small dataset.





In [11]:



    


len(image_train)



    


















    
Out[11]:








2005

Computing deep features for our images

The two lines below allow us to compute deep features. This computation takes a little while, so we have already computed them and saved the results as a column in the data you loaded.

(Note that if you would like to compute such deep features and have a GPU on your machine, you should use the GPU enabled GraphLab Create, which will be significantly faster for this task.)





In [11]:



    


#deep_learning_model = graphlab.load_model('http://s3.amazonaws.com/GraphLab-Datasets/deeplearning/imagenet_model_iter45')
#image_train['deep_features'] = deep_learning_model.extract_features(image_train)

As we can see, the column deep_features already contains the pre-computed deep features for this data.





In [13]:



    


image_train.head()



    


















    
Out[13]:








    

        id
        image
        label
        deep_features
        image_array
    

    

        24
        Height: 32 Width: 32
        bird
        [0.242871761322,
1.09545373917, 0.0, ...		
        [73.0, 77.0, 58.0, 71.0,
68.0, 50.0, 77.0, 69.0, ...		
    

    

        33
        Height: 32 Width: 32
        cat
        [0.525087952614, 0.0,
0.0, 0.0, 0.0, 0.0, ...		
        [7.0, 5.0, 8.0, 7.0, 5.0,
8.0, 5.0, 4.0, 6.0, 7.0, ...		
    

    

        36
        Height: 32 Width: 32
        cat
        [0.566015958786, 0.0,
0.0, 0.0, 0.0, 0.0, ...		
        [169.0, 122.0, 65.0,
131.0, 108.0, 75.0, ...		
    

    

        70
        Height: 32 Width: 32
        dog
        [1.12979578972, 0.0, 0.0,
0.778194487095, 0.0, ...		
        [154.0, 179.0, 152.0,
159.0, 183.0, 157.0, ...		
    

    

        90
        Height: 32 Width: 32
        bird
        [1.71786928177, 0.0, 0.0,
0.0, 0.0, 0.0, ...		
        [216.0, 195.0, 180.0,
201.0, 178.0, 160.0, ...		
    

    

        97
        Height: 32 Width: 32
        automobile
        [1.57818555832, 0.0, 0.0,
0.0, 0.0, 0.0, ...		
        [33.0, 44.0, 27.0, 29.0,
44.0, 31.0, 32.0, 45.0, ...		
    

    

        107
        Height: 32 Width: 32
        dog
        [0.0, 0.0,
0.220677852631, 0.0,  ...		
        [97.0, 51.0, 31.0, 104.0,
58.0, 38.0, 107.0, 61.0, ...		
    

    

        121
        Height: 32 Width: 32
        bird
        [0.0, 0.23753464222, 0.0,
0.0, 0.0, 0.0, ...		
        [93.0, 96.0, 88.0, 102.0,
106.0, 97.0, 117.0, ...		
    

    

        136
        Height: 32 Width: 32
        automobile
        [0.0, 0.0, 0.0, 0.0, 0.0,
0.0, 7.5737862587, 0.0, ...		
        [35.0, 59.0, 53.0, 36.0,
56.0, 56.0, 42.0, 62.0, ...		
    

    

        138
        Height: 32 Width: 32
        bird
        [0.658935725689, 0.0,
0.0, 0.0, 0.0, 0.0, ...		
        [205.0, 193.0, 195.0,
200.0, 187.0, 193.0, ...		
    



[10 rows x 5 columns]

Given the deep features, let's train a classifier





In [14]:



    


deep_features_model = graphlab.logistic_classifier.create(image_train,
                                                         features=['deep_features'],
                                                         target='label')



    


















    






PROGRESS: Creating a validation set from 5 percent of training data. This may take a while.
          You can set ``validation_set=None`` to disable validation tracking.

PROGRESS: WARNING: Detected extremely low variance for feature(s) 'deep_features' because all entries are nearly the same.
Proceeding with model training using all features. If the model does not provide results of adequate quality, exclude the above mentioned feature(s) from the input dataset.
PROGRESS: Logistic regression:
PROGRESS: --------------------------------------------------------
PROGRESS: Number of examples          : 1905
PROGRESS: Number of classes           : 4
PROGRESS: Number of feature columns   : 1
PROGRESS: Number of unpacked features : 4096
PROGRESS: Number of coefficients    : 12291
PROGRESS: Starting L-BFGS
PROGRESS: --------------------------------------------------------
PROGRESS: +-----------+----------+-----------+--------------+-------------------+---------------------+
PROGRESS: | Iteration | Passes   | Step size | Elapsed Time | Training-accuracy | Validation-accuracy |
PROGRESS: +-----------+----------+-----------+--------------+-------------------+---------------------+
PROGRESS: | 1         | 5        | 0.000131  | 1.756336     | 0.716010          | 0.630000            |
PROGRESS: | 2         | 9        | 0.250000  | 3.314907     | 0.776903          | 0.680000            |
PROGRESS: | 3         | 10       | 0.250000  | 3.908626     | 0.776378          | 0.690000            |
PROGRESS: | 4         | 11       | 0.250000  | 4.475182     | 0.778478          | 0.700000            |
PROGRESS: | 5         | 12       | 0.250000  | 4.926555     | 0.786877          | 0.740000            |
PROGRESS: | 6         | 13       | 0.250000  | 5.399979     | 0.804199          | 0.750000            |
PROGRESS: | 10        | 17       | 0.250000  | 7.091473     | 0.872966          | 0.790000            |
PROGRESS: +-----------+----------+-----------+--------------+-------------------+---------------------+
PROGRESS: TERMINATED: Iteration limit reached.
PROGRESS: This model may not be optimal. To improve it, consider increasing `max_iterations`.

Apply the deep features model to first few images of test set





In [15]:



    


image_test[0:3]['image'].show()



    


















    



























In [16]:



    


deep_features_model.predict(image_test[0:3])



    


















    
Out[16]:








dtype: str
Rows: 3
['cat', 'automobile', 'cat']

The classifier with deep features gets all of these images right!

Compute test_data accuracy of deep_features_model

As we can see, deep features provide us with significantly better accuracy (about 78%)





In [17]:



    


deep_features_model.evaluate(image_test)



    


















    
Out[17]:








{'accuracy': 0.78725, 'auc': 0.939880166666664, 'confusion_matrix': Columns:
 	target_label	str
 	predicted_label	str
 	count	int
 
 Rows: 16
 
 Data:
 +--------------+-----------------+-------+
 | target_label | predicted_label | count |
 +--------------+-----------------+-------+
 |     dog      |       cat       |  200  |
 |     cat      |       cat       |  646  |
 |  automobile  |       dog       |   5   |
 |     cat      |    automobile   |   41  |
 |     dog      |       bird      |   47  |
 |     cat      |       bird      |   82  |
 |     bird     |       dog       |   63  |
 |     cat      |       dog       |  231  |
 |     dog      |       dog       |  734  |
 |     dog      |    automobile   |   19  |
 +--------------+-----------------+-------+
 [16 rows x 3 columns]
 Note: Only the head of the SFrame is printed.
 You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns., 'f1_score': 0.7861794490122769, 'log_loss': 0.5702696641668693, 'precision': 0.785671852075494, 'recall': 0.78725, 'roc_curve': Columns:
 	threshold	float
 	fpr	float
 	tpr	float
 	p	int
 	n	int
 	class	int
 
 Rows: 400004
 
 Data:
 +-----------+----------------+-----+------+------+-------+
 | threshold |      fpr       | tpr |  p   |  n   | class |
 +-----------+----------------+-----+------+------+-------+
 |    0.0    |      1.0       | 1.0 | 1000 | 3000 |   0   |
 |   1e-05   | 0.964666666667 | 1.0 | 1000 | 3000 |   0   |
 |   2e-05   |     0.951      | 1.0 | 1000 | 3000 |   0   |
 |   3e-05   | 0.944666666667 | 1.0 | 1000 | 3000 |   0   |
 |   4e-05   | 0.936666666667 | 1.0 | 1000 | 3000 |   0   |
 |   5e-05   | 0.929333333333 | 1.0 | 1000 | 3000 |   0   |
 |   6e-05   | 0.925666666667 | 1.0 | 1000 | 3000 |   0   |
 |   7e-05   | 0.921333333333 | 1.0 | 1000 | 3000 |   0   |
 |   8e-05   |     0.917      | 1.0 | 1000 | 3000 |   0   |
 |   9e-05   | 0.912333333333 | 1.0 | 1000 | 3000 |   0   |
 +-----------+----------------+-----+------+------+-------+
 [400004 rows x 6 columns]
 Note: Only the head of the SFrame is printed.
 You can use print_rows(num_rows=m, num_columns=n) to print more rows and columns.}

















In [21]:



    


image_train['label'].sketch_summary()



    


















    
Out[21]:








+------------------+-------+----------+
|       item       | value | is exact |
+------------------+-------+----------+
|      Length      |  2005 |   Yes    |
| # Missing Values |   0   |   Yes    |
| # unique values  |   4   |    No    |
+------------------+-------+----------+

Most frequent items:
+-------+------------+-----+-----+------+
| value | automobile | cat | dog | bird |
+-------+------------+-----+-----+------+
| count |    509     | 509 | 509 | 478  |
+-------+------------+-----+-----+------+


















In [36]:



    


auto_data = image_train[image_train['label'] == 'automobile']
cat_data = image_train[image_train['label'] == 'cat']
dog_data = image_train[image_train['label'] == 'dog']
bird_data = image_train[image_train['label'] == 'bird']



    











In [37]:



    


len(auto_data)



    


















    
Out[37]:








509

















In [ ]:

Content source: jrrembert/cybernetic-organism

Similar notebooks:

notebook.community | gallery | about