Home Courses Apparel Recommendation Engine Workshop ConvNets: How to featurize an image: edges, shapes, parts

ConvNets: How to featurize an image: edges, shapes, parts

Instructor: Applied AI Course Duration: 11 mins

This content is restricted. Please Login

Prev

Next

Building a real world solution

Using Keras + Tensorflow to extract features

Text based product similarity

1.1

Converting text to an n-D vector: bag of words

14 min

1.2

Code for bag of words based product similarity

26 min

1.3

TF-IDF: featurizing text based on word-importance

17 min

1.4

Code for TF-IDF based product similarity

10 min

1.5

Code for IDF based product similarity

9 min

Text Semantics based product similarity

2.1

Word2Vec: featurizing text based on semantic similarity

19 min

2.2

Code for Average Word2Vec product similarity

15 min

2.3

TF-IDF weighted Word2Vec

9 min

2.4

Code for IDF weighted Word2Vec product similarity

6 min

2.5

Weighted similarity using brand and color

9 min

2.6

Code for weighted similarity

7 min

2.7

Building a real world solution

5 min

Deep learning based visual product similarity

3.1

ConvNets: How to featurize an image: edges, shapes, parts

11 min

3.2

Using Keras + Tensorflow to extract features

8 min

3.3

Visual similarity based product similarity

6 min

How do we measure the goodness of our solutions?

4.1

7 min

Exercise

5.1

Build a weighted Nearest neighbor model using Visual, Text, Brand and Color

6 min