You have seen the impressive accuracy of your new model. In this video, test out the model on unseen images.
- [Instructor] Now in order to test … how good our newly-trained model f_model is, … I'm going to use an image of a tulip and a sunflower. … So let's upload these images to Databricks. … So I select Data, I select Add Data, … I select Browse, I select the flower image, … which is in my Exercise files in the Images folder. … I click on the Databricks icon again, … I select Data, I select Add Data, … I select Browse, I select the tulip file … and this time I make a note of where this file was stored. … So I copy that. … I select the Databricks icon again, … and then I go back to my Notebook by just clicking Back. … And let's store the path to the tulip file … in the variable t. … So t equals tulip.jpg … and in the same way, s for sunflower … and the name of my image is sunflower.jpg. … Now what I need to do is to take these image files … and convert them into DataFrames. … So from pyspark.ml.image import ImageSchema. … And this method should already be exported … based on what we've done …
- Components in the Apache Spark ecosystem
- What is deep learning?
- Using deep learning in Spark
- Working with images in Spark
- Using pre-trained models
- Testing your model
- Deploying models as SQL functions