Welcome to Algorithmia! This guide is designed as an introduction to hosting your CNTK model and publishing it as an algorithm, even if you’ve never used Algorithmia before!
Train and save your model
After your CNTK model is trained you should save the model graph with the idiomatic
model.save(filepath). Once that’s finished saving to your local disk, you’re ready to upload it to Algorithmia.
Create a Data Collection
Here you’ll want to create a data collection to host your model.
To use the Data API, log into your Algorithmia account and create a data collection via the Data Collections page.
Click on “Add Collection” under the “My Collections” section on your data collections page.
After you create your collection you can set the read and write access on your data collection. For more information check out: Data Collection Types
Upload your Model into a Collection
After your collection is created, you’re going to want to upload your saved graph file to your newly created Data Collection.
Load model by clicking box “Drop files here to upload”
Note the path to your files:
Create your Algorithm
Creating your algorithm is easy!
- To add an algorithm, simply click “Add Algorithm” from the user profile icon.
- Name your algorithm, select the language, choose permissions and make the code either open or closed source.
Note: There is also a checkbox for ‘Standard Execution Environment’ or ‘Advanced GPU’. For deep learning models you will want to check ‘Advanced GPU’.
Note: Make sure that your version of python is the same between your development environment and the algorithm. There may be version conflicts otherwise.
set your dependencies
Now is the time to set the depenencies your model relies on.
- Click on the “Dependencies” button at the top right of the UI and list your packages under the required ones already listed and click “Save Dependencies” on the bottom right corner.
Load your Model
Here is where you load your model, which will be called by the
apply() function when you pass input to the algorithm using our API.
Our recommendation is to preload your model in a separate function before the apply(). The reasoning behind this is because when your model is first loaded it can take some time to load depending on the file size. However, with all subsequent calls only the apply() function gets called which will be much faster since your model is already loaded!
Note that you always want to create valid JSON input and output in your algorithm. For example this algorithm takes a JSON blob passing in a csv file hosted on Algorithmia, Amazon S3, or Dropbox.
Preloading your CNTK model is a great step for improving your overall performance. By downloading your CNTK model file when the container first starts, it means that any subsequent requests to the algorithm won’t have the same loading overhead!
Here’s an example of what that looks like:
Publishing your Algorithm
Last is publishing your algorithm. The best part of hosting your model on Algorithmia is that users can access it via an API that takes only a few lines of code to use! Here is what you can set when publishing your algorithm:
Set version permissions to public or private use
Set it to royalty free or set to per-call royalty
Set access permissions to have full access to the internet and ability to call other algorithms
For more information and detailed steps: creating and publishing your algorithm