Skip to content

Latest commit

 

History

History

simple-example--fibonacci-model

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
src
src
 
 
Dockerfile
Dockerfile
 
 
README.md
README.md
 
 
model_definition.yaml
model_definition.yaml
 
 

README.md

Fibonacci Model

This is an example model. Here you should describe what the model does.

This particular model generates a Fibonacci sequence. e.g.:

[0, 1, 1, 2, 3, 5, 8, 13, 21, 34, ...]

You can set three inputs:

  • SEQUENCE_LENGTH - Dictates the count of numbers in the generated sequence
  • SEQUENCE_F0 - The first number to start the Fibonacci sequence with.
  • SEQUENCE_F1 - The second number to start the Fibonacci sequence with.

The model generates a JSON file containing the sequence:

{
  "sequence": [0, 1, 1, 2, 3, 5]
}

Technical Info

This model pulls in several environment variables and passes them into a piece of Python code. The Python code is built into a Docker image using the provided Dockerfile.

There are five files here:

  • main.py - This is the file initially called by DAFNI. It is referenced in the Dockerfile.
  • work.py - This contains the main code for the model. Here, as an example, it is a simple Fibonacci generator
  • Dockerfile - Builds the container that will be run by DAFNI
  • model_definition.yaml - A machine-readable file used to define the model. This information will be shown to other users who may wish to use your model.
  • README.md - This helpful file. It should contain detailed information about the model.

Dependencies

This model requires Python and Docker to be installed in order to build and run locally.

Running the Model

You can run this example model from within the "simple-example" folder by doing the following:

python ./src/main.py

You can also run this model using Docker from within the same folder by doing:

docker build -t fibonacci-model .
docker run fibonacci-model

You don't need to run the build step every time you want to run the model, you only need to re-run it if changes are made to the Dockerfile or any of the files that go into the Docker image.

You can adjust the way it runs with the environment variables:

docker run -e SEQUENCE_LENGTH=50 fibonacci-model

or

docker run -e SEQUENCE_LENGTH=50 -e SEQUENCE_F0=1 -e SEQUENCE_F1=3 fibonacci-model

Uploading to DAFNI

You will need to create a file from your docker image to upload it. Check out the detailed instructions online at Docs but you can create it with:

docker save -o fibonacci-model.tar fibonacci-model

You can then also compress it before uploading:

gzip example-model.tar