About this guide

The following guide was written to help you get to know the cnvrg.io platform and to show you how cnvrg.io help you build and organise machine learning projects. With a real use-case, we will briefly go through the platform: from data loading and tagging to researching and experimentation to deploying models as REST APIs.

Create a project & connect a git repository

To create a new project, go to your organization's home page and click on
Start Project.
Set a name, description(optional) and click on start.

Now you have a new project in your organization!

Connecting a git project is done easily via the project home page or from project settings' page. 

From the project's home page:

And click on Save.

Now cnvrg imdb example repository is connected!

Upload Dataset

To Upload a new dataset via the web interface, simply go to your Organization's Datasets tab – and click “New Dataset”. 

  1. Give your dataset unique name and Choose type "Text".
  2. Download imdb dataset file from the following url: https://s3.amazonaws.com/text-datasets/imdb.npz
  3. Upload imdb database file to your dataset via web interface.

Running Experiments

Experiments can be run via web. This is especially useful when you want to run an experiment fast without syncing your code. To run an experiment via the web interface, simply go to your project’s experiments tab – and click “New Experiment”.

Fill "Command to Execute":  

python3 train.py --dataset_path=/data/YOUR_DATASET_NAME/imdb.npz

Choose Compute:

Choose your dataset:

Leave all other fields with default value and click "Submit".

The experiment will run for 5-10 minutes and create h5 file.

Deploying models

So we’ve ran experiment (You can run as much experiments as you want to optimized your model) and we’re good to go. Let’s deploy the model as a REST API.

To Deploy model via the web interface, simply go to your project’s Publish tab – and click “Publish”. 

Fill "File": 

predict.py

And fill "Function to execute":

predict

Choose 'cnvrg commit' related to your best experiment:

Leave all other fields with default value and click "Submit".

After Publishing complete, you can start using suggested API (under "Usage Example") in any platform you want.

All endpoints in cnvrg.io are constantly being monitored, so data scientists can see if a model needs their attention (retrain/shutdown). All Input/output of the models is being stored to enable deeper research and help reproduce predictions.  

That's it. This was a short example of using cnvrg.io platform.

Did this answer your question?