Interactive model based pricing UI Various domains such as business intelligence and journalism have made many achievements with help of data analytics based on machine learning (ML). While a lot of work has studied how to reduce the cost of training, storing, and deploying ML models, there is little work on eliminating the data collection and purchase cost. Existing data markets provide only simplistic mechanism allowing the sale of fixed datasets with fixed price, which potentially hurts not only ML model availability to buyers with limited budget, but market expansion and thus sellers’ revenue as well. In this work, we demonstrate Nimbus, a data market framework for ML model exchange. Instead of pricing data, Nimbus prices ML models directly, which we call model-based pricing (MBP). Through interactive interfaces, the audience can play the role of sellers to vend their own ML models with different price requirements, as well as the role of buyers to purchase ML model instances with different accuracy/budget constraints. We will further demonstrate how much gain of sellers’ revenue and buyers’ affordability Nimbus can achieve with low runtime cost via both real time and offline results.

Reference: [Our Demo Paper] [Our Poster] [Full Paper]

Requirements:

pip install flask

pip install bokeh

This is the view port for seller uploading files.

In ./SellerUpload, run these command in order:

export FLASK_APP=sellerupload.py

flask run --port 5000

Then open localhost:5000

This is the view port for seller.

Type in the following command from the root directory:

bokeh server Seller --port 5006

Then open localhost:5006 in browser.

This is the view port for buyer.

Type in the following command from the root directory:

bokeh server Buyer --port 5007

Then open localhost:5007 in browser.

Flask: flask run --host=0.0.0.0

Bokeh: bokeh serve buyer --port <port> --allow-websocket-origin <your-ip>:<port>