Proposal: ML + web3 model deployment survey // part 2

Name of Project:
ML + web3 model deployment survey

this is an extension of: Proposal: ML + web3 model deployment survey

Proposal in one sentence:
Conduct a hands-on survey of ML model deployment options for web3 use cases

Description of the project and what problem is it solving:

The intersection of ML and web3 is very important, but currently very green (more thoughts on that here). One of the core capabilities that will be needed is model deployment and inference.

The goal of this project is to evaluate a few of the more popular decentralized compute networks for their readiness to support ML model deployment. This will involve going through a live deployment of a CNN trained on MNIST, consuming that model from a client, and assessing the solution.

More specifically, will be evaluating

  • Ocean compute-to-data (done)
  • Bacalhau
  • Golem Network
  • Fetch.ai

For these properties

  • economics and monetization
  • inference or job types
  • hardware support
  • model and data formats
  • model management

Grant Deliverables:

  • written report on findings and results
  • code for each deployment experiment

Squad

Devin Conley


Progress since part 1:

All code is being pushed to a public repo here:

So far, have completed:

  • simple cnn model training
  • ocean asset deployment (model weights, MNIST dataset, inference algorithm)
  • ocean compute 2 data job
  • cnn inference algorithm

Survey reporting available here:

1 Like

Finally back to this and wrapped up the implementation for Fetch.ai

MNIST classifier service agent

Consumer agent

Notes dropped in the same document linked above

Golem implementation complete!
https://twitter.com/devinaconley/status/1752811785835499702


code here