iiinova — where your digital self keeps evolving

Rory Byrne

@rory.bio

539 followers 1.4K following 74 posts

Neuroscience PhD student, Cambridge UK. Confused but excited. "Everything around me was somebody's lifework" 👋 rory.bio 🔧 compmotifs.com

Posts Replies Media

Rory Byrne @rory.bio · 16/02/2025

Now, you can search your commit history to find good results and the associated code state.

110

Rory Byrne @rory.bio · 16/02/2025

Then when your experiment runs, logis will commit your code for you, with a nice commit message and experiment metadata at the bottom.

110

Rory Byrne @rory.bio · 16/02/2025

Or use the (work-in-progress) implicit API, where logis finds your parameters/metrics in the arguments and return value.

110

Rory Byrne @rory.bio · 16/02/2025

The SDK is similar to Weights & Biases, just add relevant data to your experiment's run.

110

Rory Byrne @rory.bio · 09/01/2025

There is an API for building custom @hooks - they're just functions.

100

Rory Byrne @rory.bio · 09/01/2025

You can also pre-process the raw data using @hooks. The @env:MY_VARIABLE hook loads data from an environment variable. The @value:path.to.val hook references a value in the raw input data. The @import:foo.bar dynamically imports a Python object.

100

Rory Byrne @rory.bio · 09/01/2025

The idea is simple: write an object's Blueprint once, then whenever you need that object in your @pydantic.dev models, just pass in the parameters. In #machinelearning and #neuroai, we build a lot of Tensors from various parameterisations...

121

Rory Byrne @rory.bio · 08/01/2025

Supports Dale's Law: inhibitory neurons (blue columns) have local within-module connections only.

110

Rory Byrne @rory.bio · 08/01/2025

🔧 Re-sharing an old tool: a recursive function to generate hierarchical, modular, Dalean connectivity matrices in PyTorch. Modules are locally dense, with increasingly sparse connections to more distal modules. #neuroai #neuroskyence gist.github.com/rorybyrne/dd...

171

Rory Byrne @rory.bio · 10/12/2024

I've been using a custom version of this tool to parameterise neuron models for a long time now.

110

Rory Byrne @rory.bio · 10/12/2024

1. Write a parameterisation ("Cast") and register it with the `cast.for_type(Tensor)` decorator. 2. Use `CastModel` to build your @pydantic.dev models. 3. You can now pass the _parameters_ of your tensor, and receive a built tensor.

110