Is there an existing integration?

• I have searched the existing integrations.

Use Case

Fast.ai is a deep learning library which provides practitioners with high-level components that can quickly and easily provide state-of-the-art results in standard deep learning domains, and provides researchers with low-level components that can be mixed and matched to build new cutting-edge models.

It is a framework on top of PyTorch and so native PyTorch components can also be used.

Motivation

You can use it for quick prototyping and for neck-to-neck Kaggle contests. It aims to do both things without substantial compromises in ease of use, flexibility, or performance. Fastai includes:

• A new type dispatch system for Python along with a semantic type hierarchy for tensors
• A GPU-optimized computer vision library which can be extended in pure Python
• An optimizer which refactors out the common functionality of modern optimizers into two basic pieces, allowing optimization algorithms to be implemented in 4–5 lines of code
• A novel 2-way callback system that can access any part of the data, model, or optimizer and change it at any point during training
• A new data block API
• And much more…

API Structure

flowchart BT
    classDef stroke stroke-dasharray: 20;
    subgraph a[Application]
        direction TB
        a1(Vision) & a2(Text) & a3(Tabular) & a4(Collab)
    end
    subgraph b[High Level API]
        direction TB
        b1(Learner) & b2(DataBlock)
    end
    subgraph c[Mid Level API]
        direction TB
        c1(Callbacks) & c2(Generic Optimizer) & c3(General Metric) & c4(Data Core)
    end
    subgraph d[Low Level API]
        direction TB
        d1(Pipeline) & d2(Reversible Transforms) & d3(OO Tensors) & d4(Optimized Ops)
    end 
    d --> c --> b --> a
    class a,b,c,d stroke

Fastai is organized around two main design goals: to be approachable and rapidly productive, while also being deeply hackable and configurable. This is possible thanks to a carefully layered architecture. It is built on top of a hierarchy of lower-level APIs which provide composable building blocks. This way, a user wanting to rewrite part of the high-level API or add particular behavior to suit their needs does not have to learn how to use the lowest level.

Thanks for going through the issue. If you like the idea, then kindly upvote by reacting with a 👍 emoji below.

Implementation

No response

Anything else?

No response