Unbacked SymInts

This podcast episode explores unbacked cements in PyTorch 2, symbolic integers and conditional branches, unbacked cements for export and non-zero compilation cases, unbacked simmants and broadcasting, broadcasting and shape propagation in TensorFlow, optimization of tensor handling for export, zero-one specialization in PyTorch and unbacked cements.

Outlines

Sign in to continue reading, translating and more.

Continue

PyTorch Developer Podcast

Unbacked Cements in PyTorch 2: Symbolic Integers and Conditional Branches

Unbacked Cements: Beyond Export

Unraveling the Complexities of Unbacked Simmants and Broadcasting in PyTorch

Understanding Broadcasting and Shape Propagation in TensorFlow

Optimizing Tensor Handling for Export: Non-Overlapping, Dense, and Zero-One Specialization

Unbacked Cements in PyTorch: A Deep Dive into Zero-One Specialization

Unbacked Cements: A Flexible Approach to Dynamically Sized Models

Unbacked SymInts

PyTorch Developer Podcast

00:00Unbacked Cements in PyTorch 2: Symbolic Integers and Conditional Branches

Unbacked Cements in PyTorch 2: Symbolic Integers and Conditional Branches

04:20Unbacked Cements: Beyond Export

Unbacked Cements: Beyond Export

07:06Unraveling the Complexities of Unbacked Simmants and Broadcasting in PyTorch

Unraveling the Complexities of Unbacked Simmants and Broadcasting in PyTorch

12:03Understanding Broadcasting and Shape Propagation in TensorFlow

Understanding Broadcasting and Shape Propagation in TensorFlow

14:06Optimizing Tensor Handling for Export: Non-Overlapping, Dense, and Zero-One Specialization

Optimizing Tensor Handling for Export: Non-Overlapping, Dense, and Zero-One Specialization

16:04Unbacked Cements in PyTorch: A Deep Dive into Zero-One Specialization

Unbacked Cements in PyTorch: A Deep Dive into Zero-One Specialization

19:31Unbacked Cements: A Flexible Approach to Dynamically Sized Models

Unbacked Cements: A Flexible Approach to Dynamically Sized Models