Advanced PyTorch Graph Manipulation: FX Graph Mode Quantization Coding tutorial - Part 3/3
If you need help with anything quantization or ML related (e.g. debugging code) feel free to book a 30 minute consultation session! https://calendly.com/oscar-savolainen
I'm also available for long-term freelance work, e.g. for training / productionizing models, teaching AI concepts, etc.
Video Summary:
In this video, in part 3/3 of our coding tutorial on how to quantize a PyTorch ResNet with FX Graph Mode Quantization, we look at two advanced graph manipulation techniques.
The first is how one can iteratively propagate a tensor through the graph, and access intermediary tensors in a way that is reminiscent of forward hooks. The second is how one can swap out nodes: we use the example of using BatchNorm layers into their preceding Convolution layers.
Timestamps:
00:00 Intro
02:39 Iterating through the graph
11:17 Replacing nodes in the graph
18:42 Outro
Links:
Github for code: https://github.com/OscarSavolainen/Qu...
Connect on LinkedIn: / oscar-savolainen-phd-b88277121
On this page of the site you can watch the video online Advanced PyTorch Graph Manipulation: FX Graph Mode Quantization Coding tutorial - Part 3/3 with a duration of hours minute second in good quality, which was uploaded by the user Oscar Savolainen 22 March 2024, share the link with friends and acquaintances, this video has already been watched 528 times on youtube and it was liked by 17 viewers. Enjoy your viewing!
