Name: TensorFlowの内部：TensorFlow Lite (Inside TensorFlow: TensorFlow Lite)
Uploaded: 2021-01-14T10:05:11.000Z
Duration: 37 min 31 s
Description: VoiceTubeの動画で発音を聞きながら英語表現を覚えよう！学べる英語：

JARED DUKE: Thanks everybody for showing up.

I'm an engineer on the TensorFlow Lite team.

Today I will be giving a very high level overview

with a few deep dives into the TensorFlow Lite

stack, what it is, why we have it, what it can do for you.

And if you have any questions, feel free to interrupt me.

And you know, this is meant to be enlightening for you.

and what motivated its creation, why we have it

in the first place and we can't just use TensorFlow on devices.

I'll briefly review how you actually use TensorFlow Lite.

And then talk a little bit about performance considerations.

How you can get the best performance on device

Why do you need TensorFlow Lite in your life?

Well, again, here's some kind of boilerplate motivation

But these are actually important use cases.

You can't just always be running inference in the cloud

A lot of devices, particularly in developing countries,

They want that done locally on their phone.

Streaming to the cloud, again, is problematic.

So if you can do things locally, particularly with a hardware

there are a number of challenges with memory constraints,

with the low powered CPUs that we have on mobile devices.

There's also a very kind of fragmented and heterogeneous

you have a primary provider of your acceleration backend

So again, TensorFlow works great on large well-powered devices

in the cloud, locally on beefy workstation machines.

But TensorFlow Lite is not focused on these cases.

So stepping back a bit, we've had TensorFlow

This is actually what we call the TensorFlow mobile project.

And after a lot of effort, and a lot of hours,

able to create kind of a reduced variant of TensorFlow

with a reduced operator set and a trimmed down runtime.

on where we could go in terms of the size of the binary.

could make that runtime a bit more extensible,

how we could map it onto all these different kinds

of accelerators that you get in a mobile environment.

And while there have been a lot of improvements

in the TensorFlow ecosystem with respect to modularity,

AUDIENCE: How small a memory do you need to get to?

JARED DUKE: So this is just the binary size.

In terms of memory, it's highly model dependent.

then you may be required to use lots of memory.

that we've taken into account with TensorFlow Lite

AUDIENCE: But your size, how small is it?

JARED DUKE: So the core interpreter runtime

字幕リスト動画再生

TensorFlowの内部：TensorFlow Lite (Inside TensorFlow: TensorFlow Lite)

basically

aware

advantage

audience