Everybody.

And welcome to part two of our playing with Google takeout data video miniseries thing in this video we're gonna be doing is now building some like bar graphs.

Basically, from the top 10 were prodded and go with most common words with the sliding one year window.

And later we can try different windows, or you guys could just do it on her on your own time.

But for now, we'll do a one year window.

So with that, let's go ahead and get started.

So we should you should at this point have my life dot database and that should be populated with the, you know, the primary key UNIX and then all of the unique words.

So, uh, what I'm gonna do now is I'm just gonna copy search database.

I'm just gonna rename this to be, um I don't know, images, uh, and then open super close, that old one.

So it doesn't really matter anymore.

So now, and in fact, I kind of want to keep the original just so I don't have to re type some stuff.

Um, so we still want to connect to my life database.

We don't need the search activity directory anymore.

We don't need apart states, and that should be good.

So the next thing I'm gonna do is we want we need a directory that we're gonna dump all these images into.

So I'm just gonna make a new directory real quick.

I'm gonna call this images.

Uh, let's do word images, and we're gonna start with one year window one day slide.

It's kind of a long clump directory, but we'll go with it.

That's fine.

Um, so why, uh, well, wolf work that in in a minute.

So now what we're gonna do is we're going to specify Cem starting constants.

So a day we always knows 86 486,400 seconds.

Um, and then, like, a year would be equal to 365 day, um, and then for our slide, we're going to say that is equal to one day.

Okay, so with that, what we're gonna do is just start generating or word graph.

So we're say, define define words.

Graph.

Uh, now, what we want to do is we're gonna save everything into here, so I'm gonna copy that.

Come into here and I'm gonna say, saved their equals.

That so that's we're gonna stuff all the images now we want to do is first of all, we need to starting in unending time.

So we should be able to do, like a query.

I think so.

Something like I don't know what it is.

Let's look it up for a quick sq light Max column escalate, Max.

Okay, hold on.

Find sq light.

Find Max.

We could also try men.

I just want to see the expression man select Max from Okay, So?

So the first thing that we could do is we could Look, all we want to do is find the minimum time in maximum time.

So the words sports Kraft.

And then we're going to say, um well, we're just gonna do it, si dot Execute.

Um, I guess select Max UNIX Look of like, a hair.

Something over here.

Select.

You know, Max UNIX from, um from words.

And then let me just print.

Um I'm just gonna say d equals c dot Fetch all.

And then we'll front d.

Okay, so those are max time, so then we'll just say Max time equals D.

Uh a deep.

Well, we could just say this.

We can't call to fetch ALS, but we should be able to just do that.

And then 00 right.

00 And then we should We're doing the stupid tab thing, huh?

What?

Why don't you let me do this?

There we go.

Thank you, sir.

Thank you.

That's so generous of you.

Okay, so now, Max time and then we'll do the same thing with minimum time.

So boom, change this to be men and then change this to be men, and then we'll print, um men Time, Max time.

Run that minimum time.

Okay, cool.

So that looks like it works pretty darn well.

So now we just need to start sliding over.

So are our start time will be equal to the men in time, and then our this is like our current.

So this is for our window.

So our start times, men time our end time for the specific window will be men Time.

Plus, I thought we had window.

What do we call this thing year?

Oh, we're just calling a year.

Okay.

I think I'll call.

I'll make a variable called window equals one ex year cool.

Meantime, plus window.

Okay, So, um, so that's our starting our end.

And then now what we want to do is we wanted just, like, iterated over.

Um, try to think of how exactly I would do this.

So I guess we'll go.

Wow.

Loop.

So, like, wow.

Um wow.

And is less than max time.

We want to make the query, um, for all of the words between start and end.

So, um si dot execute and we'll make a platoon f string.

Um, select word from words.

It looks like we lose our syntax highlighting when we go f string on a query.

That's okay.

Aah!

Select word from words where UNIX is greater, then start.

And UNIX is less than and beautiful.

And then we're gonna say, uh, data data equals c dot Fetch a ll, and then I'm gonna say break here, and then I'm gonna print data.

Fantastic.

So now that we have that, um, so this is like all our individual words in that timeframe now we want to do is take all of these words and throw them into some sort of list, and then we'll do.

We'll create a counter object and then do dot most common.

10.

So So before I forget, Because that's the next thing we're gonna do from collections.

Import counter.

Done.

Okay, so, Data.

So then we wantto iterated through here and upend this two words.

So while this blah, blah, blah Okay, um, we'll do words is empty to equal to an empty list.

And then what we're gonna say is four well, summer yet is Rohan data.

So for road in data, word equals ro zero, um, words dot upend row zero.

Now, that's just on the fly.

But we really could probably What could we do here?

We could say words equals, um I zero right.

Four.

I skin data, right.

We should be able to get away with that.

So I am going to do this.

I 004 Okay.

What do you mean?

Okay, cool.

All right.

So that works.

So that gives us our words.

Now we can create a counter object.

So what we can say is, um, word underscore, er counter equals counter words.

And then what we could say is, um, we could say common words equals word.

Underscore counter dot Most underscore.

Common 10.

And then, rather than printing all of the words, we can print the common words.

Why did that just happen?

Because we're printing data, I guess.

Let's try one more time.

Yes, So you can see the top of the charts.

Python Not really too surprising.

Bitcoin pi probably with raspberry pine.

It is kind of weird that we have more pie than raspberry.

Not quite sure why that would be.

Maybe I type owed raspberry.

That's probably it.

Price forgot the Pee or something.

But pie, it's fascinating.

I almost kind of want to find out if that's right.

I bet I just type owed raspberry.

But anyway, cool Lennox S and P for the S P 500 again, probably because I don't tell.

Sometimes I type S P 500 sometimes S and P 500 but 500 always is the case map.

What?

Lib?

I don't know what's happening here with Google.

That would be an interesting one and then Satan himself.

So Okay, so that's our most common.

Now we'd want to graft those, so that would be on a single time.

So So how We're gonna make a bar graph here and We need a couple of things when we make a bar graph.

I wish bar graphs were easier.

And Matt plot lib.

Which, by the way, let's go ahead and import before I forget from Lizzie.

So, import Matt plus matt plot lived up.

I plot as put from Matt plot lib, import style and style that use a good plot.

Okay, great.

Beautiful.

Couldn't ask for a better set of imports there.

So now, um, we need to kind of set up how we're gonna do this.

So common words probable.

Okay.

Common words is good.

We don't need to print commoners anymore.

Uh, and instead, what we want to say is we're gonna set the y position.

So why oppose equals?

M p dot a range?

I don't think part, Len.

Common words in private.

10.

But later, we might add that to be, like, constant that we change over time.

The important, um, pie Or have I have not import numb pie as in p.

Back to work.

Why?

Position?

Great.

Um, and then we're gonna say a word underscore counts is equal to so actually what we need to say, um, doing it common words.

What instead, What I'm gonna say is, um Topic zero for topic in this cool.

So it'll be common words.

And then the word counts will just kind of be the topic.

Uh, one.

So this will be the actual word itself in the counter, cause the counter object was a list of two bulls.

Um, where the first object in the to pull is tthe e uh, word.

And then the 2nd 1 was a count.

So cool.

Okay, so now that we have that we're ready to actually graft this and because we're gonna save it, we wantto adjust the size right away.

So we're gonna peel tea dot figure, and we're gonna say fig signs is equal to 12 and seven.

That should gives a 1200 by 700 pixel.

Um, graph.

So that's our appeal to dot figure.

Then we're gonna peel tea dot bar.

Why?

Pause.

Word counts and a line underscore her.

I'm sorry.

Not a line center, and then we're gonna give it in Alfa of not to pull 0.5 just to make it semi transparent.

Well, it will be beautiful.

And then p lt dot exe ticks.

Uh, wipo's common words.

So that way the ex ticks, rather than being like numbers, aren't said the actual word.

Because that's actually what we're trying to graph.

Here is the words.

And then how much like the Y axes will be volume.

So with that said, Peel, tea, why label people we'll call this volume and then finally, we also do a title and will display the date in the title.

So on a single frame it'll be somewhat ledge a ble.

But mostly what we want to do is make it somewhat readable while the frames are going at, like 30 or 60 frames a second.

So what I'm gonna do is make this an extreme, and I think we're gonna bring it yet another import, um, from daytime import date time.

And then we'll create a I'm trying to decide what I want to do there.

It's probably the dumbest way to do it.

Well, it's done this way, but I'm someone's gonna come up with a better way, but whatever.

So what I'm going to say here is F um, we're going to do Dave Month year because that'll be the easiest to follow.

I think so.

Just understand that's what's happening.

All non Americans are probably very happy from time stamps.

We're gonna get this daytime object from a time stamp.

Meaning you next time and we'll go with end because that's like it's like the last years of data for today.

Okay, so macro info up to today.

So we're gonna use the ending time stamp.

Um, so this and then and then what we can say is dot day, month, year dot day day.

There we go.

And that gives us our daytime objects that will give us a day like the number of the month.