Matplotlib Tutorial 9 - converting data from the internet

Key Takeaways

what is going on everybody Welcome to the ninth data visualization with python and matap plot lib tutorial video where we left off we were basically pulling data from the internet we were we got the data we've separated it and we can even unpack it into these variables here but the problem is the date stamp here is a little problematic for us because it's not like a legible date okay so so to us we understand like okay we can see okay this is 2010 this is June 29th so we can kind of understand that but we have to be be very explicit to our programs uh what that date is now a lot of programs use what's called Unix time and Unix time is just the number of seconds after January 1st 1970 mot lip however doesn't use Unix time okay it has its own little conversion thing I don't know I don't even know what the what the conversion is but it usually Corr responds to numbers like 70,000 or so so anyways uh that's that moving along um so we need to convert this and to start converting data we have to understand that it the data comes through as like a a bite string it's not even a string so we have to convert it now usually just decode will work but to convert it to a true date stamp in string form for whatever reason do decode is not enough we have to do our own kind of decoding as well so in this uh tutorial we're going to write that this decoding function here because it doesn't exist yet we'll write that and then we'll actually pull the data and we'll graph it if the star is a line so uh let's clean this function up a little bit though there's no reason to have all these white spaces all right so we're going to make this function so let's copy that come up to the top and now we're going to Define and we're going to say bite state to num and then we basically the first thing that gets passed here we're going to say is fmt that's short for format so format is this data right here it's the format of that date stamp now the next thing that we're going to do here is um basically we've got format and then we're going to actually have one more thing we're going to have the encoding and the encoding is um utf8 that's okay so that's what we need for the internet data so now we build this function we're going to say St Str converter and that's going to be equal to MD dates. stpd to num so this is strip dat to number based on what based on the format okay now just for the record if you don't understand this kind of decoding stuff that's okay that's not what this tutorial is supposed to be over back in Python 2 this was not necessary this function already kind of existed for you you didn't need to know how it worked but since it doesn't work in Python 3 for some reason we have to write our own okay and uh so this is hopefully you can understand feel free to ask about it if you want but don't be like frustrated if you don't understand it so anyway string converter that's going to be equal to M dates. stripd to number based on the format and again this is basically that that's going to strip it for the matap plot live date number that 70,000 number whatever that means then we're going to define the bytes uh conver and it's it converts b as in bites and it Returns the S which is just string and that's b. decode based on the encoding which is utf8 then it Returns the string converter converter the string converter of that new element s return not return and then uh at the end of this function we return the uh bytes converter okay so uh yeah so now return by yeah Okay so we've got that that'll return basically this answer and that should be good we should have the conversion if something goes wrong we'll we'll revisit that so now everything should be unpacked um I'm saying this as if like the program is like running live I'm just kind of running through the program in my head so we've unpacked everything with that function we have now uh converted everything and now we're actually ready to plot this stuff so um we we should be able to get rid of just like calling PLT do plot and close and stuff so let's try that and see what we do so we'll say PLT pt. plot and then the X will be a date and instead of just plot let's actually call plot plore date this notifies python that hey we're plotting dates uh and then so we plot dates and then date and then we'll plot the close P for close price and that should be fine the X label now is not X it's date y label is not yit price and that should be good let's uh let's run that and see see what happens um labels so for some reason we're passing we got this labels line which we shouldn't have gotten interesting one two three four five six so this labels is in there as well for some reason so if so let's go back up to this if values not in line and let's just add um if values not in line and labels not in line line some reason that was there I don't really recall ever seeing that one uh so if that's all true let's run this one more time okay we got a price awesome so this is uh when we use plot date for some reason the default is like more of a scatter I'm not sure why but we'll fix that right now uh so we go plot date and then just do a comma and then add the argument of just like this Dash so it's like you know your minus sign run that one more time and then now we've got just the price now we're getting this error about No Label that's because we didn't label the price so we can do um we'll just do keyword here label actually let's do here keyword argument should come last label equals um price okay good enough all right so now we've got uh we're plotting and you'll notice here we've got the dates down here at the bottom and what's kind of nifty is if we um pull this like this and like if we zoom in a specific spot you'll see that that they've expanded like originally they were just the years but if we zoom in we get the month in the year now we can keep zooming in now we've got the the the month Day in the year okay so depending on your Zoom uh that that date stamp is going to vary a little bit now what happens though if we zoom in all the way and we make this like the original size that it was we've got a little bit of overlap there on the dates um or like if we do this right so one little Nifty thing to do is to like rotate the dates and stuff and so we're going to be talking about rotating the dates and kind of cleaning up this chart a little bit in the uh next tutorial but at least for now you should be able to have this you know Price Line at least plotted up that we grabbed from Yahoo so if you have any questions comments concerns for whatever reason if your graph has not popped up please feel free to post below otherwise as always thanks for watching thanks for all the support subscriptions and until next time