Heatmap for Subscription Churn - Dash Plotly

hi everybody welcome welcome welcome to this tutorial where we are going to learn about the heat map let me just make sure that i see um the chat of my viewers so i can help you along the way there we go let's make this a little bit smaller put this in the side all right so thank you for joining everybody today we are going to learn about the heat map we are going to see how to create a heat map with plotly in python we're going to see when to use it the advantages of a heat map and anything you need to know about a heat map um if you would like to follow along because we are going to do this together i'm doing this live tutorial so we can actually do this together and i can help you out with any questions or clarifications that you need so go into this github directory that i have here i'm going to copy it in the chat and download some of this code that we're going to use the first code we're going to use is the heat map time so make sure to download here or just copy paste and into your spider your um jupiter or pycharm visual studios and we can actually get started and do this together if you're watching this video after it's been live and it's recorded i'm going to put this link under the video description so you can actually follow along as well i'm also going to put a video layout section under the video description so you can go to any section that you want and skip if you want to see the berlin crime heat map or you want to see the subscription churn heat map you can just use a video layout i'm going to give two opportunities for questions clarifications opportunities to learn about dash plotly in this live tutorial we'll do the first break after the mba season winning percentage and the second one is going to be at the very end i'm going to finish this subscription churn stay here with you and just answer any questions you might have or anything you want to know i'm here to support you if you need me to learn dash plotly and data visualization in python um hey relative i'm glad i'm glad you learned a lot oh the multi-page so that's a hard one i'm glad you're actually taking that on okay so let's get started um we're going to do this one first this heat map the nba season winner percentage um so download this heat map time from the github link that i just sent you and and let's do this together this is it there we go right here all right let's erase this i'll just practicing for you okay so the first thing we're going to do hey um so the first thing we're going to let me just keep this here in case somebody has questions um big concerns all right so don't forget the first thing you want to do is import the libraries let's import um plotly dot express as px and let's import pandas as pd okay if you don't have these libraries go into your terminal and spider visual studios and just all you have to do is pick install pandas pan does make sure to spell it correctly and then when this is done takes about two minutes pip install plotly and you should be ready to go everything should be started you can copy paste the code and everything is going to work okay let's make this a little bit smaller here next thing you want to do is um to call in the data right so we had data inside our um csv inside our github so we're going to read it because it's a csv document so all you have to do is go into the github um go into the nba data right here and then we're just going to copy paste the raw version of it and go into raw and then you'll see copy paste okay let's go back back and go here copy paste into read csv perfect and now let's print and see that we actually have it sprint the first 15 rows let's print the first 22 rows now let's go wild here with 20 rows um save let's run it okay let's see what we get so now we're printing our mba data we see that we get um all teams well we're getting celtics because it's going back to 2016 to to 19 i think 45 or 1950 and then we'll see other teams okay and then we'll see the winning percentage per per team per per year per season right celtics is very well here in 2016 2017. right so now that we have the data what we need to do is to create a pivot table if you want things to work out and to appear on a heat map just like this just all the examples we have here you have to create one way to do it the fastest way to do it is to create a pivot table so to do that just do df we'll use the pandas pivot method we'll do df dot pivot and inside in pycharm you should click control control and on the pivot with the mouse you'll see that we it has three arguments inside the pivot method a function you have um the index the columns and the values okay so the index is going to be team right this one right here we're going to choose that as the index the columns is going to be um year and then the value is going to be the winning percentage i'm just going to copy this here right now let's print it out and see what we get because we need to create this pivot table well the whole pivot table just print out the whole pivot table and now you see that we get a pivot table where the team is the index we have all these teams right here make it bigger for you the year represents all the columns and then the winner percentages represents the values so remember this this is really how you want your data frame to look like in order to put it on a heat map the fastest way possible you have index and then the columns above and to the right and then you have the values okay so now that we have our pivot table all we have to do is just create the figure so figure equals we'll use polyexpress i am show the image show which is that um um heat map and like this and then fig show and now it should work this comes from here i'm going to put it under the video as well this is the main page that will teach you all about um how to create uh all the parameters for a heat map inside of here make sure you have the latest version of plotly for this to work and you can read all about the different parameters okay um so now that we put our pivoted table inside the heat map we can just run it and it should um show us all the teams right we started this as an example with us five teams but here it should actually show us all of the teams together let's run this again a heat map wow this is loading oops um it's something that you should is very similar to to a bar chart see all the teams here in the different years um but it shows a lot more data a heat map is a two-dimensional visualization of data where the color represents a third dimension and i find that if it's a lot of data it's a lot better to use sometimes a heat map than a bar chart because sometimes it's hard to see the different heights of the bar chart and see what's high or what's not and for me sometimes to look at the colors and see different trends with the colors makes things a lot easier so here for example you see that let's take this heat map with fewer teams the very dark colors were high windy percentage precedent percentages warriors did very well in this year um and the bulls did not very well because it's very very light color and then you can um compare between teams right so if you take for example 2016 you see the warriors did a lot better than the knicks a little bit better than celtics and so on and so on so it gives you because it's like a three-dimensional um plot it gives you a lot of um a lot of information that a bar graph will not necessarily give you okay so this is how we created um the first heat map um what you want to do oops once you create this heat map um you have a few other things you can do if you wanted to for example limit the amount of teams that show on the heat map because it might be hard to see all these teams then you can use what we have here in the github where in the code where we're indexing the teams out so for example copy based here all right so you see how we're saying take the data frame or the index because the index is the teams right where the index is only equal to these five teams right so if we do that this will allow us to create this kind of chart where we have a limited amount of teams a few other things um that you can do with um with the heat map is you can actually just copy paste this here and see what else happens is all these functions this is very important in order to customize your heat map so look here in the heat map of the px i am um show uh method the heat map um you can choose your color right this you choose it from the properly expressed built-in sequential colors okay we're choosing this one right here if you have pycharm you can choose it will give you a list of all the ones that you can choose from and i'm going to show you that list right now okay let's first of all see what this color looks like the list comes from here i'm also going to add this under the under the video maybe just add it right now so you can take a look at the chat there you go take a look at this what's up how you doing undergraduate good luck with your studies yeah probably really nice all right so um go into go into here um in this link and um you'll see the different sequential colors so you can use um the default is plasma is this one i think but i chose for example right now we chose to do this one y i o r or y l i think i'm not sure so it's one of these and this one i think right so let's see what it looks like oh it's still loading sometimes when i do the video it's a bit slower than loading so i'm sorry about that let's see what that looks like okay so here the these are the the different colors it's not plasma plasma is the default this is a different color let's try one more color let's try for example anybody in the chat want a certain color tell me now tell me now going once going twice okay we'll just choose this color i think there is a five second lag no no this is hard reds let's choose uh reds so if i change that and i save it and i run it you'll see um the sequential color right here the reds one right so it's a little bit lighter than than this one that we chose before let's see um in addition the color is going to be loading in one second in addition to the colors we can also add a title with the title inside of this and then we're using the fig update layout to um increase the title font to 27 so it's bigger i'm putting title underscore x meaning 0.5 meaning i'm putting the title right in the middle of the page and by doing hover longs hover on gaps falls this is what it does if i do hover false this is original without sorry where is it okay this is the red colors you see if i hover on this i'll see different hover i'll see different information but if there's no data it won't show any hover and this is because we did hover on gaps false all right you see the title is in the middle it's a bigger font and it's a nice red color where the dark red is a high percentage of winning and the light red is a very low percentage of winning poor celtics um so this is the hovering gaps false the last parameter inside the update traces is a hover template you can use the hover template to make your template information um be whatever information you want right so as you can see we're saying that the y-axis is going to be called team we're going to have team um colon and then we're going to have the y-axis information we're going to then skip a row go break down a row and then we're going to have year colon and then we're going to have the x-axis information so you have team 76ers year 2007 and then you will have winning percentage and you will have the z access information which is the color right so winning percentage is 0.2 if you did not put this then i think team would go there in there by default but this would change to color the z is always the last row where you see winning is always color by default so this is a way to change the text inside the harvard template all right so buenas um so we just learned how to do our first heat map right here our nba heat map um we'll take a two minute break to answer any questions and to let you catch up and follow along and this break i just wanted to um highlight that many of these tutorials are possible because of the support that i get from from you guys from everybody that's watching me you ladies i get the support through the charming data patron community um and so to show your appreciation if these videos have been helping you a lot and to help me make more of these high quality videos um i would be very grateful if you could uh join my um community um the link is right here if you are uh interested and capable of doing so i'm going to add it here in the chat i'm also going to add it under the under the the video if you're watching a recording by joining this patreon community you'll get access to different dash potly tips and tricks and i'm also going to offer these um two codes in git gitlab i'm going to offer these two python files or you can do also a heat map of a dash data table and you can also do a heat map of um with dash so we're going to actually add a drop down and you can add as many teams as you want or take away any teams that you want to take away okay i hope to see you there but thank you for joining me so far um any questions anything you want to know before we move on to the next section which is the berlin crime distribution to see how to create this and how to create a heat map if you did not have a pivot table we're going to see that in about two minutes if you're seeing a recording of this just um fast forward to to this section the berlin crime and um under the video and we'll um you'll be able to see this or just stay for the questions okay um andy lindsey here and devashish how are you how you doing ron welcome um any questions about the heat map so far about the code about when to use it how to use it other things i'll give you about a few more seconds if there's no questions i'm glad that means everything is clear and i'll move on to the next heat map um we'll give it ten more seconds and then we'll move to the next um heat map to see how to build this these two heat maps together the subscription and the berlin crime heat map okay ten more seconds see if there's any questions okay so um in this case what we're going to do is i'm going to copy paste because a lot of the code is very similar so i don't have to write everything out if you have questions along the way please let me know aws it's it's aws is is is a bit a season book or complicado it's a bit complicated but i have a tutorial on on um on heroku and um how to upload to heroku this next week i hope to do a tutorial on upload your your web app to um upload your dashboard app to the web using python anywhere so hopefully i'll get this done by next sunday um soon to come yeah this is so nice to see you hi okay um vamos let's get let's get started let's do the second heat map we're going to do the um this heat map right here the berlin crown distribution so going to the github in case you don't have it let me copy paste it so now you have it in the chat right here and then um just copy paste the heatmap.pi file um all of it let's do the first part here let's do all of it and then we'll we'll show here and then we are going to um run it so actually before we run it i'm going to hashtag this out okay so as you can see save this run run there you go okay so as you can see here we are downloading the data from um from the github where i have it it's berlin data um let's actually see how it looks like print df 15. let's see what this data looks like it's a lot better when you when you see the data it's a lot better to understand how to how to manipulate it so the first data before we do anything is year and then disk check we see that we have multiple locations you can't see it here but you can see it on the um on the github you'll see that um berlin crimes um it has multiple locations for each district and this is important because we are going to um group um group by you see every district midte has like multiple locations inside inside this this district of berlin and this is for multiple years from 2012 all the way down to 2019 so what we are going to do you see then you have graffiti and then you have the number of cases per location per district the number of drug cases per location per district and so on and so on so what we are going to do to create this heat map right here is you can create it in different ways what we are going to do is we are going to group by district because i want to take um these four um crime types inside the data frame the data and i'm going to take only the the median of these crime types so i'm going to take the median average of all the years combined from 2012 to 2019 take the median average of these four per district so i'm not going to see all these locations i'm only going to see mitte like i see right here you see after i group by and i print it this print is bringing out this so there's per district i see the four different columns of four different crimes and the case is an average median average case for all the last seven years all right so now that i grew by this is um i have to melt it before i turn it into a pivot table i have to melt it okay because i need all these crimes to be in one column so i can't have a wide data frame of many many different columns i'm going to have a narrow data frame of one column that represents crime that represents all the different graffiti robbery um aggravated assault and burglary and one column and the value is in a different column this will make it easier to pivot it maybe you can pivot it from this um if you're a pandas expert um or if there is a way to do it i didn't know so i melted it first right so you see the id the district the id variables are going to remain to the left right here and then the values that are going to be under uh value variables under one column that is called crime is going to be all these that all these um values the graffiti all the columns right and then the last column is the value column which has all the numbers per crime per district so now that we have that right here we can actually pivot it right and remember the pivot is the index is going to be crime the column is going to be district and the values are going to be just from the value column so that's why we have if we do the pivot we end up with this right here right the crime is a district if the crime is the index the district is the columns and then all the rest of the values represent them so because you have the pivoted table all set up and done now you can just run your if you wanted to you can just do like this fig equals um px i am show and you can just run df just like this right and just let's show it you don't need everything here let's run it and see what happens taking the second to load the first time it's always slower when i load it on the first time oops there we go okay so let me close this out um the map that we're uh came up with with the crimes in berlin uh oh shoot is still loading one second while it is loading sorry with the video with live it just a bit slower but while it is loading let me teach you a few other things inside of here so as we saw before um you can you can have a color a continuous scale if you want to change the color that's going to go inside your your heat map now we're going to use plasma which is a default color okay and then we added a title and then we're going to have we're going to put the title in the middle we're going to give it a font i'm here we're changing the uh we're changing the hover template we're changing the information inside the hover if you do not have this i'll show you what happens if you don't have it run this okay give it a 10 seconds it should load here and you'll see what happens if you did not change the hover information oh i think i'm actually doing a new hit yes perfect close all the rest shoot everybody i don't know why it's so slow um all right let me just tell you this is what we're creating right if you don't have um the hover template if you don't have this section right here hover template you won't have you see how it says district crime and then cases instead of cases it would say color because color goes there by default so if you want to add um change the the text just change the the z right here the z column um information on the z-axis which is a color just change and put the text that you want in there we're going to put cases in there okay okay let's give it um one last try and then um and then we'll move on to the next one because i don't want to spend too much of your time if it's not uh if it's too slow here i'm just going to copy paste here from github give him one last try and see what happens all right all right so this is our berlin crime um distribution with the plasma and as you can see we have um you can compare between districts and you can compare between crimes so you see there's a lot more aggravated assault on average in the last seven years than robbery so that's why i like um i like the heat map so much because there's so many different ways and so much information you can see just by looking at colors all right now this is an important section let's hashtag this out this is a very very important section to learn at the very bottom here and just do all this again so this which is the bottom part of the of the code this is a way to actually build a heat map where if you do not have a pivot table you don't like working with a pivot table or you're not sure how to create a pivot table from all the data that you have the only thing you need to do really is to create a list of lists you see this list of lists that i created and it's inside the data object once you create a list of lists then you can put that inside the px.i am show with the heat map method and it will create let's hashtag this out it will actually create your um your heat map just by using a list of lists okay this is very important to know because sometimes you can't create a pivot table so if you cannot create a pivot table just create a list of lists and this is what happens you see i didn't put any x-axis didn't put any y-axis labels but i do have the same color that i have right here it's just in a different in a different um in a different way right so this is just a lift a list of lists right so this is what i did i grew by the district just like i did above this is the average like like right i did right here right grew by so this is the average of all the four um crimes over the last seven years per district right and then what i did i took these four crimes and i just did um i created i took their values and i put them inside a list so the result if i print the list the result is this this is this is important to know it's a list of different lists so you see i have one list a list of one list two lists three lists you have a total of 12 different lists here y12 because we had 12 rows of the 12 different districts right after we grouped everything inside inside districts we only have 12 different rows in this data frame you could see that if you just print here the the data frame um and so for every row every district represents a list and inside the list every every number represents a crime right the average median of the crime for the last seven years so this will represent graffiti this represents robbery there's aggravated assault and so on and so on but this is important to remember because now you know that you can create your labels and your x-axis and y-axis because in in the heat map in this method if you just pass a list of lists inside the data now you know that the x-axis always refers to the number of digits inside the the lists right so if you have um the first list has four digits and and obviously every list has four digits you know that you need four different variables on the x-axis if you don't have to call it graffiti robbery you know it refers to this but you can call it whatever you want you can call it um d a something and e right if you call this way this is what you this is what will show up on the graph below right instead of one two three you will have these values but because we know that these refer to oops these refer to graffiti robbery and burglary we might as well put here graffiti robbery and burglary okay now and the y-axis refers to the list of districts because the y-axis refers to the number of you see here the the the list the number of lists inside the big list so there's 12 lists here and this refers to the y-axis so i know if there's 12 lists i know that i need 12 variables here so i can do a list list variable number one variable number um two oops two a very number we'll call the third variable whatever i could put 12 variables in here or 12 different strings or i can just put the districts because i know they refer to the district right um so important to remember that and then this is the labels this labels refer to the hover right when this is the information that's going to be on the hover and then the color scale is a regular default color scale that we chose even if you took it out you wouldn't have it will be the same because this is a default color all right so now you can see that we have graffiti in the oops crime types in the x-axis and on the y-axis we have our different districts okay if you want to flip this around you'll have to flip your list of lists you'll just have to reshape it and this is why we have this in the code reshape the list of lists to swap the x-axis and the y-axis right if you want so just do this as a list comprehension of the data and if we print it out if you do this you have to swap the x-axis and the y-axis so just do this will be the y and this oops and this will be the x and now it should work because remember the number of lists um inside the main list right here right here refers to oh come on remember the number of lists inside the main list right here refers to the y-axis so we have one list here i think we have a total of four right then we're going to have to have four different um variables in this case it's the the crime and we if we have the numbers inside each list the number of values that here i have about 12 values refer to the to the x-axis so here i need to have 12 different variables so the x-axis is going to be district and that's why now if if the lows sorry it's lower in the video you'll see that um that it creates this heat map that's more horizontal and less vertical because we just we swapped we reshaped the list of lists okay give me one last shot one last shot by charm we're not giving you more than one more opportunity you've been slow today okay do this close these i think these are heavy on them there we go so see how instead of this way we reshaped our list of lists using this list comprehension and we changed our x-axis and y-axis with the data appropriately and now instead of this we have this um nice chart you can add a title to it and you can add whatever you want where you see the heat map of crimes in in districts in berlin all right so now that we have this we are going to move into our last subscription uh churn heat map and here what we are creating here is um we're using fake data to see how many people uh these are all the people that bought a subscription in january and then when um how many cancelled per week so zero people cancel the subscription in week 22 zero people in week 28 but there was one person that canceled will be 29 and nine people that cancel in week 38 right so we're going to put the weeks number of weeks on the x and the month that onboarding month of subscriptions on the y-axis so you have everything here inside the github code that i sent you just go into the heat map churn we'll copy paste this into your python ide like this i'm going to copy it into i'll just put in the same in the same file like this hope that it um blows a little bit faster and here you'll see we're doing we're doing a few things what we're doing here is we're changing um we're downloading the data right this is this is a churn data fake data that i created and then we're grouping by the onboard month and the week subscribe and we're taking the size why are we doing this because um we want to know inside this data frame we want to create a pivot table and to create a pivot table you can do it with unstack it's another way to create a pivot table okay so when we do unstack we're creating a pivot table we're filling the empty values with zero and this is a new thing that we're doing by filling an empty value with zero what happens is that um instead of having this empty screen like gray in the background um now everything will be blue in the background because um blue is the zero value when there is no actually no values now it's blue because it's zero so this is what you get with um by doing by doing the field value zero okay here we go so you see that we just got this loaded so instead of having a gray background here it has blue because we change it to zero right and we yeah um and then we're going to re-index this because if you would print here print the df you would see that april comes first it goes by alphabetically april and then i can't remember the second month so i'm re indexing everything so this january comes first and then december comes last and you can see that i printed it out right here and you get this pivot table the onboard month is the index the week subscribe is the y column and then the values are the you know the values inside the pivot table and then i'm just going to show it here like we did before we're going to show the df we'll choose this color sequential color um title font put in the middle and just change the the hover template okay i want to show you what happens um if i don't change the hover template let's see what happens by default if you did not change the hover template you will not have um cancellations in the last row it will just say color okay if it loads i will show you um so this is this is this is pretty much it i wanted to show you how to create a heat map when it's the best time to use it and um and and the short way and then the fast way with um with um pivot table is a short way and fast way and then the the longer way but you have a little bit maybe more control is with the list of lists that we just went over if you enjoyed this live tutorial um stick around i'll have some room for questions and so we can chat a little bit um click the like button on the video um subscribe below and turn on your notifications so you can receive a video every week about data visualization um dashboards uh in in python and how to build them and join my patron um i really you know appreciate your support i appreciate people that join my community to learn more about dash plotly and to learn all about the different um ways that you can create your dashboard app i'm also going to add these to um python files inside uh to the community so they can have more more information and more data they can they can work with all right thanks everybody for joining me let us see if we have any questions camilo hey how are you camilo thank you thank you i appreciate it let me close this close this sleigh fed thank you thank you schleifer i appreciate it germany going full lockdown in three days in germany you're going full lockdown in three days i know sergio i agree um heat maps are tricky and there's different ways you can do heat maps you can do also heat maps on on the data table you can do heat maps on regular maps but they are definitely um they are tricky thank you abel i appreciate it i appreciate it yeah um i hope i hope it was helpful uh in my pc let me read some comments here sergio you said you prefer vim in [Music] what are you what are you saying there what does that mean let me see i'm sorry were you asking me which one is more efficient hey um i don't well you're not you're not too late for the questions part um we are we're still answering some questions here but i went over the whole tutorial so you can you can watch it from the beginning if you want and download the code from from here if you want access to all to all the code using pivot or not actually oh um which one is more i i think using pivot is more efficient because it's just it's just faster you don't look at this when we created this new heat map um table or here we only did we only we read the data it was it was built in a way that all i could use just to pivot uh method immediately and then and then i built my right after this i could just build my um figure just doing polyexpress i am show and just put the f and that that would be my that's it that would be my figure that would be my heat map so it's a lot more um quicker when you when you can use a pivot table but if you don't um want to you can use the list of lists and just put the data and then put the x x-axis and y-axis labels so they know what they're um uh so that so the graph looks pretty oh set for the text editor okay okay after i prefer the oh okay why do you prefer pim in the text editor pankaj welcome you're welcome although i use pycharm oh i got it okay sergio there's a simple text editor for debian okay got it hey davis um welcome i'm glad you you enjoyed and i'm glad you were able to join i'll try to do this every not every sunday because i have to also live it takes a lot of time but i'll try to upload a video every sunday and whenever i can i'll do a live video as well a live tutorial so we can get to talk and share things with each other um yes uh yes yes davis um you can for example this is the interactive part so you go here i'm sharing this with i'm going to put this in my patreon community but here i'm adding a drop down and i'm adding different there's different teams right so this drop down will add teams to your heat map so now you can see as many teams as you want or you can take teams out if you don't want to see certain teams okay so this is written with dash and it's using i'm using this uh heat map underscore dash python file so this is how you would build it with um to add more interactivity to your to your your heat map you're welcome it's me you're welcome it's me you're gonna have to change your youtube your youtube name because i i i see so many different youtube i answer so many youtube questions and followers i i can't remember the names unless unless you have the name on there i remember your logo i remember it's me seeing you a million times i'm talking to you a million times but i'm sorry um doesn't i have automatic correction and price yeah pycharm yeah python has automatic um um correction what i like about pycharm is that it will tell you if you have an error so look at this for example if i'm making a mistake here and i'm putting color let's put this first of all you you'll see automatic um auto auto complete so now it'll tell you you know the parameters you can use inside im show and then it will tell you if you have an error you see now i put sd whatever some object no string and it tells you have an error here so if i go into hover over the error it's going to hover over the red part it'll tell me unsolved reference see unsolved reference sadf so i can click on it i can go there and i'll see there's an error here and i can and i'll fix it with however it needs to be fixed so that's i used atom before but i much prefer pycharm it just it's just more um just gives me more uh help because i'm when i'm coding to understand where i have my mistakes if you you are welcome um yeah the documentation it i finally documentation documentation and dashboard is comprehensive uh but but yes it's helpful when you have tutorials and you can see videos and explanations on it um and second yeah i'll try and do this on sundays every now and then maybe once twice a month so join and we can talk more donato um title of the video where where you show dash cytoscape i've looked for it like ah no leonardo i don't think i showed that cytoscape i don't think i want to do a dash tutorial on cytoscape an introduction but i do not think i i did that if anybody is interested what lornado is talking about is a dish this this dash let's do matchfigure cytoscape is a great library inside dash go into the main dash website and then go into um dash cytoskate right here you'll see under dash open source libraries click on this and you'll see dash cytoscape overview i'm going to copy paste the code so yep the link so you have it right here in the chat now donato is talking about this right here which is um which is a library inside dash it allows you to create nodes uh highlighting them labels oops whoops zoomed out too much um let's see some layout functions or callbacks and really cool things inside of the nose come on come back to me come back to me here you go ah i'm zooming out too much let's go inside here see okay different nodes different cities you can move them around you can do circles and different ways so if you know the callback if you do the crop manipulate the callback very well you you know how to use it then you can create cytoscape library you can use that to create these kind of nice networks so yeah i don't know i hope that answers your question um yeah i know i know it's me if we discuss many many questions it's me as well i just don't remember the name because i talked to a lot of people i'm sorry um yeah i agree it does i well i don't know vs code i heard really good things about it so i i can't say pycharm is better i just um i i haven't tried uh vs code yet but i've tried pycharm and atom and i like pycharm more um jonathan yes you need to install you need to install to run the library you need to install a plotly and and panda if you are referring to the tutorial um you need to install those two things my code editor is pycharm is it the one yes i use pycharm or for nodes i will use dash but i use i'll put everything on pycharm okay morning here in new york a lot of sunlight finally it was so cold in the last week or so but now it looks a lot better how you doing nice to see you i thought about we were learning we were learning heat maps but i started my tutorial at 9 45 so um so we we ended it finished now i'm just answering questions but um yeah today we went over a heat map so we learned how to do um heat map with subscription churn we learn how to do this heat map of berlin crimes and how to do a time index heat map time on the x-axis of nba percentages um and i'm sure i shared the code here my github so if you need it here you go this is the code where are you you can use the code einstein is there a free way to include metrics in deployment in heroku i don't think i'm not sure einstein i'm not sure there's a free way to include metrics in deployment in heroic i wouldn't be surprised if they charge for that because that's how they make their money thank you yuvraj um thank you you raj you're welcome you're welcome here davis um charm gives you a virtual environment option starting a new problem that is a charming figure about python does anyone know if this code doesn't automatic start project would be nice to know thank you it's me thank you thank you very much um davis that's a good question does anybody know if if you can use um vs code you can do virtual environment on vs code i would be surprised if you couldn't but i'm yeah i don't know i don't didn't use vs go but if anybody know please let davis know if you can use virtual environment with um with vs code inside vs code i emailed you yesterday i um i sent you some information on the core path um i just have a long to do list of tutorials uh maybe i'll get to it one day for for corporate from different countries but look at that file that i sent you that that's on twitter that should be that should be helpful all right everybody any other questions any other last comments um any clarifications anything else you want to know um from for me about dash about plotly about getting started you're welcome afaro i hope it helps it's it's very comprehensive um a lot of information but i think it would help you create create um different corporate maps on different uh in different countries it shows you how to use geo json files thank you very much i appreciate i am glad it was very helpful um yeah like i said next next week i hope to do i hope to do a video on python anywhere and how to deploy your app on python anywhere so we'll try and do try and do that if i have the time jill gill um i did not do a video on integrating dash app to existing flask app but maybe in the future i'll do it sergio um sergio what do you mean what what kind of in-depth subjects with graph animation you just you just want to see graph animation on different on different graphs like i did one on on scatter plot and i think i did on on a bar graph you want to see different types of animation i did this in the beginning of the last year so you can see the beginning of the videos my playlist i have some animations um yeah it's me i what i do beside beside videos um i play with my cat i have dinner and conversation with my wife and i do youth development so i work in the youth development field and i help children find mentors i build matches between children and mentors and tutors that take them out that show them different paths in life and different things they can do and are there to support them so i i managed one of the departments inside that organization that helps youth um find mentors in life and um and become successful become become you know successful adults feel they have somebody there for them um davis said have a great week don't know how to do it i'm trying to keep uh i would like to see a waterfall gantt chart solution f oh yeah the gantt chart okay i'll add it to thank you davis for the suggestion i'll add it to my uh my to-do list animated bargain which the text i i position for each bar keeps falling from the left side of the screen oh that's weird yeah so feel free to send send me my um your code to i'll try and take a look at it sometimes at night i have time um i'll see if i can help but it's um charmingdata.org um but if not always you can ask on plotly forum and then somebody will probably answer you quickly than i can but i'll try and help you um no leonardo i haven't used uh altair or or streamlit they are my um my worst enemy um to that no i'm just kidding uh they i heard they're very good i heard i don't know all the air that well i heard streamlit is very good um it creates grass and interactivity um very very quickly very fast from what i understand i looked into it a little bit um it doesn't have as many components or as much flexibility and customization opportunity that dash gives you so you we streamlit you can start off very quickly and and then very powerfully at the very beginning but if you want to customize your graph change your page and do all that i think dash still holds an advantage there um so that's what i know but i don't want to say anything bad about stream stream live because i really don't know it enough that's what i that's what i heard and i saw from the very beginning um yeah i yeah that's that's that's what i hear people saying that streamlit is is very simple to build apps but do you do you know how if it has the capabilities that dash has or plotly when creating the number of graphs and and how you can customize it and what you can do with it um when you want to when you want to um really change your dashboard around and do other things is it as comprehensive as as dash maybe i should look more streamlined i know it's good so please you were saying jill showcase how we would work deploy with databases um i don't know i'll have to see um jill or guild24 if that would be ideal that would be a very good tutorial but it would it takes a long time to do tutorial um on on [Music] on how to deploy on uh your app to the web and so i only have time to work on at nights and on the weekends because i have a a nine-to-five job with youth development and helping new york city children so if i have the time i'll add a database to it but it's it's not likely it's i don't think i'll have the time for that maybe i'll do a different um tutorial on that in the future nice davis um and had a app game oh nice app game to in helping to develop children to make a good decision that sounds pretty interesting actually davis that app really nice good for you thank you for helping the the community davis and the children's and the elders are so so important did not say a lot just just a period so i hope i hope to read more from you thank you jill thank you okay of it that's good to know streamlight um to build dashboard is very fast that's good to know i'll i'll i should take a look at it although you prefer dash you because it provides ease to integrate the process with front end using the django flask yeah i was wondering about stream leaders you can actually it's easy to upload is it easy to upload to the web is it is it is it complicated is it fast um how can you actually upload it to the web a streamlit um dashboard do you know ale or artharva have you ever have you ever uploaded a streamlet um dashboard to the web um no camilo i think i wonder what other people say to camilo i really just know python very well so i'm going to stick to python but um let camilla know if you're if you're thinking of using dash for other languages does anybody use dash for r does anybody use dash for julia i only use it for python because i only know python let me write this i i prefer plotly it just has more functions than book in my opinion young sean have you uploaded the stream dashboard to the web yeah i agree python seems for me when i started learning programming python seemed a bit easier than r so i just i stuck with python you're welcome daniel no i have no idea how to create a couple of parallel graphs with cross filter between them i don't i don't actually know what you mean exactly daniel um but feel free to write my write me to my email um and and i can try and take a look and see how it can help pair a little grass with cross filter if it just grasps with filter between the graphs i think it's fairly fairly doable you can just write me all right everybody so it's been a while i'm gonna have my i'm gonna go get my breakfast eat some call my dad call my grandparents have some family catch up time um you can use flow davis says and build a flow of your process and it can convert your floor to more different languages to create sonic oh interesting for sure here you're welcome avail but they have an easy way of deploying steps but they have an easy plus you can use streamlight you can have an easy way of deploying and you can use different plotting libraries in the same streamlet app nice that's a big advantage that's very good yeah thank you i'll take a look at it i just i don't know if i if i should like how would everybody feel everybody that's right now on this how many people do we have here um where is it it's about 21. how my channel is dedicated to doing data visualization so far in dash if i switched it up a little bit and also added some tutorials on on doing data optimization with streamlit would that be something that would confuse people too much will that take away from the specialty of or the the niche were you just focusing on dash what would you feel your recommendations are about using charming data to create also tutorials on streamlink goodnight it's me from india have a good have a good yeah i'm glad i bet it won't change your love for um dashing plotly thank you camilo so okay everybody so i am going to oh no thank you for the recommendation and uh davis you find dash to be very eloquent not been afraid of screening but it's easier it's where the you can never you will never be stuck in this industry yeah you're right i agree davis yeah maybe i should do a tutorial one two one streamline and see see what people think because it does seem to be it's easy oh you would like to import your class in spanish have a good one okay guys so i'm going to take off i'm going to um have some breakfast um thank you for watching thank you for um uh subscribing and joining my channel and just following along um i really enjoy talking to you and uh i hope you have um i hope you have a good day espanol america latino again you

Channel: Charming Data

Views: 3,130

Rating: 5 out of 5

Keywords: heatmap, heatmap with plotly, heatmap tutorial, heat map, heatmap python, how to a heatmap, how to graph a heatmap, bar chart, bar graph, plotly, matplotlib, python, data analytics, how to make a bar chart, plotly dash, plotly python tutorial, plotly tutorial, gráfico de barras, the bar chart, code bar chart in python, python matplotlib, python tutorial, data analysis, data visualization python, python data science, dash tutorial, bar chart with plotly, plotly bar chart

Id: RgRwsKjkJnU

Length: 70min 9sec (4209 seconds)

Published: Sun Dec 13 2020