Hi Sir, On the whole UA-cam you are the best teacher mentor I have ever seen with no cost, I request you to please come with dash basic structure videos
Thank you sir, I used dash and plotly to make a dashboard as an internship selection test. I did not get selected, but I updated and made it better and got an internship with a new dashboard. All thanks to your tutorials, I was in a hurry initially, but now, I have decided to watch every video of yours. Suggestion: The python file in the end was around 800 lines. I found ways to break the whole file into different files by callbacks, layouts, others utility functions etc. That was much easier to navigate.
Not sure why, but for some reason simple and effective Dash tutorials are super hard to come by on the internet and your videos make it super simple and easy to follow. Thank you! If there's an opportunity to request certain topics on Dash, I'd love to provide feedback!
Hi Hon luu, I have several Dash tutorials planned for the upcoming months (pattern matching callbacks, Dash with postgreSQL, Bootstrap, etc.). But if you have specific topics you think many people are interested in, I would love to hear your feedback. Thanks for watching
Charming Data for me, id love to see a tutorial on 1) creating a dashboard that lets an end user upload any dataset and let them explore the data via the dashboard, 2) custom function and aggregation methods and 3)dashboard for a predictive model (I.e clustering with sliders for parameters).
@@semidevilz Thank you for the suggestions Hon. These are very relevant. I appreciate it. Can you tell more a little more about suggestion number 1, please. Are you envisioning an app that allows users to upload their CSV/Excel file so that a datatable is created; and then, they explore the datatable with pre-built graphs?
Charming Data there’s definitely a lot of ways to think about it. I imagine like a homemade version of powerbi or tableau where the user would upload a cleaned csv file and all combinations of scatter charts would populate, along with summaries etc etc. but I think in general, I’m excited to make It where dash can interact with all datasets types , and not just the ones that you predefine. It’s definitely a bit of reinventing the wheel knowing that we have powerBI, but I think the satisfaction is much higher and you get control to every piece of the dashboarding
All your tutorials are so well made! I learned so much from them and thanks to you I made through my first quite complicated Dash web app for a client. Thanks a lot for your work! :-)
Really enjoy your videos on Dash Plotly charts, diagrams. Could you please also make a tutorial on histogram with dropdown for aggregation function choices(a drop down where you can choose a certain aggregation function such as sum, count, mean)? I'm stuck with a project for creating a dropdown for histogram with Plotly Dash:( And I can't find a good example online, the official website does have an example but when I supply my x and y values in my dataframe with the code provided, it's complaining about "dictionary update sequence element #0 has length 5; 2 is required"
Hi @junyi i have very long list of tutorials to do. But feel free to connect with me on Patreon if you'd like to discuss something as specific as that histogram with a Dropdown. www.patreon.com/charmingdata
@@CharmingData Ohh Thanks for the compliment Adam... It's my niece and my kid in the profile photo... Used your above tutorial for data analysis for one of my project... Output was exactly what I was hunting for... Heartly thanks again... Take care God Bless
Thanks so much for making this video. I really appreciate you taking the time to lay everything out step-by-step; it's definitely helping me understand Dash and reinforce good habits. I do have a question concerning the layout for the x-axis: I copied your code exactly, including maintaining the update_layout() with your selected parameters ({‘categoryorder’: ‘total ascending’}). The problem is that on my app, the x-axis, when 'Month Call Made' is selected, isn't ordered chronologically. I've tried changing the value for categoryorder and checking the docs-while choosing different values does change the order slightly, nothing I've tried successfully sets the months to ascend along the x-axis, beginning with January. Any ideas for how to fix this? Any tips are appreciated!
@@CharmingData Hi CharmingData, thanks for getting back to me! So the values aren’t different-they seem to be appropriately paired with their corresponding months. Just to run you through a more specific example of what I’m seeing on mine vs. yours: @ 16:25 when you select x-axis seeing all month calls made calls (y-axis set to time spent), the values themselves are accurate in my app (August having the most time spent, January the least). The issue is that on my app, it's not ordering the months in chronological order along the x-axis (i.e. Jan(1), Feb (2),....Dec(12)), but instead ordering the x-axis to ascend by the value of time spent (i.e. Jan(1), May (5), ....Aug (8)). So it seems clear to me that in my app, the update_layout x-axis categoryorder arg is *working*, but it's working on the actual values (in this case, time spent), instead of acknowledging the month order. Just to double check myself (in case I may have changed a line while following along), I copied your code again from the github repo into a new file, ran it, and it’s still having this issue. The only difference I can detect is that I’m running different versions of the modules on my machine: Pandas=1.0.3 Dash=1.20.0 Plotly=4.14.3 Could that be a factor..?
@@s.d.fogarty6221 Ummmm... that should still work. I'm not sure. It could be because of the different versions but that would need to be tested. I recommend testing with other versions. If that doesn't solve the issue, try posting a reproducible example on Plotly forum. I'm sure you'll get help there.
Wonder if you've solved this problem or not . If you haven't , try to change " total ascending " into " category ascending ".Hope you work out alright . (Hope I didn't misunderstand your description of that problem .) Code : xaxis={"categoryorder":"category ascending"} And thank you so much for all those videos you made, Mr. Charming Data. I am fascinated when I first saw your video . The good thing about you is that you encourage us to read documentation of different modules and functions , not only to repeat what you teach in the videos . Thank you so much !
Hello Charming . In a bargrah like this, that the rectangles are stacked on top of one another, how can you put the total y-axis value on the top of each bar as a text? Thank you!
Hi @Uni 13 Yes, that's possible. You should be able to do that like I do it in the "All about the Bar Chart" video tutorial ua-cam.com/video/N1GwQNatOwo/v-deo.html Add the text parameter inside the barchart=px.bar(text='y_axis') . Then below, you need to update the layout: barchart.update_layout(uniformtext_minsize=14, uniformtext_mode='hide')
@@CharmingData Thanks for the answer. I tried this and i get one text per rectangle, which means i have many texts in every bar. What i want is only the total count of those mini rectangles , at the top of each bar.
@@uni1357 I don't think that's possible because the stacked bars are traces of different bars stacked together, and every bar inside the stack has it's own number. If you just want the total of each stacked bar, I think you can re-sort/group the pandas dataframe, so they are grouped into one trace per bar (x-axis) before plotting them.
Wow, this tutorial is wonderful and detailed. I would like to see dash applications on machine learning and artificial intelligence I am a newbie in dash and would love to connect with you. Any chance i can get your email or preferred mode of communication. P.s I have subscribed to your content. Very good stuff.
Hi Sir, On the whole UA-cam you are the best teacher mentor I have ever seen with no cost, I request you to please come with dash basic structure videos
Thank you sir, I used dash and plotly to make a dashboard as an internship selection test. I did not get selected, but I updated and made it better and got an internship with a new dashboard. All thanks to your tutorials, I was in a hurry initially, but now, I have decided to watch every video of yours.
Suggestion: The python file in the end was around 800 lines. I found ways to break the whole file into different files by callbacks, layouts, others utility functions etc. That was much easier to navigate.
Congrats on your new internship, @john
@@CharmingData Thanks to you sir.
Not sure why, but for some reason simple and effective Dash tutorials are super hard to come by on the internet and your videos make it super simple and easy to follow. Thank you! If there's an opportunity to request certain topics on Dash, I'd love to provide feedback!
Hi Hon luu, I have several Dash tutorials planned for the upcoming months (pattern matching callbacks, Dash with postgreSQL, Bootstrap, etc.).
But if you have specific topics you think many people are interested in, I would love to hear your feedback.
Thanks for watching
Charming Data for me, id love to see a tutorial on 1) creating a dashboard that lets an end user upload any dataset and let them explore the data via the dashboard, 2) custom function and aggregation methods and 3)dashboard for a predictive model (I.e clustering with sliders for parameters).
@@semidevilz Thank you for the suggestions Hon. These are very relevant. I appreciate it. Can you tell more a little more about suggestion number 1, please.
Are you envisioning an app that allows users to upload their CSV/Excel file so that a datatable is created; and then, they explore the datatable with pre-built graphs?
Charming Data there’s definitely a lot of ways to think about it. I imagine like a homemade version of powerbi or tableau where the user would upload a cleaned csv file and all combinations of scatter charts would populate, along with summaries etc etc. but I think in general, I’m excited to make It where dash can interact with all datasets types , and not just the ones that you predefine. It’s definitely a bit of reinventing the wheel knowing that we have powerBI, but I think the satisfaction is much higher and you get control to every piece of the dashboarding
But what you mentioned is also something I had in mind
All your tutorials are so well made! I learned so much from them and thanks to you I made through my first quite complicated Dash web app for a client. Thanks a lot for your work! :-)
Really enjoy your videos on Dash Plotly charts, diagrams.
Could you please also make a tutorial on histogram with dropdown for aggregation function choices(a drop down where you can choose a certain aggregation function such as sum, count, mean)?
I'm stuck with a project for creating a dropdown for histogram with Plotly Dash:(
And I can't find a good example online, the official website does have an example but when I supply my x and y values in my dataframe with the code provided, it's complaining about
"dictionary update sequence element #0 has length 5; 2 is required"
Hi @junyi i have very long list of tutorials to do. But feel free to connect with me on Patreon if you'd like to discuss something as specific as that histogram with a Dropdown.
www.patreon.com/charmingdata
Your explanation on plotly and dash is simply the best... Keep on going... @charmingdata
Love your profile picture, Azhar 💙
@@CharmingData Ohh Thanks for the compliment Adam...
It's my niece and my kid in the profile photo...
Used your above tutorial for data analysis for one of my project...
Output was exactly what I was hunting for...
Heartly thanks again...
Take care God Bless
this man is amazing.
Thanks so much for making this video. I really appreciate you taking the time to lay everything out step-by-step; it's definitely helping me understand Dash and reinforce good habits. I do have a question concerning the layout for the x-axis: I copied your code exactly, including maintaining the update_layout() with your selected parameters ({‘categoryorder’: ‘total ascending’}). The problem is that on my app, the x-axis, when 'Month Call Made' is selected, isn't ordered chronologically. I've tried changing the value for categoryorder and checking the docs-while choosing different values does change the order slightly, nothing I've tried successfully sets the months to ascend along the x-axis, beginning with January. Any ideas for how to fix this? Any tips are appreciated!
Hi S.D. are the month values in your app different than the ones in my app?
@@CharmingData Hi CharmingData, thanks for getting back to me!
So the values aren’t different-they seem to be appropriately paired with their corresponding months. Just to run you through a more specific example of what I’m seeing on mine vs. yours:
@ 16:25 when you select x-axis seeing all month calls made calls (y-axis set to time spent), the values themselves are accurate in my app (August having the most time spent, January the least). The issue is that on my app, it's not ordering the months in chronological order along the x-axis (i.e. Jan(1), Feb (2),....Dec(12)), but instead ordering the x-axis to ascend by the value of time spent (i.e. Jan(1), May (5), ....Aug (8)). So it seems clear to me that in my app, the update_layout x-axis categoryorder arg is *working*, but it's working on the actual values (in this case, time spent), instead of acknowledging the month order.
Just to double check myself (in case I may have changed a line while following along), I copied your code again from the github repo into a new file, ran it, and it’s still having this issue.
The only difference I can detect is that I’m running different versions of the modules on my machine:
Pandas=1.0.3
Dash=1.20.0
Plotly=4.14.3
Could that be a factor..?
@@s.d.fogarty6221 Ummmm... that should still work. I'm not sure. It could be because of the different versions but that would need to be tested. I recommend testing with other versions. If that doesn't solve the issue, try posting a reproducible example on Plotly forum. I'm sure you'll get help there.
Wonder if you've solved this problem or not . If you haven't , try to change " total ascending " into " category ascending ".Hope you work out alright .
(Hope I didn't misunderstand your description of that problem .)
Code : xaxis={"categoryorder":"category ascending"}
And thank you so much for all those videos you made, Mr. Charming Data. I am fascinated when I first saw your video . The good thing about you is that you encourage us to read documentation of different modules and functions , not only to repeat what you teach in the videos .
Thank you so much !
I'm really enjoying your videos and learning a lot from them, could you make video about dcc upload and to plot graphs with
Thanks @Shafiq i hope to see you in future videos and live sessions
Hello Charming . In a bargrah like this, that the rectangles are stacked on top of one another, how can you put the total y-axis value on the top of each bar as a text? Thank you!
Hi @Uni 13 Yes, that's possible. You should be able to do that like I do it in the "All about the Bar Chart" video tutorial ua-cam.com/video/N1GwQNatOwo/v-deo.html
Add the text parameter inside the barchart=px.bar(text='y_axis') . Then below, you need to update the layout: barchart.update_layout(uniformtext_minsize=14, uniformtext_mode='hide')
@@CharmingData Thanks for the answer. I tried this and i get one text per rectangle, which means i have many texts in every bar. What i want is only the total count of those mini rectangles , at the top of each bar.
@@uni1357 I don't think that's possible because the stacked bars are traces of different bars stacked together, and every bar inside the stack has it's own number. If you just want the total of each stacked bar, I think you can re-sort/group the pandas dataframe, so they are grouped into one trace per bar (x-axis) before plotting them.
Hi
Hope you r doing well
I need one help in plotly but unable to solve it it could be grateful if any have good experience for the same
Wow, this tutorial is wonderful and detailed.
I would like to see dash applications on machine learning and artificial intelligence
I am a newbie in dash and would love to connect with you.
Any chance i can get your email or preferred mode of communication.
P.s I have subscribed to your content. Very good stuff.
of course. I'm on plotly forum. Just look for Charming Data and send me a message. community.plotly.com/
0 dislike 🙌😎