How to Build Your Own AutoML App | Streamlit #17

[@DataProfessor]

👉Watch Part 1 of the Streamlit tutorial series kzbin.info/www/bejne/kIuXc2OHiq2ohMU ** Support this Channel If you like this video, please make sure to LIKE, SUBSCRIBE, COMMENT and SHARE! 🌟 Buy me a coffee www.buymeacoffee.com/dataprofessor 🌟 Download Kite for FREE www.kite.com/get-kite/? 👉 Subscribe to this KZbin channel kzbin.info 👉 Join the Newsletter of Data Professor newsletter.dataprofessor.org

[@imadata8043]

This is sick!

[@kareembelle8947]

Just found your channel today and I have to say I’m already a fan. Great work!

[@DataProfessor]

Welcome aboard! 😊

[@luciferkhusrao]

Absolutely Lovely. Even I had completed making exactly a similar thing and had completed on 30th Jan 2021. But, I must say, your work is far more complete and concise and I would love to follow your application only. Thanks a lot for sharing.

[@DataProfessor]

Wonderful! You just gave me an idea for a new series on the channel. Would you be interested in submitting your projects and if it is selected, I will do a walkthrough of the project in a video.

[@luciferkhusrao]

@@DataProfessor definitely sir. I will just put it up on github and share you the link. Thanks a lot. 🙂

[@praneethkumar4791]

Really exciting when u your introducing a new things for us...love this channel so much❤️❤️❤️

[@DataProfessor]

Thanks for the kind words!

[@bernard2735]

I am really enjoying your channel - it has a great mix of new libraries and simple explanations. Many thanks.

[@DataProfessor]

Thanks for the kind words, glad the contents are helpful!

[@notknown9307]

Nice to see you again

[@DataProfessor]

😁

[@landonwallaz]

This is absolutely incredible. fantastic work.

[@DataProfessor]

Thank you 😊

[@CristiVladZ]

really insightful! thanks and keep em up!

[@DataProfessor]

Thanks!

[@jahudini]

I appreciate you Boss! Providing us content and making our job easier as DS. Thank🙌 you Sir!

[@DataProfessor]

Wow, thanks for the encouragement! 😆

[@madhavanjn]

Very useful video professor, keep posting more nd more videos 😀

[@DataProfessor]

Thanks for watching

[@madhavanjn]

@@DataProfessor plz post bioinformatics related videos, it's very useful for us, huge biological data available in net but we don't know how to use this data for drug discovery.

[@janelle3183]

thank u for the great tutorials! just wondering, is there a way to use st.multiselect to display different types of graphs using the same dataframe?

[@Julio_Zambrano]

Amazing and inspiring! Thank you Dr!

[@DataProfessor]

My pleasure!

[@idanmorad4769]

Great video. You should know that in order for the app to work the git repository should be cloned and not just download (wget) the requirements file.

[@DataProfessor]

Thanks for help clarifying 😆

[@vin84iitr]

Thanks for this wonderful content. However, not sure to handle categorical data with lazypredict!!

[@user-or7ji5hv8y]

What are the downsides of using such solutions? Do they work well? Just seem so easy then. Can you discuss in future.

[@madhu1987ful]

not able to run the requirements.txt file, throwing lot of errors even though i followed all the steps right from creating new env on conda

[@hassanbelarbi5185]

Thank you so much ,very helpful content I appreciate your efforts

[@DataProfessor]

Glad it was helpful!

[@akshitagarwal1309]

This is HUGE!! Thanks!

[@DataProfessor]

Thanks for watching!

[@salikmalik7631]

Data Professor, Is better to do a sentiment analysis project as a beginner? Need suggestion.

[@DataProfessor]

Although tabular datasets may be a bit trickier than tabular datasets, I certainly think that this is doable. There are several good datasets on Kaggle and I would recommend to check out the corresponding Notebooks and see how others are implementing their workflow to analyze the data that you could use to jumpstart your own project.

[@salikmalik7631]

@@DataProfessor I am thinking after do a project on sentiment analysis I will deployed on streamlit by the help of your streamlit playlist. Is'nt good Idea sir?

[@DataProfessor]

@@salikmalik7631 Great idea! My friend Jaemin has made a couple sentiment analysis tutorial video, check it out at his channel kzbin.info/www/bejne/epi3eGWHnJKrZs0

[@salikmalik7631]

@@DataProfessor ohh, Thanks alot. I was searching for that.

[@divyakarade4378]

That's again a very useful video...Thanks :)

[@DataProfessor]

A pleasure, thanks for watching!

[@myreisparstory3906]

Please how did you create the app.py? Did you just run "streamlit app.py" ?

[@DataProfessor]

Hi app.py is a file containing the code. To run app.py we use the command "streamlit run app.py".

[@myreisparstory3906]

@@DataProfessor thanks, I'm trying to run now with ngrok it's saying the connection was successfully tunneled to your ngrok client but the client failed to establish a connection to the local address...

[@DataProfessor]

@@myreisparstory3906 Hi, I'm guessing you're using ngrok tunneling to run the web app from within Colab? I haven't tried that though, but I recommend to watch Jesse's video on how to do that kzbin.info/www/bejne/rmGxlY2hlsicrdU

[@myreisparstory3906]

@@DataProfessor Thanks Data Prof

[@sunritjana4573]

This video is awesome!! I was also writing the code from scratch as per the logic taught in video, And I'm experiencing some issues. Code section: ax1 = sns.barplot(y=predictions_test.index, x="R-Squared", data=predictions_test) Error: ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all(). Why is it being caused, and How do I fix it?

[@DataProfessor]

Hi, I would recommend to check Stack Overflow for prior solutions. This is what I found, stackoverflow.com/questions/36921951/truth-value-of-a-series-is-ambiguous-use-a-empty-a-bool-a-item-a-any-o

[@sunritjana4573]

@@DataProfessor I did check, but I was unable to get any answers, because I don't have any comparison operator. Moreover I got another error, that tuple model is erroneous, which is weird because I followed same steps

[@sunritjana4573]

This is the exact warning `tuple model failed to execute`

[@DataProfessor]

Hi, I would recommend to try by trial and error, cutting out blocks of code from the streamlit app and try to get it to run. For example, try cutting out the block of code that is responsible for creating the plot, then see if the app runs, if it does then it means the plot making code has the bug that will need to be debug.Then slowly add the plot code into your app line by line while testing to see if the app is working. I would typically do this when debugging the code.

[@foundersgurukul]

Hey I am also facing the same error, what are the steps u took to debug it... If u can share the code link..

[@techworld3043]

amazing content

[@DataProfessor]

Thanks!

[@yosolonopuedo]

Nice! Will you cover Bokeh+Panel on any of your videos?

[@DataProfessor]

That's the plan! Thanks for the suggestion!

[@yosolonopuedo]

@@DataProfessor awesome!! Looking forward to it!!

[@siddhigolatkar8558]

Awesome..

[@DataProfessor]

Thanks for watching!

[@Sawaedo]

Thank you!

[@DataProfessor]

A pleasure, thanks for watching!

[@shubhamnaik9246]

love the video absolutely amazing but i am getting an error after running the app everything works fine till rendering the train daset but after that test dataset doest load and the i get the following error The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all(). can anyone please help

[@DataProfessor]

Hi, is this error occuring for an uploaded CSV dataset? or from the example dataset (where you click on the "Press to use Example dataset" button).

[@shubhamnaik9246]

@@DataProfessor first of all wow. i was not expecting this quick reply. thanks a lot for that. and no I tested it with uploading my own dataset as well as the example dataset button same problem. been through the code with fine tooth comb still cant find whats the source. any pointers i can get from you will be helpful. Thanks All the algorithms are running just fine on the training dataset the error comes on running the algos on the test dataset and the rest part that is plots etc follows since it requires the dataframe thats been created as output for running algorithms on the test set

[@DataProfessor]

@@shubhamnaik9246 If running the example dataset works but when using your own dataset does not work, then I would recommend to inspect your dataset to see if it is read in properly, is there missing data, etc.

[@shubhamnaik9246]

@@DataProfessor yeah i checked running on example dataset doesnt wok either. keep getting the same error. If you can check the SS of the error code i have attached to the issue i raised on the github repo.

[@aeigreen]

Can we please have data base tutorial sql/nosql with streamlit.

[@DataProfessor]

Thanks for the suggestion

[@salikmalik7631]

That's great.

[@wildrifter2583]

Nc video! As always.

[@DataProfessor]

Glad you enjoyed!

[@username42]

*Argh.* This app has gone over its resource limits. Please try again in a few minutes.

[@DataProfessor]

Yes the app is resource hungry 😅

[@prasuna530]

Hello, is this tutorial still valid? I am not able to run the app. Here is the message I am getting. Is anyone have the same issue and got resolved? Traceback (most recent call last): File "C:\kaiser\anaconda\envs\lazypredict\lib unpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "C:\kaiser\anaconda\envs\lazypredict\lib unpy.py", line 85, in _run_code exec(code, run_globals) File "C:\kaiser\anaconda\envs\lazypredict\Scripts\streamlit.exe\__main__.py", line 7, in File "C:\kaiser\anaconda\envs\lazypredict\lib\site-packages\click\core.py", line 1130, in __call__ return self.main(*args, **kwargs) File "C:\kaiser\anaconda\envs\lazypredict\lib\site-packages\click\core.py", line 1055, in main rv = self.invoke(ctx) File "C:\kaiser\anaconda\envs\lazypredict\lib\site-packages\click\core.py", line 1657, in invoke return _process_result(sub_ctx.command.invoke(sub_ctx)) File "C:\kaiser\anaconda\envs\lazypredict\lib\site-packages\click\core.py", line 1404, in invoke return ctx.invoke(self.callback, **ctx.params) File "C:\kaiser\anaconda\envs\lazypredict\lib\site-packages\click\core.py", line 760, in invoke return __callback(*args, **kwargs) File "C:\kaiser\anaconda\envs\lazypredict\lib\site-packages\streamlit\cli.py", line 233, in main_run _main_run(target, args) File "C:\kaiser\anaconda\envs\lazypredict\lib\site-packages\streamlit\cli.py", line 249, in _main_run command_line = _get_command_line_as_string() File "C:\kaiser\anaconda\envs\lazypredict\lib\site-packages\streamlit\cli.py", line 244, in _get_command_line_as_string cmd_line_as_list.extend(click.get_os_args()) AttributeError: module 'click' has no attribute 'get_os_args'

[@DataProfessor]

Hi, can you try to update click to version 8

[@prasuna530]

@@DataProfessor I updated Streamlit to latest version and it worked. I will update Click too. Thank you.