couldn't stop watching your videos. interesting and very much helpfull. really helpfull for someone like me who are looking to switch career to data science. keep up the good work.
thanks Satyajit this is really great insight. Can you please give some guidance I am currently working in sales in healthcare industry. how can i prepare my self for the maximum result. I am having a 4+ Year of experience in this field. Thanks. Saurabh
thank you for this wonderful session sir, one request can you please explain an end to end data analysis project (start to end) , just to feel and understand real life companies work , thank you in advance
This is a great video. I wanted to ask on line 85 whether the predictor is something we define or is defined within python? Naturally I would like it to iterate through all the columns. An error messages arises with "name 'predictor' is not defined." I am not sure what to do next. It would be laborious to do one at a time. I'd appreciate your help. Thanks.
i and predictor are something that the Python interpreter understands. Is that line giving you an error? If yes, can u paste the error here, or else ask it on stackoverflow, and give me the link
@@SatyajitPattnaik also sir i am from mechanical background and i do ansys analysis. Can u pls suggest some use case where I can project as data analyst. Bcus i want to utilise my current job experience and move into data analyst field
Auto increment columns can be blindly dropped, apart from that, we need to see if there's any column with 35-40%+ null values, if they have any dependency with other columns, then we analyze, else we drop them.. So there's various steps we follow in EDA
@@SatyajitPattnaik okay Thank you!!! but when we find correlation then its difficult to analyze the columns if we have 200+ columns. how to deal with that, sir?
after running < base_data['NoShow'] = np.where(base_data.NoShow == 'Yes',1,0) > and < base_data.NoShow.value_counts()> it is showing me 0 110527. It has summed up both 1 and 0 values together. how do i fix this?
Want me to work on some projects on your domain?
Leave a comment and I will come back with a video soon ❤️❤️
Predictive maintainence and preventive maintainence in manufacturing sector
Yes, If possible please upload an EDA on Construction Industry.
Thank you so much sir I was searching whole net for health care projects thank you so much
The way you teach sir is honestly the best kind of teaching. I got so much to learn from you... Thank you ❤
Thanks 🥹
very nicely explained, in a simple language. thanks a lot.
this video is very helpful to me in my project🥰🥰🥰
couldn't stop watching your videos. interesting and very much helpfull. really helpfull for someone like me who are looking to switch career to data science. keep up the good work.
I request you to make end to end data analysis project.
thanks Karthik
@@karthiksr8990 have u done a medical degree like mbbs so now you wanna add this skill to it ❓❓❓❓❓
PATTNAIK JI THANK YOU again.
thanks Satyajit
this is really great insight.
Can you please give some guidance I am currently working in sales in healthcare industry.
how can i prepare my self for the maximum result.
I am having a 4+ Year of experience in this field.
Thanks.
Saurabh
thank you for this wonderful session sir, one request can you please explain an end to end data analysis project (start to end) , just to feel and understand real life companies work , thank you in advance
Sure 👍
Sir make a video on which data cleaning is required like to use mean, median,fillna,etc
Hi Satyajit, your video is really helpful and informative. Can you make a video on handling imbalanced data using sampling techniques
Thank You so much @Satyajit Sir.
Keep upload more n more.. in healthcare.
It's really a clear understandable video.
Sure Nishi
Hi sir, This video is suitable for the presentation of the Medical data science project in an interview??
yes ofcourse
@@SatyajitPattnaik Thank you so much for your quick response 🙏
What is Scholarship meant to be in dataset??
Plz make a video on Cervical cancer dataset
Pls send me the dataset link here..
hi satyajit , if u don't mine can u have teach the eda on pharmaceutical industry project.
Can anyone please explain me where is AL part starts ? I didn't understood.
Please
Where can we find this dataset?
Check the video description 😀
For plotting of missing values, im getting an error "TypeError: pointplot() got multiple values for argument 'data'". How to proceed?
This is a great video. I wanted to ask on line 85 whether the predictor is something we define or is defined within python? Naturally I would like it to iterate through all the columns. An error messages arises with "name 'predictor' is not defined." I am not sure what to do next. It would be laborious to do one at a time. I'd appreciate your help. Thanks.
i and predictor are something that the Python interpreter understands. Is that line giving you an error? If yes, can u paste the error here, or else ask it on stackoverflow, and give me the link
if we can use tableau for EDA ,then why we need to use python/sql?? i am confused.
We can do 1 task in multiple tools, depends on what tool your company is using
@@SatyajitPattnaik also sir i am from mechanical background and i do ansys analysis. Can u pls suggest some use case where I can project as data analyst. Bcus i want to utilise my current job experience and move into data analyst field
how can we convert diff files into csv
Can you provide pdf of this project?
Which machine learning algorithm is used
EDA is exploratory data analysis, it doesnt involve ML
How can we contact you??
I already have my details in the description of all my videos
Hello Sir, Make a video on EDA (which contain 30k rows or more and 200 features(columns)) .Please please
Credit card EDA is one of such examples, I will make a video on it soon..
@@SatyajitPattnaik yes waiting, post it soon!!! Thank you so much!!
How to decide which column is unnecessary.
Auto increment columns can be blindly dropped, apart from that, we need to see if there's any column with 35-40%+ null values, if they have any dependency with other columns, then we analyze, else we drop them..
So there's various steps we follow in EDA
@@SatyajitPattnaik okay Thank you!!! but when we find correlation then its difficult to analyze the columns if we have 200+ columns. how to deal with that, sir?
@@NehaYadav-hs1po There's no other way than creating some logic to have absolute values of correlation which is above a certain value
@@NehaYadav-hs1po for ex: stackoverflow.com/questions/29294983/how-to-calculate-correlation-between-all-columns-and-remove-highly-correlated-on
@@SatyajitPattnaik Thank you !!!!!!!
after running < base_data['NoShow'] = np.where(base_data.NoShow == 'Yes',1,0) > and < base_data.NoShow.value_counts()> it is showing me 0 110527. It has summed up both 1 and 0 values together. how do i fix this?