Join this channel membership to get access to materials: ruclips.net/channel/UCNU_lfiiWBdtULKOw6X0Digjoin check out the complete end to end project playlist ruclips.net/p/PLZoTAELRMXVPS-dOaVbAux22vzqdgoGhG
krish sb, i just saw microsoft 365 ai, it was able to do simple data analysis tasks eg:when user typed an english sentence "find subset of data w/highest company revenue." I hope a time doesnt come when programmers have to stage a mass gilet jaune styled protest against ai engineers for taking away jobs. My nightmare is, imagine one only needs to type in english that "I need a program that can calculate the best tax rate" and an AI just creates it for you. Meaning programmers become worthless just like google translate has meant human translators arent needed. 😭 Seems like creativity,human emotions are the only things that even if an AI manages to show humans wont care. Like who wud care to watch movies chruned out by an AI.
Thanks Krish! As always, very helpful videos. I just have one request - can you please upload a flow diagram which shows all these various python files, classes and function are calling each other when train or predict gets triggered? This will be very helpful to understand how all these things actually working.
Thank you for the amazing efforts you put in to teach us Krish. I've gained confidence with respect to end-to-end projects just because of your videos. Also would like to highlight a small error in the HTML file: name="writing_score" is assigned to Reading score label and vice versa. This can result a slightly inaccurate predictions based on user inputs.
@Krishnaik06 Thank you for sharing such a detailed series of videos, it have taught me a lot. I have only one comment on the last video about the prediction pipeline. I think it would be better to read the model and the preprocessor from the pickle file in the predict_pipeline class as a class variable instead of reading those each time you make a prediction. Greatly appreciate you sharing all the good stuff
I have two questions Krish sir: 1. Where do we add the code if there are feature engineering steps involved like merging data from multiple sources, defining a new variable etc.? 2. This feature engineering functionality is also required in the Prediction pipeline. How do we implement the same? Would be grateful to your suggestions.
Great content. Why go through the trouble of making a separate method in the CustomData class? Couldn't it be designed such that on initialization the variables are assigned AND they are placed in a dataframe that is returned. I don't use OOP much, but several times in the series I have had a similar question, which is, why not place more into each method, or all of it into the initialization function? Debugging? Reuse?
Hi Krish sir, I am getting error like IndexError: Replacement index 0 out of range for positional args tuple for data_scaled=preprocessor.transform(features) this line , please let me know where I am going wrong?
Hi Thank you for your amazing series, I have tried the samething but getting the error "[X has 7 features, but ColumnTransformer is expecting 8 features as input.]" how i can encounter, anyone can help?
sir, when we give 100 for both reading score and writing score, sometime the prediction of math score goes above 100 (I got 100.64). so somewhere we need to set the maximum prediction should be 100. right? results[0]=min(results[0],100) return render_template('home.html',results=results[0]) , hope this will work?
@@debojitmandal8670 not yet . Im still trying to make it work in deployed stage . I have tried in both elasticbean and heroku . But the prediction part doesn’t work in both. And works absolutely fine in local system . Im still trying to figure out whats the problem
"When I press Predict", im getting this error: Internal Server Error The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.
Thanks Krish! Your tutorials are amazing, very simple to understand and straightforward. Keep doing a good job! Thanks I keep getting this error "Traceback (most recent call last): File "C:\projects\MLproject_end2end\app.py", line 6, in from src.pipeline.predic_pipeline import CustomData, PredictPipeline File "C:\projects\MLproject_end2end\src\pipeline\predic_pipeline.py", line 4, in from src.utils import load_object ImportError: cannot import name 'load_object' from 'src.utils' (C:\projects\MLproject_end2end\src\utils.py)" Do you have any idea?
@@sahilansari1045 try to go to control panel/ programs/ programs and features there you will get Turn Windows feature on or off , check mark on Internet infromation services and internet information services hostable web core. using this your local machine can be a web server for any local development by you.
hi krish i am getting an error after putting the data on webpage. Internal Server Error The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.
sir i am making a multiple fruit detection app website , i have trained the model using yolov8 and labelled it using labelimg ,i am having problem in getting the detected image on front end of website (with bounding box ,name of fruit and accuracy),while i am getting the predicted image on visual studio code ,where i have the whole code in runs folder and each time we click on predict,it saves it in vs code as pred1,pred2 ... and so on and same for txt file of bounding box class to get the class of fruit,please help
Hello Krish Sir, I'm facing an error which I'm unable to resolve. My preprocessor pickle file has been loaded but it does not transform the data. Because of which it throws an internal server error. Can you help me resolve this please?
@@muhammadfaheem4024 Recheck how the column transformer was build and better approach would be to just open a new jupyter file and load the pickle file there and use it to see the results you are gettingn
src.exception.CustomException: Error in python script name [C:\Users\HP\Documents\GitHub\mlops-01\src\pipeline\predict_pipeline.py] line number [21] error message[Found unknown categories [None] in column 1 during transform] any solutions
The problem comes from the fact that the mapping beetween variables name need to be perfect. In the home.html file you need to replace = "ethnicity" by "race_ethnicity" otherwise you will always get None for the race_ethnicity features. You can check it out by printing the pred_df
Hello sir, thankyou for the help but I have replaced "enthnicity" with "race_enthnicity" in home.html file but still getting the same error. It seems like the error is occuring in column 1 which is "gender". Thankyou again for the help sir! @@medounelearning2025
Sir please me give replay of This message. Sir when I am put gender : male , parental education: master, food: standard , read score :100, writing score : 100 . So why its's give me math score 103.25. the exam marks in between 0-100 so why it gives me 103.25 . Please reply this message. Because of this one error I am trying to found ....please give me reply as soon as possible..
@krishnaik06, i am getting error like "Internal Server Error The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application."
The name of the preprocessor pickle file that you saved is different from the path name that we given in prediction ,in my case due to spelling of the file name different ,after correction of file name ,the error resolved
Can anyone help? I am getting this error message "Internal Server Error The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application." Each time I click on predict math score. Every other things seems okay
@krishnaik06 , i am getting below error after enter values - "Internal Server Error The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application."
Got this error when trying to predict data : Internal Server Error The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application. Anybody know the solution for this error please le t me know
@@lionsinescanor405 it is running but when i am predictig the score after passing the values , getting an error ['OneHotEncoder' object has no attribute '_drop_idx_after_grouping'] Traceback (most recent call last)
src.exception.CustomException: Error occured in python script name [E:\AI project\mlproject\src\pipeline\predict_pipeline.py] line number [16] error message[Error occured in python script name [E:\AI project\mlproject\src\utils.py] line number [59] error message[CustomException.__init__() missing 1 required positional argument: 'error_detail']] how to fix this error
Join this channel membership to get access to materials:
ruclips.net/channel/UCNU_lfiiWBdtULKOw6X0Digjoin
check out the complete end to end project playlist
ruclips.net/p/PLZoTAELRMXVPS-dOaVbAux22vzqdgoGhG
krish sb, i just saw microsoft 365 ai, it was able to do simple data analysis tasks eg:when user typed an english sentence "find subset of data w/highest company revenue."
I hope a time doesnt come when programmers have to stage a mass gilet jaune styled protest against ai engineers for taking away jobs. My nightmare is, imagine one only needs to type in english that "I need a program that can calculate the best tax rate" and an AI just creates it for you. Meaning programmers become worthless just like google translate has meant human translators arent needed. 😭
Seems like creativity,human emotions are the only things that even if an AI manages to show humans wont care. Like who wud care to watch movies chruned out by an AI.
Thanks Krish! As always, very helpful videos. I just have one request - can you please upload a flow diagram which shows all these various python files, classes and function are calling each other when train or predict gets triggered? This will be very helpful to understand how all these things actually working.
exactly we need this
This whole playlist was wonderful. Sir please create one end to end project on deep learning also.
Thank you for the amazing efforts you put in to teach us Krish. I've gained confidence with respect to end-to-end projects just because of your videos.
Also would like to highlight a small error in the HTML file: name="writing_score" is assigned to Reading score label and vice versa. This can result a slightly inaccurate predictions based on user inputs.
@Krishnaik06 Thank you for sharing such a detailed series of videos, it have taught me a lot. I have only one comment on the last video about the prediction pipeline. I think it would be better to read the model and the preprocessor from the pickle file in the predict_pipeline class as a class variable instead of reading those each time you make a prediction.
Greatly appreciate you sharing all the good stuff
amazing playlist,going to add this project in resume as it will be easier for me to explain it in detail
Very very practical and to the point tutorials ... Thanks
One of the best projects I have ever seen. Thank you for this content 😇
Thank You sir for your contents. wonderful contents l have learnt so many things from this playlist.
Please make the next project on deep learning
I have also learned web dev using nextjs and express js. So, I have used postman alot of time
I have two questions Krish sir:
1. Where do we add the code if there are feature engineering steps involved like merging data from multiple sources, defining a new variable etc.?
2. This feature engineering functionality is also required in the Prediction pipeline. How do we implement the same?
Would be grateful to your suggestions.
Love it! Thanks for this type of content. 🙏
Its very amazing Krish sir
Thank you so much sir . Was eagerly waiting for it to drop
Thankyou krish sir for this series
Great content. Why go through the trouble of making a separate method in the CustomData class? Couldn't it be designed such that on initialization the variables are assigned AND they are placed in a dataframe that is returned. I don't use OOP much, but several times in the series I have had a similar question, which is, why not place more into each method, or all of it into the initialization function? Debugging? Reuse?
i got error when press predict button ColumnTransformer' object has no attribute '_name_to_fitted_passthrough' what wrong with it ?
Cool! Thank you
@krishnaik06 sir, for the train_pipeline, can you give an overview on what to do? it's a little confusing. I'd like to work on this.
Thanks a lot for this amazing content
Hi Krish sir,
I am getting error like IndexError: Replacement index 0 out of range for positional args tuple for
data_scaled=preprocessor.transform(features) this line , please let me know where I am going wrong?
Thanks
amazing mind blowing hundred millions dollar
Well Done Sir!!
Next project on deep learning.. plzz sir...
Thank you sir❤
Thank you!
we can use gradio for this
Hi Thank you for your amazing series, I have tried the samething but getting the error "[X has 7 features, but ColumnTransformer is expecting 8 features as input.]" how i can encounter, anyone can help?
sir,
when we give 100 for both reading score and writing score, sometime the prediction of math score goes above 100 (I got 100.64). so somewhere we need to set the maximum prediction should be 100. right?
results[0]=min(results[0],100)
return render_template('home.html',results=results[0]) , hope this will work?
@@debojitmandal8670 not yet . Im still trying to make it work in deployed stage . I have tried in both elasticbean and heroku . But the prediction part doesn’t work in both. And works absolutely fine in local system . Im still trying to figure out whats the problem
"When I press Predict", im getting this error: Internal Server Error
The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.
U got the solution for this I am also facing same problem
@@SuprithKorishettar not yet 🤷🏻♂️
did you get it?@@KokkeOP
you got it ???@@kanangulati8276
I am also facing same error
Thanks Krish! Your tutorials are amazing, very simple to understand and straightforward. Keep doing a good job! Thanks
I keep getting this error "Traceback (most recent call last):
File "C:\projects\MLproject_end2end\app.py", line 6, in
from src.pipeline.predic_pipeline import CustomData, PredictPipeline
File "C:\projects\MLproject_end2end\src\pipeline\predic_pipeline.py", line 4, in
from src.utils import load_object
ImportError: cannot import name 'load_object' from 'src.utils' (C:\projects\MLproject_end2end\src\utils.py)"
Do you have any idea?
@@sahilansari1045 did you get the solution?
@@sahilansari1045 try to go to control panel/ programs/ programs and features there you will get Turn Windows feature on or off , check mark on Internet infromation services and internet information services hostable web core. using this your local machine can be a web server for any local development by you.
hi krish i am getting an error after putting the data on webpage.
Internal Server Error
The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.
Will u be showing mlops?
In function initiate_data_ingestion is where there is training pipeline or, kindly explain
When are you going to start next project?
sir i am making a multiple fruit detection app website , i have trained the model using yolov8 and labelled it using labelimg ,i am having problem in getting the detected image on front end of website (with bounding box ,name of fruit and accuracy),while i am getting the predicted image on visual studio code ,where i have the whole code in runs folder and each time we click on predict,it saves it in vs code as pred1,pred2 ... and so on and same for txt file of bounding box class to get the class of fruit,please help
Sir, what if the no. Of columns are too large???
Hello Krish Sir,
I'm facing an error which I'm unable to resolve. My preprocessor pickle file has been loaded but it does not transform the data. Because of which it throws an internal server error. Can you help me resolve this please?
Hi, I'm also facing the same issue. Were you able to find any solution ?
Same Error
@@muhammadfaheem4024 Recheck how the column transformer was build and better approach would be to just open a new jupyter file and load the pickle file there and use it to see the results you are gettingn
@@sanyamjain4378 Okay, I'll give it a go!
when i am running python app.py nothing is happening. There is just total silence no error nothing. Don't know what's wrong.
Did you get the solution for this bro...Please let me know
I'm also facing the same issue
src.exception.CustomException: Error in python script name [C:\Users\HP\Documents\GitHub\mlops-01\src\pipeline\predict_pipeline.py] line number [21] error message[Found unknown categories [None] in column 1 during transform]
any solutions
get your solution or not?? same error here from my side. kindly share you get the solution.
The problem comes from the fact that the mapping beetween variables name need to be perfect. In the home.html file you need to replace = "ethnicity" by "race_ethnicity" otherwise you will always get None for the race_ethnicity features. You can check it out by printing the pred_df
Hello sir, thankyou for the help but I have replaced "enthnicity" with "race_enthnicity" in home.html file but still getting the same error. It seems like the error is occuring in column 1 which is "gender". Thankyou again for the help sir!
@@medounelearning2025
@@medounelearning2025 thank you very very much, I was debugging from past 20 mins.
@@medounelearning2025 Thank you so much. I thought I will not get the the solution and leave this project in the middle.
Krish sir, pls upload new content on tech neuron. Its almost 5 months now..
'CustomData' object has no attribute 'get_data_as_data_frame'
i keep getting this error .
how do i solve it
Sir please me give replay of This message. Sir when I am put gender : male , parental education: master, food: standard , read score :100, writing score : 100 . So why its's give me math score 103.25. the exam marks in between 0-100 so why it gives me 103.25 . Please reply this message. Because of this one error I am trying to found ....please give me reply as soon as possible..
Will u show data monitoring and data versioning
i think he will cover the data version
'GradientBoostingRegressor' object has no attribute 'transform'. Getting this error, does anyone have solution?
@krishnaik06, i am getting error like "Internal Server Error
The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application."
Is your error solved?
The name of the preprocessor pickle file that you saved is different from the path name that we given in prediction ,in my case due to spelling of the file name different ,after correction of file name ,the error resolved
Can anyone help?
I am getting this error message "Internal Server Error
The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application." Each time I click on predict math score. Every other things seems okay
Hi Did u find any solution for this error?
@krishnaik06 , i am getting below error after enter values - "Internal Server Error
The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application."
Were you able to figure it out?
i'm getting this error also..it u trying to figure it out...
Got this error when trying to predict data : Internal Server Error
The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application. Anybody know the solution for this error please le t me know
i am also getting the same error
@@siddharthtyagi1254 I got the solution bcoz i forgot to import os in predict_pipeline
@@lionsinescanor405 ok i will try to do the same
@@lionsinescanor405 it is running but when i am predictig the score after passing the values , getting an error ['OneHotEncoder' object has no attribute '_drop_idx_after_grouping']
Traceback (most recent call last)
@@siddharthtyagi1254 There must be some kind of syntax error like spelling mistake in the code
src.exception.CustomException: Error occured in python script name [E:\AI project\mlproject\src\pipeline\predict_pipeline.py] line number [16] error message[Error occured in python script name [E:\AI project\mlproject\src\utils.py] line number [59] error message[CustomException.__init__() missing 1 required positional argument: 'error_detail']]
how to fix this error
🙏💯
Need DL project
Hi Sir,
sorry to say but aap bohot bura samjhate ho
aapka content achcha h
pr skill as a teacher 'fail' sir
aree hadd h
i get this error when i load pickel file [ is not a known BitGenerator module.] any body knows how to resolve it
Thank you soo much sir
Thankyou very much sir 😊