Recommendation Systems or any kind of chatbots can be a good option.

Hey, even I'm in the process of learning and researching about GenAI. So once I'm done, I'll make videos on GenAI concepts and projects in detail.

Yes, uploading soon!

Thanks for appreciation! I'll soon make that but on a different dataset. As the lung cancer one becomes too heavy for processing after augmentation.

@@InsightsByRish Okay mam, thank you, but please can you use the Histopathological images dataset, actually I am a graduate student researching cancer cell detection using Histopathological images. so I will get some help. thanks!

Ok, will make soon.

Yes, soon

Thank you! And yes, I'm making more such videos. Stay tuned!

Will definitely try!

Please try as soon as possible 🙏. It will be very helpfull

Will try! But I have the machine learning code here: github.com/RishaRane/Market_Segment_Clustering.git

drive.google.com/file/d/1dMT88UCeBOIRlMv78l-a5Z201kylCax4/view?usp=sharing

Thanks a lot ​@@InsightsByRish

I think it's likely a combination of both factors(overfitting and imbalance). But still I would say 'Overfitting' can highly be a factor (as that's a small dataset getting trained on high no.of epochs).So you might try adding few dropout layers or Batch normalisation layer. After adding that your accuracy will be dropped a bit but atleast it wouldn't fluctuate.

@@InsightsByRish After experimenting with parameters, it is not fluctuating --> I reduced the number of data augmentation techniques, changed the learning rate to 0.002 and also changed strides to 1. It is still overfitting though may be the dataset is small need to work on it.

Hey, I've provided the GitHub repository link in description box. Dataset is available there.

Will definitely make 👍🏻 But let me also inform you that I already have all this things made and uploaded in different parts on my channel.

What's the error?

@@InsightsByRish model.fit(X_train, y_train) Input X contains NaN. LinearRegression does not accept missing values encoded as NaN natively. For supervised learning, you might want to consider sklearn.ensemble.HistGradientBoostingClassifier and Regressor which accept missing values encoded as NaNs natively. Alternatively, it is possible to preprocess the data, for instance by using an imputer transformer in a pipeline or drop samples with missing values.

​@@Luminouss12096 I think your input data frame (X) still consist of missing values. Try removing that before building the model. If you still encounter the error then mail your .ipynb file on [email protected].

Hey, I've provided the link of Github repository for the code in the description box.

Thank you ma'am thanks a lot 🙏

I have the code for that. I can mail you that if you want.

It's not select_datatype('object'). It is select_dtypes('object').

Yes I tried this one also but it's showing issue... While I run the encoder it shows an error that you have to select int or str

I tried it also but it's showing error while I run the encoder part its showing an error that you should select int or str in encoder.fit(df[cat_col])​ @@InsightsByRish

@@vikashsingh6888 Hey, not really able to understand what's going wrong with your code, but I would suggest that restart your kernel and again implement the code from beginning line by line as it is (I've provided the code on GitHub).

If you will provide your insta or something for contact even email I will send you the ss .. if you can it will help me 🙏

Welcome 😊. If your theoretical concepts are clear then I would say the best way is to watch a lot of implementation videos where every person explains with different methods. At the end of the day you at least get the core gist of things altogether.

Ok, will make and upload somewhere around coming Tuesday.

@@InsightsByRish ok

Yes, will make 👍🏻

Thank you. Also that's a good suggestion, will definitely include statistics project.

Thank you! 😊

@@InsightsByRish kindly upload generative ai videos

Right now I am actually just making beginner and intermediate level projects, but will also try to upload advanced levels as well 👍🏻

Logic is the columns that don't help us in any way for making predictions are removed. Like the User ID column which has zero correlation with the target column. And for rows they are either removed if they contain null values or outliers.

@@InsightsByRish can you make separate video to discuss which features we have to take and which should remove , in our particular problem statement.you r videos are so helpful

Thank you! Will make one on that topic soon.

@@InsightsByRish Great 👍

Please provide me a link for this particular data set

Yes, it should be Q3. I've mistakenly considered it as Q1.

Yes you can use Standard Scaler as well. I used MinMaxScaler as I wanted to transform the values only between the range 0 to 1. Basically when you are predicting something which is continuous in nature, you apply Regression Model. Eg: House Price Prediction, Temperature Prediction etc. Here the values/output can have any value like 25K, 45K etc. But when you have specific values in your target column then it's a classification model. Like 'Loan Prediction' where the output can either be 'Yes (Will get Loan)' or 'No(Won't get loan)'. So here the output won't take any value but either 'Yes' or 'No'. So that's a classification model.

ThankYou @InsightsByRish . Now it’s crystal clear 🙂

Ok I'll try!

Thank you so much! I have provided the link of GitHub repository for the code files in the description box.

Hey, I have provided the link of GitHub repository for the code in the description box.

It's 'Google Colab'.

Yes, I have provided the link of dataset in the description box.

Ok.

@@InsightsByRish Mam you upload video on this.kindly upload it soon

@@user-vn5gs2pb3j Actually the issue is the dataset for that is too large. My current GPU won't be able to process so much of data.... Still I'm trying.

@@InsightsByRish Oky mam Thank you

Partially yes! Though this steps can be applied to any predictive model but there might be few additional steps as well depending upon your dataset.

Yes sure..!

Yes sure! Stay tuned.

@@InsightsByRish Ok Thanks

Hey, I have provided the link of GitHub repository for the code in the Description Box.

Thank you so much​@@InsightsByRish

Thank you. Pleased to know you liked the content.