Video

Can I have access to this colab code please?

Is this an IDS?

this is the best tutorial for any learners who wants to study Uplift model, big thanks!!

thank you for this Tutorial!

How can I save the model to an Azure Blo Storage?

Perfect! Thank u...

For anybody watching this, it is worth saying that doing hyperparameter tuning by selecting parameters with the smallest CIs, as this video appears to be recommending, is total garbage statistically. In this case, what you're doing here is just making your CIs less accurate.

I am getting KeyError 0 when trying to run the model with class weight technique. My y_label has 0 & 1 and I have assigned weight manually as weight = {0:1, 1:32} but the code throws KeyError. Tested everything but not able to figure out the issue.

causal inference tutorial

How was the true difference 10 calculated?

What does feature-complete training mean?

How to deploy code from GitHub to Databricks workspace? cd part

you should try to speak yourself instead of using AI, otherwise good content

Very clear and helpful

Thanks for the indepth tutorial. I am wondering if any of these can be used on categorical variables. The only algorithm I know of for working with categorical data is K-Prototypes. Thank you in advance.

nice tut, thank you

Fantastic video …huge help. I’m hoping I can ask a semi related question. I wish to upload my own image (cat) from either iCloud or external hard drive. I can get the image in, but there is no right click “copy link”. I’ve opened in new tab, but as soon as I want to edit (cartoon), everything disappears. Certainly I watch others do it, but I’m missing something or is this just a no with a Mac. I’d be thrilled if you had advice. Txs. Kate

Thanks!!!Great content!

Thanks so much bro..can i use undersampling technique before Splitting the dataset into training and testing because it doesn't introduce any data leakage like in oversampling technique ?

Why doesn't power analysis require the total population to determine sample size? Ignoring the total population, wouldn't a large population have the same sample size as a small population?

Great content! Thank you! Where can I find datasets?

V informative and well-explained. Thank you!

Is there any evaluation model for this?

Thanx for the video....

How to run neural network in noteable or is there any other plugin which we can use to automate deep learning or autoML for ANN?

Hello, another great video about Prophet modeling for time series! I have a question: as here you follow the train/test split for training and eventually evaluating the model, can you also make a video where you would just use the whole time series to train the model? On this latter approach, I don't understand how to add extra regressors, because if I use the make_future_dataframe I would just get Nan values for the extra regressors. This isn't clear for me! Also, after you trained and evaluated the model, how can you make predictions into the future (beyond test data)? Thank you very much!

Sweet video! This tutorial seems to have the most comprehensive, hands-on summary on how to build and use a Prophet model. I am finding all of the info you presented on a time series project on my own. I really liked the through explanation and the code you shared. Thank you!

Thanks so much! it really helps! But I do have a question regarding the finding similar user part. the user-item matrix I have is containing a lot of missing values as well, just like yours. Howver, when I tried to use the pearson correlation to get the correlation, why I did not get NaN for those entries with missing value; instead, I still got numbers. That's so weird cuz based on your illustration, it should return NaN for missing values correlation.

Thank you very much! For a very long time I have been looking for information about embedding categorical variables in boosting models, but I found almost nothing or the information was incomprehensible. Now I understand everything. Thank you!

notebook link is no longer available😭

Thank very much for the content. Please what application do you use for reading the text?

prophet was proven to suck in time series forecasts

Im using tableau public , is it possible to connect to google drive ?

🏪 Purchase data science and computer science themed products in my Amazon store: amzn.to/40HUTsl Products mentioned in the video: ✅ Christmas tree sweatshirt: amzn.to/47sXfxU ✅ Naughty data scientist sweatshirt: amzn.to/47Cfcd6 ✅ Naughty computer scientist hoodie: amzn.to/49NVYms ✅ Christmas tree pillow: amzn.to/40PjeMM ✅ Data calling phone case: amzn.to/40U1dx4 ✅ Data calling t-shirt: amzn.to/416ok7L ✅ AI nerd t-shirt: amzn.to/3sI3kHB ✅ AI nerd sweatshirt: amzn.to/40QgcYC ✅ Computer Science Era tote bag: amzn.to/40Se5n3 ✅ Data Science Era tote bag: amzn.to/47GTmF9 ✅ Computer Science Era sweatshirt: amzn.to/3QPxGji ✅ Data Science Era t-shirt: amzn.to/49KwxlH ❤ Blog post for this video: medium.com/@AmyGrabNGoInfo/top-5-christmas-gifts-for-data-scientists-and-computer-scientists-0fe131f37486 🚛 GrabNGoInfo Machine Learning Tutorials Inventory: medium.com/grabngoinfo/grabngoinfo-machine-learning-tutorials-inventory-9b9d78ebdd67

Your channel is very underrated. Great work.

Great collection for quick grab-and-go for interviews. Can you please make one similar for SQL, advanced SQL preferably?

This was one shitty intro. Maybe next time do it by yourself and not just read some pdf

Thanks for the video, very useful content

Thank you for your efforts! I have one question. You gave a condition that keep only the movies with greater than 100 ratings. could you explain why you give a condition like that?

Thanks for your video. I do not undestand cross validation. You have to say to algorithm what was de model, but then using an initial part of the series the algorithm makes a new model and makes a forecast. Then more data is added to the series and a new model is made again? If in each new iteration a new model is made, what is the idea of ​​telling the algorithm what the original model was, what is the use of this?

❤ Blog post for this video: medium.com/@AmyGrabNGoInfo/how-to-access-llama-2-free-generative-ai-llm-alternative-to-chatgpt-api-359569b27c3a 🚛 GrabNGoInfo Machine Learning Tutorials Inventory: medium.com/grabngoinfo/grabngoinfo-machine-learning-tutorials-inventory-9b9d78ebdd67 🎞 GenAI playlist: studio.czcams.com/users/playlistPLVppujud2yJpfKPS1IFyAvudXQAa8Eouk/videos 👩🏻‍💻 Follow me on LinkedIn: www.linkedin.com/company/grabngoinfo/

Hi, I keep getting error for spark_forecast.show(5) do you have any idea about this?

This is really helpful. Thanks a lot

This is really helpful. Thanks alot.

Could you please share link to the notebook?

Quick question - The MAPE for the log-transformed data is 0.67%. How can you compare that directly with 4.45%. 4.45% is with the data on original scale right?

Is this with machine voice-over? It's very mono-tonal.

I am not able to access the whole code bro.

Clear and easy to follow

Awesome video! But why the anomaly data in X_train is used to build the model? As I know, only the normal data should be used to build the model.