simple_trick

[daijob00]

Simple_trick is not offered as an option in the first linear regression function. In the second linear regression function, it is called, but the arguments is incorrect (no learning rate).

[dhern023]

Hi @daijob00. The main difference between the two is the presence of the error collection, but you can call the simple_trick function in the first one (see how below).

If you want to use simple_trick in the second linear_regression function, you'll need to comment out the learning_rate argument as well and adjust the parenthesis like so because the simple_trick only takes four arguments.

        # Uncomment one of the following 3 lines to use the simple, the absolute, or the square trick
        price_per_room, base_price = simple_trick(base_price,
        # price_per_room, base_price = absolute_trick(base_price,
        # price_per_room, base_price = square_trick(base_price,
                                                  price_per_room,
                                                  num_rooms,
                                                  price,
                                                  #learning_rate=learning_rate)
                                                  )

Also, feel free to try the refactored version as well. I did my best to make switching between the different functions a bit more intuitive.

[vilas273]

Hello,
I am getting an error of attribute name, even after installing the utils (pip install python-utils) package
utils.plot_points(features, labels)

AttributeError: module 'utils' has no attribute 'plot_points'
may i know how to fix this.?

[dhern023]

@vilas273 Don't confuse python-utils with utils.py provided with the code.

How are you running it? Can you check the refactored version to see if you're getting the same issue?

[ArmanbekAlkin]

Hello @dhern023 Can you advise what Python software to use with your book (Grokking Machine Learning)? PyCharm or VS Code. I tried to use your coding in Jupyter, or Google Colab, but with files you provided I guess it's better to use an installed software.

[dhern023]

@ArmanbekAlkin What breaks for you?
Did you install the requirements.txt in your conda environment before you ran the code on juptyer? You can also to the same thing on Google Collab.
Afterwards, did you try the refactored version(s)?

[ArmanbekAlkin]

@dhern023 Hi Luis. Thank you for the reply. I haven't seen the requirements.txt file. Let me check your repository for this file, and look at the requirements. I am reading your book (Grokking Machine Learning), and trying to use your codes examples in PyCharm software application. I haven't used Refactored versions, as I noticed the difference with the book and didn't know if they are working. I will try them. My problem was, that for example, I use linear regression coding file from Chapter 3 and trying to execute code line by line and haven't seen the graphs you showed in the book. Now I see that you in refactored version of the file you have plt.show() functions. Let me play with refactored files and see how it goes. I like to use PyCharm to keep all my files and projects in one folder and track the difference with GIT.

[ArmanbekAlkin]

@dhern023 Hi Luis. I tried non-refactored version and it worked just fine, looks like I just didn't install some libraries. Although Turicreate is not working (I realized it's not working on Win10 OS), but it's OK. I am able to move forward to next topic. Thank you!

[dhern023]

I'm not Luis or the author. I'm just the author of the refactored versions since turicreate was a crippling issue to non-mac users.

…

On Mon, Feb 19, 2024, 6:44 AM ArmanbekAlkin ***@***.***> wrote: @dhern023 <https://github.com/dhern023> Hi Luis. I tried non-refactored version and it worked just fine, looks like I just didn't install some libraries. Although Turicreate is not working (I realized it's not working on Win10 OS), but it's OK. I am able to move forward to next topic. Thank you! — Reply to this email directly, view it on GitHub <#12 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB4XDP6A6YLC725WAJBLPLDYUM3H7AVCNFSM5CKZWWW2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOJVGIZDONRRGQYA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[ArmanbekAlkin]

@dhern023 Oh, my apoloigies! I thought this is the Author's repository and that he is replying to me )) I'm sorry my friend. I will play with refactored versions then as well to understand the concept. Thank you one more time for your support.