Comments for EML

Comment on The Best OS For Data Science [From An Ex-Data Scientist] by Armando

Armando — Sun, 17 Sep 2023 14:29:34 +0000

Thanks for the eloquent and concise post on the virtues of macOS for Data Science.

How much more difficult (if at all) has working on macOS become for ML practitioners following nVidia’s withdrawal of CUDA support a few years ago?

Cheers!

Comment on Multivariate Polynomial Regression Python (Full Code) by Ody

Ody — Sat, 08 Apr 2023 19:54:59 +0000

Hi Dylan,
Thanks so much for this code, I used it and it worked for multivariate polynomial regression. However, if I want to visualise it to see the polynomial fit, I get the the error that my X (2 variables) and y (1 variable) are not of same dimension. How can I view my polynomial fit using scatter or line plot.
Thanks

Comment on GPT-3 vs Bloom [Real Code & Real Results] by Dylan

Dylan — Thu, 09 Feb 2023 02:57:15 +0000

In reply to DZoh.

You’re on the right track, though I don’t think Bloom would want their product compared to a lesser GPT.

I know most of the hype surrounding Bloom was being self-hosted, so we went with a model version that could be… self-hosted.

The download on even just this one was atrocious, and I couldn’t imagine downloading a more expansive model.

Comment on GPT-3 vs Bloom [Real Code & Real Results] by DZoh

DZoh — Wed, 08 Feb 2023 19:54:44 +0000

If I understand correctly you haven’t download the “real” bloom model (which is more than 300gb) but a smaller one. I’m not sure it would really fit inside a regular computer anyway.
Therefore it seems normal than the generated output are… well… more than disappointing since it’s like comparing Gpt2 to gpt3.
Bloom (the large one) has an API accessible on Huggingface. Even for this large model I still find the results difficult to compare to the one provided by GPT3. They are less usable.
Yet I think that the price of GPT3 token generation (and the time it takes) would remain a problem if you plan to use it intensively in any deployed app.

Comment on Multivariate Polynomial Regression Python (Full Code) by Dylan

Dylan — Fri, 20 Jan 2023 22:08:02 +0000

In reply to Renald.

Hello! what do you mean by the output?

You’ll need to use it to predict something to get an output for a model.

Dylan

Comment on Multivariate Polynomial Regression Python (Full Code) by Renald

Renald — Fri, 20 Jan 2023 14:07:01 +0000

Dear Dylan,

Thank you for great explanation. I need to ask 1 thing.
I want to check the regression output after fitting the model. How to do that?

Thanks

Comment on Data Science vs. Bioinformatics: [They’re Not The Same] by Albert M Mahaffey III M.S. in Biotechnology with Bioinformatics

Albert M Mahaffey III M.S. in Biotechnology with Bioinformatics — Tue, 03 Jan 2023 20:08:22 +0000

Bioinformatics is data science, unlike biotechnology, bioinformatics is purely dry lab work which predominately focuses on developing algorithms and codes to process, analyze, and assemble biological data. Bioinformaticians uses already available tools and terminal(shell) to execute data but have minimal understanding and do not code. Biostatistics on the other hand requires coding in the R language as well as SAS and ANOVA at the advance level so yes programming is used in R. A Bioinformaticists is a programmer that utilizes Python, R, Perl, Java, ML, ANN, database systems, etc. to visualize and manipulate biological, chemical, and pharmaceutical data. To truly master bioinformatics or biostatistics you need to be proficient in a programming language. My advice to anyone getting into Bioinformatics is to master R, Python, and ML because they are very few positions for just Bioinformaticians and typically most employers want you to have a master’s in bioinformatics but if you master the primary programming languages you should be able to find work in data science fairly easy.

Comment on Chi-square Test of Independence In Python (Full Code) by Jarrod

Jarrod — Thu, 22 Dec 2022 04:27:25 +0000

Great delivery. Sound arguments. Keep up the great effort.

Comment on Multivariate Polynomial Regression Python (Full Code) by Dylan

Dylan — Thu, 01 Dec 2022 19:34:48 +0000

In reply to ChienChi.

Hi ChienChi,

poly_model is defined in the for loop!

A normal “fit” method won’t have any output, as it’s not transforming (as seen in the for loop).

a fit_transform is a little different than a fit method.

Dylan

Comment on Multivariate Polynomial Regression Python (Full Code) by ChienChi

ChienChi — Thu, 01 Dec 2022 03:22:23 +0000

Dear Dylan Kaplan,

The code is useful and thanks for sharing it.
I put the sample code below and need your help to
confirm one question:

At line 10, the parameter values passed to mean_squared_error()
are “y_value” and “y_pred”.
“y_value” is the result fetched from the Excel file.
“y_pred” is the result returned by line 8.

But at line 4, “poly_model.fit(poly_x_values, y_values)” is executed.
What does this line take effect?
This line doesn’t generate any result.
What is the relation between line 4 and line 10?

Thanks very much.

======================================================

1 poly_model = PolynomialFeatures(degree=degree)
2
3 poly_x_values = poly_model.fit_transform(x_values)
4 poly_model.fit(poly_x_values, y_values)
5
6 regression_model = LinearRegression()
7 regression_model.fit(poly_x_values, y_values)
8 y_pred = regression_model.predict(poly_x_values)
9
10 plt_mean_squared_error.append(mean_squared_error(y_values, y_pred, squared=False))

Sincerely
ChienChi