Stan Si🇺🇦

Dive deeper in data processing

[Stan Si🇺🇦 @StanSi_]

The second post in a series about image processing in Python with #numpy, #scipy, and #skimage. There are some basic thresholding approaches, that poorly segment a target object, but notable creepify a sample image. https://t.co/3ETgKmTiXt

[Stan Si🇺🇦 @StanSi_]

@JessMead7 @PhDVoice The most important skill in Academia is grant application writing and fund raising. If you sure that can do this with great outcome, then choice Academia. "Publish or perish" is shit. The proper way is "raise money, gather team, communicate, seek ideas, publish OR perish".

[Stan Si🇺🇦 @StanSi_]

There are nine useful tools for LaTeX and MathJax user. https://t.co/9OKeQEV6A2
#LaTeX #MathJax #Overleaf

[Stan Si🇺🇦 @StanSi_]

@matplotlib Is it possible to get the effect of plt.axis("off") by changing in rcParams? I want to work with images and disable axis for plt.imshow(...) in an entire notebook.

[Stan Si🇺🇦 @StanSi_]

@marktenenholtz For CPU code, it is not true. Python reduce performance twicely, because of the imports and objects calls that request low-level code. However, C/C++ code for ML cannot be supported. Each pipeline on C requires 2 weeks of work.

[Stan Si🇺🇦 @StanSi_]

@driscollis scipy and scikit-image. I use Python for processing of biomedical images and signals.

[Stan Si🇺🇦 @StanSi_]

@hapyresearchers Don't give up! You should try again and agian, If your belive in the importance of your studies.

[Matt Dancho (Business Science) @mdancho84]

I think the message in Data Science needs to be: Don't believe everything you read.

#rstats

[Stan Si🇺🇦 @StanSi_]

The other point of view. In "Blindsight" written by P. Watts, aliens have no concsciousness and work like the Chinese room; however, that do not prevent them from being hostile;) Such techs may be non-santient in real, but dangerous in applications.

[Tom Gara @tomgara]

This discussion between a Google engineer and their conversational AI model helped cause the engineer to believe the AI is becoming sentient, kick up an internal shitstorm and get suspended from his job. And it is absolutely insane. https://t.co/hGdwXMzQpX

[Stan Si🇺🇦 @StanSi_]

Useful idea, but I see trouble here. The simplest backward transform is a matrix. Matrix series has an issue with condition numbers, eigenvectors, fixed points, etc. I assume that not more than 20 backward transforms will really be available on practice.

[The Struggling Scientists @TheStrugglingS4]

Academia, where money is always a problem no matter how many titles and awards you have😅

@OpenAcademics #phdlife #phdvoice #phdmemes #gradschool #AcademicTwitter #PhD #AcademicChatter #phdchat @PhD_Genie

[Stan Si🇺🇦 @StanSi_]

@FerrerCfd @hapyresearchers In really, only PhD students from the first 300 better universities publish in high impact journals. Most PhD students in the world have no access to proper research facilities or can't pass paywalls...

[Deb Raji @rajiinio]

Many of ML's major benchmarks have already become obsolete. It's getting pretty urgent to re-think ML evaluation.

[Stan Si🇺🇦 @StanSi_]

@mm_johari Your recent study is really great. At least twitter promotes it;) I wish you to find a good next position.

[Stan Si🇺🇦 @StanSi_]

Image processing with Python. Part 1. Image reading, clip, flip, and channel extraction. #numpy #Python #ImageProcessing #MachineLearning

https://t.co/qCzefdWVjd

[Parvez Shah @parvezshahshaik]

What’s Like a Data Scientist in Silicon Valley?
Growth & Learning that benefit every aspiring DS

#DataScience

#IoT #IIoT #PyTorch #Python #RStats #TensorFlow #ReactJS #GoLang #CloudComputing #Serverless #Linux #Programming #Coding #100DaysofCode

[Maarten van Smeden @MaartenvSmeden]

NEW PAPER
The harm that can be done with SMOTE, oversampling and undersampling to "correct" for class imbalance, illustrated using logistic regression
Open access: https://t.co/n3kDAqCHoD

[Data Science Dojo @DataScienceDojo]

100+ Free #DataScience eBooks (downloadable PDF) for Beginners and for Experts: https://t.co/UcHTRm5wlP