Sabitlenmiş Tweet
Medical student | Data analyst | Python in healthcare
Exploring medical datasets, analyzing patterns, and sharing projects & insights.
Check out my work: github.com/sheriff2005
Open to feedback & collaboration.
English
Sheriff
592 posts
Sheriff
@DataWithSheriff
Med student | Data analyst | Python in healthcare | Exploring medical datasets & sharing projects |aspiring data scientist
Nigeria Katılım Şubat 2026
58 Takip Edilen39 Takipçiler
most people think being good at math is talent.
wrong.
it is training culture.
how chinese learn mathematics by lianghuo fan et al.
looks at how chinese students build mathematical ability from the inside.
not through shortcuts.
through:
repetition
worked examples
deep practice
teacher guidance
conceptual variation
problem solving discipline
the important pattern:
they don’t treat math as something you “get” instantly.
they treat it as something you grind until structure appears.
first you imitate.
then you understand.
then you generalize.
western education often worships creativity too early.
but real creativity in math comes after fluency.
you cannot make elegant moves
if you don’t first master the basic forms.
English
@PhysInHistory Concept of hayflicks theory and senescence theory,
Death is a natural occurrence birth - death cycle
English
Is death a biological inevitability or just an engineering problem we haven't solved yet? 🧠
English
@SolvingForZ Well explained 🔥, that's why we use the cost function to estimate the error in the model , with regards to over fitting or under fitting in model
English
Overfitting occurs when a machine learning model learns the training data too well — capturing not just the underlying patterns but also the noise and random fluctuations specific to that dataset. The result is a model that performs excellently on training data but fails to generalize to new, unseen data. Think of it like a student who memorizes every answer from past exams verbatim: they’ll ace a repeated test but struggle the moment a question is phrased differently. In technical terms, the model has low bias but high variance — it’s overly sensitive to small changes in input.
The usual remedies for overfitting include regularization (like L1/L2 penalties that discourage large weights), dropout in neural networks, cross-validation to detect the gap between training and validation performance, early stopping during training, and simply using more training data when possible. Feature selection and pruning (in decision trees) also help by reducing model complexity. The core idea across all these techniques is the same: constrain the model so it captures the signal, not the noise.
English
day 5 of learning machine learning.
today, i focused on model evaluation, which is one of the most important parts of machine learning.
i learnt classification metrics and evaluation techniques like accuracy, precision, recall, f1 score, confusion matrix, and roc auc. these metrics help you understand how well a classification model is performing beyond just saying “the model is correct.”
i also studied regression metrics and evaluation techniques like mean absolute error, mean squared error, root mean squared error, and r squared. these help measure how far a model’s predictions are from the actual values.
then i went into clustering evaluation metrics, both internal and external. internal metrics help evaluate clusters without using true labels, while external metrics compare the clusters with already known labels.
i’ve not really been posting daily or catching up every single day because of university activities, but the learning is still going on.
slow progress is still progress, and i’ll keep pushing.
Faithiology@faithiologyyy
day 4 of learning machine learning. today, i studied clustering and how it applies to real world problems. i learnt that clustering is an unsupervised learning technique that helps group similar data points together without using predefined labels. it is basically about finding patterns in data when nobody has already told the model what each group means. i also learnt k means clustering. k means works by choosing a number of clusters, finding the center of each cluster, and grouping data points based on how close they are to those centers. i also went through dbscan and hdbscan. dbscan groups data based on density, so it can detect outliers instead of forcing every point into a cluster. hdbscan improves on that by handling clusters with different densities, which makes it better for more complex datasets. day 4 helped me understand that machine learning is not always about prediction. sometimes, it is about discovering hidden structure in data. we keep learning 💪
English
@Marvee_da_Great That's really nice, just started my ML this week, happy to connect with you
English
Day 66 of #100DaysOfCode 🚀
Learnt introduction to Machine Learning
Then built a Linear Regression model in Python:
- Pandas, Matplotlib, Scikit-Learn
- Predicted home prices from square footage
#MachineLearning #Python
#LearninginPublic
English
Are you looking to learn either:
- Cybersecurity
- Data Analytics
- Product Design
- Data Science E.t.c?
I'm hosting a small scholarship aimed at training anyone looking to learn a Tech skill!!
Please only serious minded people.
Comment below your preferred Tech skill 😁
English
Hey @X
algorithm
I'm looking to #connect with people interested in:
- Frontend
- Backend
- Full stack
- DevOps
- AI/ML
- Data Science
- UI/UX
- Freelancing
- Startup
- Saas
Say hi 👋 & Let's grow together #BuildingInPublic #CONNECT
English
Another day to learn.
Another day to improve by 1%.
Medicine.
Python.
Data science.
Repeat.
Slow progress is still progress.
#LearnInPublic
English
@thedevchandra I keep posting but I got no engagement, any tips on how to improve
English
Don't waste time on this:
Excel - 100%
SQL - 0%
PowerBI/Tableau - 0%
Python/R - 0%
Instead strike a balance by aiming for this:
Excel - 25%
SQL - 25%
PowerBI/Tableau - 25%
Python/R - 25%
You don't need to know everything straight away.
English
Big realization: Machine learning is not about maximizing training accuracy.
It’s about building models that survive contact with unseen data.
#MachineLearning #Python #DataScience #AI #ScikitLearn #LearningInPublic
English
Day 46 of my Data Analytics journey📊.
Yo!😁
Congratulations to me🥂🎉🎉….
I got my first internship program😊
If you don’t start, you will never know where you could get to👌.
@ezekiel_aleke @TechSphereAcad
#DataAnalytics
#LearningInPublic
#TechJourney
#Intern
English