Top Tweets for #GradientBoosting
Day 70 of ML !!
Gradient Boosting (Regression + Classification)
From math → residuals → log-odds → full implementation
code : [github.com/DiwanshuG/Machine…]
#MachineLearning #GradientBoosting #LearnInPublic
![TheRealDiwanshu's tweet photo. Day 70 of ML !!
Gradient Boosting (Regression + Classification)
From math → residuals → log-odds → full implementation
code : [github.com/DiwanshuG/Machine…]
#MachineLearning #GradientBoosting #LearnInPublic](https://pbs.twimg.com/media/HFyqHPsacAETGaM.png)
Day 88: Advanced Boosting.
Hands-on with Gradient Boosting: trained both Classifier (GBC) and Regression (GBR) models
Then deep dive into the in-depth intuition of XGBoost for classification, understanding the core logic that makes it so powerful.
#ML #GradientBoosting #XGBoost
Day 162: Data Science Journey
->GB: Uniform prob plane tags all class 1; residuals-> errors
->3D scatter + contourf steering stumps to carve adpt bound, min log loss.
->Fix: F(x)=prev + γ*tree (shrink γ<1), tunes 2x accuracy, kills overfit!
#DataScience #ML #GradientBoosting

Day 161: Data Science Journey
->Flat pred plane at mean(y)=0.56 via Plotly 3D; all points classed as 1.
->Scatter3D real pts + Surface for constant pred.
->Resid. r=y-p: vertical errors guide weak learners; start simple, correct via residuals.
#DataScience #ML #GradientBoosting

Well, stayed up a bit late and experimented with new readings and models.
GradientBoostingRegressor is overfitting it seems.
Lasso seems the best one considering the test data score(R² score 2)

Ran this titanic prediction project on my Lazyvim setup.
Looks like .py works really well.
That means....no more jupyter notebooks from an IDE (VS Code ain't working) and a chance to try out Colab.

Happy Sunday
Till now i did
--> Laundry -_ -
--> 1-Lec on XGboost 🔥
--> 3-pages of Gradient bOOsting Research paper
--> KNC & PCA on Fashion mnist data
--> Now applying GradientBoostingClassifier on mnist dataset.

PCA--> Principal components Analysis {DONE}
KNeighborsClassifier {DONE}
Highest accuracy reached 86 :(
Time for some other alogs.

ProtBoost: protein function prediction with Py-Boost and Graph Neural Networks - CAFA5 top2 solution
1. ProtBoost emerged as the 2nd best solution in the CAFA5 competition, leveraging Py-Boost, a novel gradient boosting algorithm optimized for multi-target tasks, and Graph Neural Networks (GCNs) to achieve a score of 0.58240, outperforming over 1,600 teams.
2. Introduced Py-Boost, the first gradient boosting method capable of efficiently predicting thousands of targets simultaneously, with computational speedups of over 50x compared to alternatives like XGBoost and CatBoost.
3. Employed cutting-edge protein language models (Prot-T5, ESM2) to generate embeddings that effectively capture protein sequence properties, achieving superior functional similarity predictions compared to traditional alignment methods.
4. Developed Conditional Probability Modeling (CondProbMod), a novel approach to account for hierarchical relationships in Gene Ontology, significantly enhancing prediction accuracy by incorporating ontology structure directly into model training and ensembling.
5. Utilized a GCN-based stacking strategy to integrate multiple models and external annotations, such as QuickGO and UniProtKB, achieving a substantial performance boost from 0.59 to 0.62 in ablation studies.
6. Demonstrated superior scalability and flexibility of ProtBoost, enabling prediction of up to 13,000 Gene Ontology terms, with models trained on diverse datasets of 142,246 proteins spanning eukaryotes, prokaryotes, and viruses.
7. Highlighted the efficacy of machine learning in protein function prediction tasks, addressing challenges of large-scale multi-target modeling and paving the way for advancements in bioinformatics through collaborative competition frameworks.
💻Code: github.com/btbpanda/CAFA5-pr…
📜Paper: arxiv.org/abs/2412.04529
#Bioinformatics #ProteinFunctionPrediction #GradientBoosting #GraphNeuralNetworks #CAFA5 #MachineLearning

Most Popular Users

Elon Musk 
@elonmusk
240.2M followers

Barack Obama 
@barackobama
118.6M followers

Cristiano Ronaldo 
@cristiano
114.4M followers

Donald J. Trump 
@realdonaldtrump
111.8M followers

Narendra Modi 
@narendramodi
106.9M followers

Rihanna 
@rihanna
98.6M followers

NASA 
@nasa
92.2M followers

Justin Bieber 
@justinbieber
91.7M followers

KATY PERRY 
@katyperry
89.9M followers

Taylor Swift 
@taylorswift13
83.9M followers

Lady Gaga 
@ladygaga
75.4M followers

Virat Kohli 
@imvkohli
73.3M followers

Kim Kardashian 
@kimkardashian
70.8M followers

YouTube 
@youtube
68.6M followers

Neymar Jr 
@neymarjr
66.3M followers

Bill Gates 
@billgates
65.2M followers

Selena Gomez 
@selenagomez
63M followers

The Ellen Show
@theellenshow
62.1M followers

CNN 
@cnn
61.7M followers

X 
@x
60.6M followers

![TheRealDiwanshu's tweet photo. Day 70 of ML !!
Gradient Boosting (Regression + Classification)
From math → residuals → log-odds → full implementation
code : [github.com/DiwanshuG/Machine…]
#MachineLearning #GradientBoosting #LearnInPublic](https://pbs.twimg.com/media/HFypvfvbkAA-VZC.png)
![TheRealDiwanshu's tweet photo. Day 70 of ML !!
Gradient Boosting (Regression + Classification)
From math → residuals → log-odds → full implementation
code : [github.com/DiwanshuG/Machine…]
#MachineLearning #GradientBoosting #LearnInPublic](https://pbs.twimg.com/media/HFypvfraMAAFSzl.png)
![TheRealDiwanshu's tweet photo. Day 70 of ML !!
Gradient Boosting (Regression + Classification)
From math → residuals → log-odds → full implementation
code : [github.com/DiwanshuG/Machine…]
#MachineLearning #GradientBoosting #LearnInPublic](https://pbs.twimg.com/media/HFypvf2aEAAe3Mz.png)











