Myself Sagar | Full stack developer by day & night ππ». I Always on the lookout for the next challenge to crack. | Let's build something amazing together! π‘
Learned RAG basics and how LLMs use real data
π Document loaders
π§ Embeddings
π§² Vector stores
π Retrieval with context
Building smarter LLM apps π€
GitHub π https://t.co/r7IgwUJ7nD
#RAG#LLM#GenAI#AI#MachineLearning#DeepLearning#Python
LLM basics finally clear
π€ What is an LLM
βοΈ Prompting
π‘οΈ Temperature
π’ Tokens
π HuggingFace and OpenRouter
Learning GenAI step by step π§
GitHub π https://t.co/r7IgwUJ7nD
#LLM#GenAI#AI#MachineLearning#DeepLearning#Python
Learning NLP from the ground up :
Tokenization
Embeddings
Built a small sentence similarity project and understood how text turns into meaningful vectors β¨
Code here π https://t.co/pDP3jEZpgM
#NLP#AI#MachineLearning#DeepLearning#Python#Developers
Attention feels powerful and now I understand why they outperform RNNs
Learning HuggingFace basics made text tasks feel simple
Feeling confident with Transformer intuition
#Transformers#Attention#DeepLearning#AI
GitHub:https://t.co/pDP3jEZpgM
Learning CNNs boosted my confidence in computer vision tasks
From convolution filters to feature maps and pooling the entire pipeline finally makes sense
Running inference with MobileNet and ResNet was the real moment where everything clicked.
#DeepLearning#CNN#AI#ML
Learning CNNs boosted my confidence in computer vision tasks
From convolution filters to feature maps and pooling the entire pipeline finally makes sense
Running inference with MobileNet and ResNet was the real moment where everything clicked.
#DeepLearning#CNN#AI#ML
Built a simple feed-forward neural network today and trained it on a dummy dataset. Learned how tensors, model layers, loss functions and the full training loop work together in PyTorch. A solid step toward mastering deep learning.
#DeepLearning#PyTorch#AI#ML
Built a simple feed-forward neural network today and trained it on a dummy dataset. Learned how tensors, model layers, loss functions and the full training loop work together in PyTorch. A solid step toward mastering deep learning.
#DeepLearning#PyTorch#AI#ML
Built a small Payment Fraud Detection model today ππ³
Cleaned data, handled imbalance, trained models and tested fraud predictions.
Code here: https://t.co/TRNVsPrgUE
#MachineLearning#AI#FraudDetection
Just finished my Diabetes Prediction ML project.
Cleaned data, handled zero values, scaled features and trained a Logistic Regression model. Evaluated using accuracy, precision, recall, F1 and confusion matrix.
Reached around 74% accuracy a solid baseline.
#MachineLearning#AI
Just finished my Diabetes Prediction ML project.
Cleaned data, handled zero values, scaled features and trained a Logistic Regression model. Evaluated using accuracy, precision, recall, F1 and confusion matrix.
Reached around 74% accuracy a solid baseline.
#MachineLearning#AI
Learnt how to evaluate ML models using accuracy, F1 and MSE.
Also understood how precision, recall and F1 are connected to the confusion matrix.
Helpful to see the exact mistakes a model makes.
#MachineLearning#AI#ModelEvaluation#DataScience
GitHub: https://t.co/mdmtmH4mwz
Completed Clustering and K-Means today.
Learned how clusters form, how centroids move, inertia, elbow method and how to visualize cluster groups with simple plots.
Also built a mini K-Means project.
#MachineLearning#AI#Clustering#KMeans
GitHub: https://t.co/mdmtmH4mwz
Finished learning Decision Trees and Logistic Regression and built a Titanic Survival Prediction model using Scikit learn.
Touched 80 percent accuracy and learned how to evaluate a classifier.
Project code here : https://t.co/I8w6jRAzP9
#MachineLearning#AI#Python#MLProject
Built a small house price prediction project using python and linear regression.
Cleaned the data trained the model and tested it on unseen rows.
Good first step in applied ML.
Check the project : https://t.co/cxVQ4FmMON
#python#ai#ml#datascience
Supervised Learning finally clicked today
Covered:
regression classification,
linear regression
logistic regression
KNN
model training prediction evaluation
loss functions
Feeling more confident after running everything with real examples
On to the next day
#AI#Python
Today I completed one of the most important parts of AIML which is data preprocessing.
I worked on handling missing values scaling encoding outlier detection train test split and basic pipeline building.
It feels great to understand how much clean data improves the model.
Working on my AIML fundamentals and broke everything into clean day wise sections
Day 1 Python Fundamentals
Day 2 NumPy and Pandas
Day 3 Math for ML
Day 4 Data Visualization
All notes are on my GitHub if you want to check them out.
Link : https://t.co/Pvr23ujpBs
Wrapped up my Data Visualization study for AIML.
From preprocessing to interactive dashboards, and from Matplotlib to Plotly β learned how a clean chart can explain what pages of text canβt.
Good visuals = clear thinking.
#AIML#LearnInPublic#GenAi
Completed my Calculus module for AI and ML.
Feeling strong with derivatives, gradients and optimization.
Ready for the next step in my AI learning journey.
#AIML#LearnInPublic