Agent:
  Langchain: Create Agent
  Langsmith: Monitor Agent Logs (Langsmith=Sumologs)
Create Virtual Env in Python
Run ML on AMD GPU
Machine Learning & Types:

Supervised Learning
  |- Regression(Linear, Logistic, Decision Tree(Random Forest))
  |- Classifier(Binary, Multiclass)
Unsupervised Learning
  |- Clustering
Algorithms(KNN, ANN(NSW, HNSW))

Terms: AI vs ML vs DL, Activation Function, Bias/Sampling Bias, Conda, CNTK, feature, Overfitting, Underfitting, Optimizer, label, Large Language Model(LLM), Layer, Learning Types, Loss, Matplotlib, Metrics, Neuron, Neural Network, Tensor: Add Vector & matrix, Dot Product, Tensor Reshaping stochastic gradient descent (SGD) Tensorflow Variance, Loss, MAE(Mean Absolute Error), MSE(Mean Squared Error), RMSE(Root Mean Squared Error)
Code:
  Opencv: Draw rectangle around faces
  speech_recognition: Convert Voice to Text
  OpenAI APIs: Call openai APIs from code/Postman
ML Model
  Claude Code
  Ollama: Open Source Model Runner
ML Libraries
  Keras(build Neural Networks):
  Introduction: Models(Sequential, Functional, Subclassing)
  Layers(Dense, Convolutional..)
  Functions: Dense(), compile(), fit(), predict(), evaluate()
  Code:
   Classify grayscale images into(0 to 9)
   Classify movie Review as +ve or -ve based on 50k review.
   Linear Regression
   Binary Classification
  Pandas:
  Introduction(series, dataframes), Conversion(dataframe to numpy array), Plots(Histograms, boxplot and save in png file)
  Functions: any(), read_csv(), describe(), drop(), dropna(), head(), isNull(), loc[], select_dtypes(), shape(), sum()
  scikit-learn(build light weight models)
  Introduction
  Code:
   1. Calories Predictor during workout using: DecisionTreeRegressor, Random Forest Regression model
   2. Filling Missing values using Imputator
   3.Anomaly Detection using scikit learn
   4.California House Price Prediction
  Functions: fit(), fit_transform(), Imputator, SimpleImputator, transform(), train_test_split(), RandomForestRegressor()
  Tensorflow
RAG (Retrieval-Augmented Generation)
 RAG Pipeline: Introduction, Code, Evaluating RAG Pipeline(Precision, Recall), Chunking, User Query Rewriting
 Searches in RAG (Keyword, TF-IDF, BM25)
 RAG Pipeline Observability
 Companies working in RAG Space
  RAGAS(Retrieval Augmented Generation Assessment)
  Introduction
  Testing RAG using RAGAS
Protocols:
  MCP(Model Context Protocol)