Machine Learning For Chemists: Practical Applications

Machine Learning For Chemists: Practical Applications
Published 9/2025
MP4 | Video: h264, 1920×1080 | Audio: AAC, 44.1 KHz

Use Data Science to solve chemistry problems: predict molecular properties, classify compounds, and resolve spectra

What you’ll learn

Understand the fundamentals of machine learning and how they apply to chemical problems. Confidently handle real-world chemical datasets for ML applications.

Select and implement supervised and unsupervised learning algorithms for chemistry. Apply regression, classification, and clustering methods to chemical data.

Interpret model performance using evaluation metrics relevant to chemical research. Visualize chemical datasets and model predictions effectively

Use Python libraries (scikit-learn, pandas, numpy) for ML tasks in chemistry, Apply ML to predict molecular properties such as Solubility, HOMO-LUMO gap

Gain practical skills to integrate ML into chemistry. Use ML-driven insights to support research publications and projects.

Separate the spectra (infra-red spectra) of each compound from a mixture of spectra using Machine learning approaches

Requirements

Description

Overview

Section 1: Introduction

Lecture 1 Introduction

Lecture 2 Machine Learning (Theory Lecture)

Lecture 3 Key Concepts of Machine Learning (Target, model training)

Lecture 4 Machine Learning Types, a graphical presentation

Section 2: Supervised Learning: Predicitn boiling points of alcohols through Machine Learni

Lecture 5 Regression method of supervised learning (Theory)

Lecture 6 Linear Regression example of Boiling Pont of alcohols explained

Lecture 7 Linear Regression Model equation

Lecture 8 Analyzing dataset of boiling points of alcohols in NotePad

Lecture 9 Placing the dataset file in Google Drive and opening Google Drive

Lecture 10 Setting up Google Colab Environment

Lecture 11 Mounting Google Drive in Colab

Lecture 12 Reading dataset of boiling point through Pandas and reading head.

Lecture 13 Deta Preprocessing approaches

Lecture 14 Cleaning the dataset before ML

Lecture 15 Feature Engineering

Lecture 16 Feature Selection

Lecture 17 plotting boiling point against possible features (no ML)

Lecture 18 Importing libraries for supervised machine learning

Lecture 19 Defining features and Target

Lecture 20 Completing the code for predicting boiling points

Lecture 21 Model evaluation Metrices (Theory)

Lecture 22 Evaluating the model and plotting the results

Lecture 23 Effect of test size on the performance of ML Model

Lecture 24 Graph of predicted vs actual boiling points

Lecture 25 Residual Plot

Lecture 26 Predicting bp of an unkonw alcohol

Lecture 27 Model II of Regression (Ridge Model): with standard Scalar for boiling Point

Lecture 28 Robust Scalar for Ridge Model and coefficients

Lecture 29 Outliers and How to handle them

Lecture 30 Model III: Lasso Model of Regression for predicing Boiling Points of Alchohols

Lecture 31 Cross Validation (Theory)

Lecture 32 K-fold Cross Validation In Machine Learning

Lecture 33 Scalars (minmax, standard and Robust) and effect on coefficients of features

Section 3: Predicting solubliity of compounds using Random Forest model of supervised ML

Lecture 34 Installing RdKit library in Google Colab

Lecture 35 Importing Libraries required for the problem

Lecture 36 Extracting features as descriptors for molecules using smiles in RdKit

Lecture 37 Loading dataset and preprocessing the data

Lecture 38 Grid search and the best grid values for the problem

Lecture 39 Model evaluation and the performance

Lecture 40 Plotting the results (feature importance, residual frequency, etc)

Section 4: Supervised Machine Learning for Classification of Compounds

Lecture 41 Supervised Learning Classification (Theory) I

Lecture 42 Supervised Learning Classification (Theory) II

Lecture 43 Classification of compounds example explained

Lecture 44 Predicting the classification of compounds (importing Libraries)

Lecture 45 Evaluation of the model throgh Confusion Matrix

Lecture 46 Dimensionality Reduction (Principal component analysis PCA)

Lecture 47 Clustering example explained in theory

Lecture 48 Analyzing Wine data

Lecture 49 Unsupervised learning Clustering import libraries

Lecture 50 Clustering, Plotting PCA1 against PCA2

Lecture 51 Silhoutte Score

Lecture 52 DBSCAN Approach for clustering

Section 6: Unsupervised Learning for blind signal separation

Lecture 53 Blind Signal Separation example explanation

Lecture 54 FAST ICA 1: importing libraries and data of mixture spectrum in python

Lecture 55 FAST ICA II: Completing the code for independent component analysis

Lecture 56 Plotting two extracted spectrum from the mixture spectrum in parallel

Lecture 57 Plotting all extracted spectra from the mixture spectrum

Lecture 58 Identifying cyclohexane by comparing the extracted spectrum with the reference

Lecture 59 Identifying all components of the mixture by comparison with the reference spect

Lecture 60 Plotting all four spectra in one figure

Lecture 61 Plotting all four spectra in one figure (2nd approach)

Graduate students and researchers in chemistry or chemical engineering,Data scientists and ML engineers interested in scientific applications,Academic researchers wanting to integrate machine learning into chemical research,Computational chemistry professionals seeking to expand their ML toolkit,Laboratory managers implementing data-driven approaches to experimental design