<aside> 🔗
<aside>
</aside>
<aside>
</aside>
<aside>
</aside>
</aside>
<aside> 📝
Purpose
The goal of Pharmaceutical Formulation project is to explore and create different models to predict the viscosity of the product. This can be divided into two parts:
Exploratory Data Analysis: Provides an analytical look at the dataset and allows us to determine what type of transformations and preprocessing might need to be done to the dataset prior to modelling.
Model Training: Create and train the models on the training dataset, and then compare the accuracy of the models by predicting on the rest of the data.
Method
The dataset was manually created by me based on the Design of Experiment(DoE) that my lab did. The data was explored and preprocessed in R and then transformed in Python to prepare the final datasets using Feature Engineering. The models were trained in Python using Scikit-Learn and the model accuracies were compared graphically
Outlook
This project is a simple, but powerful way to quickly create visuals for datasets with multiple independent variables and dependent variables. The variables being modeled or explored can be quickly switch through the use of functions and the amount of data can be easily scaled up to fit thousands of experiments and provide insights. The models can be used to create new combinations for real life process development.
</aside>
<aside> 📝
</aside>