VIPUL1306 / Comet-for-Data-Science

Comet for Data Science, published by Packt

Comet for Data Science

This is the code repository for Comet for Data Science, published by Packt.

Enhance your ability to manage and optimize the life cycle of your data science project

What is this book about?

This book provides concepts and practical use cases which can be used to quickly build, monitor, and optimize data science projects. Using Comet, you will learn how to manage almost every step of the data science process from data collection through to creating, deploying, and monitoring a machine learning model.

This book covers the following exciting features:

• Prepare for your project with the right data
• Understand the purposes of different machine learning algorithms
• Get up and running with Comet to manage and monitor your pipelines
• Understand how Comet works and how to get the most out of it
• See how you can use Comet for machine learning
• Discover how to integrate Comet with GitLab
• Work with Comet for NLP, deep learning, and time series analysis

If you feel this book is for you, get your copy today!

Instructions and Navigations

All of the code is organized into folders.

The code will look like the following:

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from datetime import datetime

Following is what you need for this book: This book is for anyone who has programming experience, and wants to learn how to manage and optimize a complete data science lifecycle using Comet and other DevOps platforms. Although an understanding of basic data science concepts and programming concepts is needed, no prior knowledge of Comet and DevOps is required.

With the following software and hardware list you can run all code files present in the book (Chapter 1-11).

Software and Hardware List

Python 3.8 Python libraries: | 0. comet-ml==3.23.0

1. findspark==1.4.2

2. gradio==3.2.2

3. matplotlib==3.4.3

4. numpy==1.19.5

5. pandas==1.3.4

6. pandas-profiling==3.1.0

7. pyspark==3.2.1

8. scikit-learn==1.0

9. seaborn==0.11.2

10. shap==0.40.0

11. spark-nlp==3.4.4

12. sweetviz==2.1.3

13. tensorflow==2.8.2
    Windows, macOS, or Linux If you are using macOS, please make sure that the chip is not Apple M1. To overcome this problem, you can use Google Colab

Java SE Development Kit 17.0.2 (optional for Chapter 4, Workspaces, Projects, Experiments, and Models)
| Java libraries (optional for Chapter 4,|Workspaces, Projects, Experiments, and Models):

1. comet-java-sdk-1.1.10 weka 3.8.6

R software (optional)

R libraries (optional):

1. caret
2. cometr

Docker Kubernetes git Java 8 (required for Chapter 9, Comet for Natural Language Processing) Apache Spark 3.1.2

You should notice that the code examples described in Chapter 4, Workspaces, Projects, Experiments, and Models, require a different version of Java with respect to those described in Chapter 9, Comet for Natural Language Processing. In addition, to make Comet work, you need to sign up to the Comet platform (https://www.comet.com/signup) and create an account.

Part 1: Getting Started with Comet

Chapter 1 - An overview of Comet

• First Use Case
• Second Use Case

Chapter 2 - Exploratory Data Analysis in Comet

• EDA Techniques

Chapter 3 - Model Evaluation in Comet

• Model Evaluation

Part 2: A Deep Dive to Comet

Chapter 4 - Workspaces, Projects, Experiments and Models

• Example in R
• Example in Java
• First Use Case Advanced
• Second Use Case Advanced

Chapter 5 - Building a Narrative in Comet

• Example 1
• Example 2

Chapter 6 - Integrating Comet into DevOps

• Comet REST API
• Docker Example
• Kubernetes Example

Chapter 7 - Extending the Gitlab DevOps Platform with Comet

• Docker Example
• Test Models

Part 3: Examples and Use Cases

Chapter 8 - Comet for Machine Learning

• Model Selection
• SHAP Value

Chapter 9 - Comet for Natural Language Processing

• Sentiment Analysis

Chapter 10 - Comet for Deep Learning

• Deep Learning

Chapter 11 - Comet for Time Series Analysis

• Basic Concepts
• Time Series Analysis

We also provide a PDF file that has color images of the screenshots/diagrams used in this book. Click here to download it.

Related products

• Applied Machine Learning Explainability Techniques [Packt] [Amazon]

• Practical Deep Learning at Scale with MLflow [Packt] [Amazon]

Get to Know the Author

Angelica Lo Duca is a researcher at the Institute of Informatics and Telematics of the National Research Council, Italy. She is also an external professor of data journalism at the University of Pisa. Her research includes data science, data journalism, and web applications. She used to work on network security, semantic web, linked data, and blockchain. She has published more than 40 scientific papers at national and international conferences and journals and has participated in many international projects and events, including as a member of the Program Committee. She is also part of the editorial team of the HighTech And Innovation Journal. She owns a personal blog, where she publishes articles on her research interests.