End-to-End Machine Learning Leveraging Cloud Service Providers : A Comparison of AWS and GCP
Mäkelä, Roope (2022)
Mäkelä, Roope
2022
Bachelor's Programme in Science and Engineering
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2022-05-10
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202204263698
https://urn.fi/URN:NBN:fi:tuni-202204263698
Tiivistelmä
Cloud service providers give users resources as service products through the internet. Using cloud services has many benefits compared to physical systems. The high scalability means services automatically adjust to the demand of the user. Within machine learning, where data sets are very large and can change from project to project, the scalability of cloud services is particularly advantageous.
This thesis examines how the user can leverage specific services from cloud service providers in order to implement an end-to-end machine learning project. To achieve this, a theoretical example with Google Cloud Platform is conducted. Mainly the capabilities of the services are examined, without going deep into the technology behind them.
The thesis compares two cloud service providers for implementing a machine learning project: Amazon Web Services and Google Cloud Platform. The comparison focuses on the different services provided for each stage of implementation and how the capabilities of the services differ across the platforms. The comparison is made using existing research and information from the websites and product offerings of the respective platforms.
The comparison shows that neither platform is universally the better choice. The decision will largely depend on the specific project and what factors are most important to the user. While Amazon generally excels in machine learning algorithms and incorporating external models, Google offers a larger amount of relevant services across the implementation process. In the end, the platforms are ranked in terms of individual factors, but a general overall choice cannot be made.
This thesis examines how the user can leverage specific services from cloud service providers in order to implement an end-to-end machine learning project. To achieve this, a theoretical example with Google Cloud Platform is conducted. Mainly the capabilities of the services are examined, without going deep into the technology behind them.
The thesis compares two cloud service providers for implementing a machine learning project: Amazon Web Services and Google Cloud Platform. The comparison focuses on the different services provided for each stage of implementation and how the capabilities of the services differ across the platforms. The comparison is made using existing research and information from the websites and product offerings of the respective platforms.
The comparison shows that neither platform is universally the better choice. The decision will largely depend on the specific project and what factors are most important to the user. While Amazon generally excels in machine learning algorithms and incorporating external models, Google offers a larger amount of relevant services across the implementation process. In the end, the platforms are ranked in terms of individual factors, but a general overall choice cannot be made.
Kokoelmat
- Kandidaatintutkielmat [7052]