Machine Learning sample of use

This week has been very intense. A training trip, a five-day challenge from DataWorkshop, the first “customer”, and at the weekend a school. It’s good that I have a day off next week. It was happening and it was time to take stock of the week.

As part of 5DayChallange, I learned how to prepare and use ML models that can interpret images. Very interesting issue. It turns out that building a complicated model and firing calculations on a “strong” machine is not enough. It is then easy to “teach” the model. This means that the algorithm will perfectly recognize the graphics on which he learned, but it will not cope with new images at all. That’s not the point. You have to build a model in such a way that it recognizes the characteristics, but does not learn the images by heart. This is a difficult task. It requires a good knowledge of algorithms, how they work and how they will affect the final image of the model. In any case, I can use one ready-made algorithm to identify what is in the picture. I also know that this algorithm is great at recognizing sewing machines, worse with flowers.

As part of my ML learning, I learned how to run Web Scraping. Thanks to this, I learned that there are things implemented on my website that would have to be removed. Since I mention the settings, it turned out that you can prepare a web store for the customer, which the customer will not be able to handle. Insanity, but he met my friend so I helped. It was not easy, because the complexity of the service is high. OK, the store offers a lot of functionality, but so what if the customer is not able to handle it? I am glad that I found out about the case because it motivates me to learn. There is probably nothing better than being able to help someone in need. By the way, this situation showed me that there is still a place for companies in the Web developer market that will first talk to the client, and then adjust the solution to the skills and needs of the client. With particular emphasis on skills.

The training trip meant that I had less time to continue the Git course, but I wrote two drafts of blog entries. The first of them will appear next week, and it will talk about how he organizes his day to find time for hobbies. The second is more existential-philosophical, and I’ll probably throw it in two weeks. Going back to Git for a moment, I learned how to resolve conflicts when merging repositories. This is important because I led a repo for Jupyter’s notebooks to the first conflict. Then I gave it to myself because I did not want to stop working on the model until the conflict was resolved. We will see if after a week of work on the model it will still be possible to successfully merge in this repository.

So much for this week. To sum up, I spent on the hobby:

  • 1:30 hours to learn Python including ML
  • 6: 35 hours to learn Git

I do not count working with an ultra-complicated store, although I know that this is my mistake. It took me a few hours to figure out what was squeaking in the store. Just the way back home and a weekend full of school.

Share the Post:
Cover of book AI in Project Management

What's inside

Explore the transformative power of AI in project management with Krzysztof Nyrek’s “AI in Project Management.” This essential guide offers in-depth knowledge on automating routine tasks, optimizing resource management, and utilizing AI for risk estimation and decision-making. Learn how to implement AI tools for improved project efficiency, cost reduction, and enhanced quality management. Ideal for project managers and IT professionals, this book provides actionable insights and practical examples to harness AI technology effectively. Enhance your project outcomes and stay ahead in the digital era with this indispensable resource.

Embrace the future of project management with AI-driven solutions that save time, reduce costs, and boost project success.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.