Introduction to AI with Python

Introduction to AI with Python

“Congratulations, you have reached the end of the course. Thank you very much for joining me! For the course project you will have to make a Machine Learning model that serves to classify and detect your own images, using neural networks. To get started, select which object or objects you want to classify with your classifier. It should be something that you can have lots of example images of, or you can get them from the internet (eg animals, kitchen utensils, etc). Then, get as many images as you can for the training process, and put all the images from each category in a different folder. For example, if you are going to classify pens and pencils, you should create a folder "Pens" and another "Pencils", each with their images. Leave a few separate images to use for predictions (not used during training). Follow the process similar to how we saw it in the course. I recommend trying 2 ways: Creating your own model from scratch, and another with transfer learning using an existing model, to compare the results. Remember that you can try different hyperparameters (number of layers, number of neurons) and see which one gives you the best results. Next, create your data set and separate it into Training and Test sets. Train your models. Use the data and graphs to find the best hyperparameters to use, as well as define an appropriate number of epochs to train your model without overfitting/underfitting. Finally, use your model to make predictions, you can use the images that you left separate at the beginning, or you can search for additional images on the internet, or take other photos if it is a model that you trained with your own photos. Once your project is ready, I would love for you to upload it to the forum by clicking on "Create your project". I will be giving you feedback as soon as possible. However, please be patient as it may take a while for answer. While you wait, observe the work of the rest of your classmates. I invite you to exchange ideas with them, always in a respectful way. In this way we can help each other and learn from each other. Thank you very much for completing the course, see you soon in the forum.”

Partial transcription of the video

“And with this. we reach the end of the course. I hope you enjoyed it and learned a lot. but before you leave. i want to give you some tips for your final project. Remember. you can create your own model from scratch or find and use an existing model to adapt. Try different hyperparameters. add or remove layers and neurons from your network. and measure which gives you the best results. If your model tends to overfit rapidly. consider using additional training images or applying techniques discussed in the course. such as dropout. for improvement. And in case you're struggling a bit with Pyt...”

This transcript is automatically generated, so it may contain mistakes.

Course summary for: Introduction to AI with Python

• Level: Beginner
• 98% positive reviews (415)
• 26708 students
• 6 units
• 29 lessons (4h 22m)
• 16 downloads

• Category

  Artificial Intelligence, Web & App Design

• Areas

  AI programming, Artificial Intelligence, Programming, Web Development