Peter

ML Zoomcamp 2023 – Deep Learning – Part 10

Adding more layers It’s possible to add more inner layers after the vector representation. Previously we had one inner layer before outputting the prediction. This layer does some intermediate processing of the vector representation. These inner layers make the neural network more powerful. Adding one inner dense layer Usually adding one or two additional layersContinue reading “ML Zoomcamp 2023 – Deep Learning – Part 10”

ML Zoomcamp 2023 – Deep Learning – Part 9

Checkpointing Checkpointing is a way of saving our model after each iteration or when certain conditions are met, f.e. when the model achieves the best performance so far. This is a nice way because when the model starts to oscillate, the model after 10 iterations is not necessarily the best one. Saving the best modelContinue reading “ML Zoomcamp 2023 – Deep Learning – Part 9”

ML Zoomcamp 2023 – Deep Learning – Part 8

Adjusting the learning rate What’s the learning rate ? Let’s use an analogy. Imagine that learning rate is how fast you can read and you read one book per quarter. That means you can read 4 books per year. Somebody else can read 1 book per day, so he can read many books per year.Continue reading “ML Zoomcamp 2023 – Deep Learning – Part 8”

ML Zoomcamp 2023 – Deep Learning – Part 7

This is the second part of Transfer Learning section. While in the last article we looked at reading image data this article covers the training part. Transfer Learning – Part 2/2 Train Xception on smaller images (150×150) (Better to run it with a GPU) So far for reading the data, now let’s train a model.Continue reading “ML Zoomcamp 2023 – Deep Learning – Part 7”

ML Zoomcamp 2023 – Deep Learning – Part 6

This part is devided into 2 sections. This article is about the first part where we want to read image data using the ImageDataGenerator. The second article covers the second part where we want to train an Xception model. Transfer Learning – Part 1/2 This time we want to use an already trained network thatContinue reading “ML Zoomcamp 2023 – Deep Learning – Part 6”

ML Zoomcamp 2023 – Deep Learning – Part 5

This part is about Convolutional Neural Networks (CNN). In the last part we already touched this topic while we were using a pre-trained neural network. This part is a bit more about the theory. Convolutional neural networks Convolutional neural networks are mostly used for images and they consists of different kind of layers. Let’s imagineContinue reading “ML Zoomcamp 2023 – Deep Learning – Part 5”

ML Zoomcamp 2023 – Deep Learning – Part 4

In this part we’ll use a pre-trained convolutional neural network to understand what is on the image we load previously. Pre-trained convolutional neural networks This time we want to take an image and an off-the-shelf neural network that was already trained by somebody, so we can use it. Now we want to use a specialContinue reading “ML Zoomcamp 2023 – Deep Learning – Part 4”

ML Zoomcamp 2023 – Deep Learning – Part 3

TensorFlow and Keras TensorFlow and Keras are powerful tools in the field of deep learning. TensorFlow, an open-source machine learning library developed by Google, provides a comprehensive platform for building and deploying machine learning models. It facilitates the creation of complex neural network architectures and supports both training and inference across various devices. Keras, onContinue reading “ML Zoomcamp 2023 – Deep Learning – Part 3”

ML Zoomcamp 2023 – Deep Learning – Part 2

This part is not discussed in the course ML Zoomcamp, but it could be useful how to setup GPU support for your local machine. How to find the right Cuda version? First you need to know which CUDA Toolkit works with your NVIDIA graphics adapter. The easiest way to find out the supported version isContinue reading “ML Zoomcamp 2023 – Deep Learning – Part 2”

ML Zoomcamp 2023 – Deep Learning – Part 1

What is Deep Learning and why it is important for this chapter? Deep learning is a subfield of machine learning that involves neural networks with multiple layers (deep neural networks). It’s particularly powerful in handling complex tasks such as image processing, speech recognition, natural language processing, and more. Handling images is exactly what we willContinue reading “ML Zoomcamp 2023 – Deep Learning – Part 1”