⚠️ Lab under construction! ⚠️
Welcome to my lab! Well, maybe it is a bit pretentious to call it a lab... for now, let's leave it as a playground. I will use it to document my journey through Deep Learning and its applications in VR/AR, computer graphics, BCI, etc. (You can find more about my interests in the AboutMe section). I am learning (I hope I always will be), so these posts aim to help me consolidate my knowledge and share my work with the community in case anyone can help me or, even better, it can be useful for someone. There is no point in keeping everything I learn for myself, so let's explore together!
Posts
First post on the NLP with Transformers series. We will apply the concepts of Chapter 2 to a different problem. Will DistillBert be capable of distinguishing between a paraphrase and a sentence with a different meaning?
Oct 30, 2022
This is a version of the codelab from chapter 3 of the Efficient Deep Learning book which covers Data Augmentation for vision applications. In the original PyTorch version we simply train the model for 100 epochs. Although, when using Fastai we normally train the models in a different way. Intrigued?
Jun 9, 2022
While trying to implement the previous post about quantization I faced some difficulties. So it is better to write them down for everyone to know (including a future me). Also, I think the process of solving these issues may apply to other problems. Without further do, time to fix🔨!
May 30, 2022
Quantization is one of the most popular techniques when compressing a Neural Network. This post shows how to perform Post Training Quantization with PyTorch. It is based on one of the codelabs from the Efficient Deep Learning book, which uses TensorFlow. If you are more confortable using PyTorch, now you have no excuse, let's quantize🐜!
May 29, 2022
This is the second part of the post ResNet for Traffic Sign Classification With PyTorch, in which we will try to get close to the results of the previous model but with a smaller one. To do so we will apply the knowledge distillation technique, want to know more? Let's dive in!
Apr 24, 2022
This is my implementation of an existing blog post on Traffic Sign classification. The previous work done by Pavel Surmenok was based on Fastai v1, so I have tried to update it to Fastai v2. If you are interested in computer vision in autonomous driving, and you want to learn more about Fastai this woul be a great starting point!🚗
Apr 21, 2022
First post explaining my plans for this blog.
Apr 18, 2022