alejo-gonzalez-garcia

bayesian-analysis-formula1-circuits-rstudio-

The prediction of a pit stop depending on the circuit based on bayesian and frequentist techniques

deep-autoencoder-based-on-dense-neural-networks

In this project we have developed a Deep Autoencoder using Dense Neural Networks to perform dimensionality reduction on MNIST and FMNIST datasets. The project includes training, saving, and evaluating models using PyTorch. Utilized the Weight & Biases library for monitoring and comparison of model performance

electric-gas-fuel-type-consumption-statistical-learning-rstudio

Analysis of fuel type consumption, electric vehicles, including Factor Analysis, Hierarchical Clustering, PCA and Feature Engineering

f1-material-transportation-with-minimum-cost-flow-problem-non-linear-optimization

The FIA has to satisfy the demand of materials for each race week but trying to minimize the cost. To do so, we are solving a Minimum Cost Flow Problem applying optimization. On the second part, we are solving a simple Non-Linear problem, where we minimize the Drag Force for some given parameters.

linear-optimization-of-f1-wind-tunnel-budget-assigment-of-the-fia

On this optimization problem we are the FIA and we have optimized the budget assignation that the F1 has to do each season by using Wind Tunnel data. Depending on the cost of each piece and the time improvement on each of the cars, we must assign more or less money to maximize the performance.

microprocessors-autonomous-robot-stm32l

We have developed an autonomous wheeled robot using STM32L, that is capable of traveling through the floor avoiding obstacles and varying its speed. We can control the speed at which it travels and fully control de robot from our mobile/pc

optimization-problem-with-numerical-methods

we have solved an optimization problem applying Composite Trapezoid Rule and Gradient Descent methods

reinforcement-learning-pac-man

In this project we have developed machine learning techniques to allow the pacman to learn and improve it´s performance to its maximum capabilities, in different maps, with different speeds and achieving really good results

route-lookup-switching

Next Hop Search Problem solved by implement Route Lookup algorithm based on a multi-bit tree

shared-links-statistical-learning-rstudio

Target is to predict the number of shares of an article in social media. Data Visualization, EDA, LDA, QDA, Naive Bayes, Random Forest, Gradient Boosting, Ensembled Learning, Outliers Removal,

sinking-titanic-eda-svm-rf

Exploratory work about the sinking of the Titanic regarding embarkation port and survival rate

text-preprocessing-vectorization-and-classification-applying-nlp

We have performed a multi-class classification task of literary poems, which will be assigned to a period. Raw data has been collected from the web and processed the in order to apply Natural Language Processing and Machine Learning tools, such as feature extraction and selection, topic modeling, text preprocessing and classification

understanding-calibration-in-cnns

Explored calibration in Convolutional Neural Networks (CNNs) using the CIFAR-10 dataset, focusing on binary classification of birds and cats. The project encompasses data preprocessing, model training, and evaluation, with a deep dive into calibration techniques. Weight & Biases library for monitoring training processes and model performance.

zoo-web-page

In this project we have created a zoo website where you can reserve activities, buy tickets and see the animal catalogue. AI has been used to generate all the images shown.