- π Iβm a student specialized in Machine Learning and Applied Mathematics.
- π‘ Iβm interested in research in AI (ML/DL) but also Data Science projects.
- Email: [email protected]
- LinkedIn: Marijan Soric
- Website/Portfolio: Marijan's Website
- π€ Hugging Face: Marijan Soric
Here are some of the projects I'm currently working on or have contributed to:
- LLM-text-generation-work: LLMs survey. Transformer architecture with BERT, GPT. Fine-tuning method (LoRA).
- LLM-based-App: Creation of an app based on open source LLM that includes: chatbot, document chat and summarizer.
- TDT4265_StarterCode_2024: Exercices from Deep Learning Course at NTNU.
- Softmaxregression on the MNIST dataset.
- Multi-layer neural network on the MNIST dataset.
- Implementing CNN to perform image classification of images in the CIFAR10 dataset.
- DiffusionModel: Implement the architecture from the paper Denoising Diffusion Probabilistic Models. Car image generator. Demo App on π€ Hugging Face Space "Cogito Diffusion Model"
- RealEstate: Supervised learning on real estate prices with a dataset from Institut Louis Bachelier.
- NTNU-ML-course: Implementing Logistic Regression and K-means from scratch. Exercices from Machine Learning Course at NTNU.
- Time-Series-Forecasting-Solar-Energy-Production: Kaggle competition to predict solar production of pannels for ANEO.
- PAi23: FDIR. Unsupervised learning on flight detection: anomaly detection.
- Hopfield-Network-Attractor: Discuss about discrete attractors in Hopfield Network. Essay from Neural Network Course at NTNU (Neuroscience faculty).
- CatMatch: Create a recommendation system content based filtering for the CatMatch app.
- School-Project: Projects done at Centrale Lyon: Image Compression, Automatic Change, Hangman Game, Hotel Management, Website Listing the Bridges of the World.