markmartorilopez,Mark Martori Lopez,github

👋 Hi, I’m @markmartorilopez
👀 I’m interested in how Machine Learning tackles client's needs with a direct outcome, mainly in Computer Vision and NLP.
📁 Recent Projects in 2022:
- Automatize the creation of on-demand chemical reactions data sets using Vision Transformers and OCR.
- Object detection and classification benchmark with Vision Transformers, Faster-RCNN, RetinaNet, YOLOv4 and feature detectors.
- Creation of faces that do not exist with a Progressive-Growing GAN (Team Project).
- Playground for Deep Reinforcement Learning experiments, mimicking behavioural tasks in neuroscience.
🌱 In my most recent project, as a ML Scientist I learnt the in-depth concepts of Vision Transformers in Deep Learning with IBM Research under the supervision of Dr.Teodoro Laino in the AI for Scientific Discovery Team. I developed the digital tool "OchemR", check it's usability out at https://rxn.res.ibm.com and a demo of an early-staged code in my github.
💞️ I am looking to collaborate in enthusastic projects concerning Computer Vision and NLP.
📫 How to reach me: LinkedIn -> markmartorilopez

Mark Martori Lopez's Projects

asvz_database_sql

Akademische Sportverband Zürich Database simulation using both Oracle and PostgreSQL with Schema design and relationships creation.

bigdataanalytics_r

Big Data Analysis to Cancer Prediction Dataset - Gene Expression.

detr

End-to-End Object Detection with Transformers

dl_recognition_detection

Basics of Deep Learning to get hands-on experience on a real example.

ds_structure

Main structure to follow when Long-Term Project in place. ( Production, Staging, Development )

ellis_ml4moleculediscovery_manuscript

Poster presented at ELLIS Workshop - Machine Learning for Molecule Discovery.

machine_learning_notebooks

Personal Machine Learning Notebooks when following courses in Coursera, Udemy or at IBM.

From a chemical reaction image, detect and classify molecules, text and arrows by using the Vision Transformer DETR. Comparisons with well-established CNNs (RetinaNet and FasterRCNN also provided). The detections are then translated into text "OCR" or into SMILES. The direction of the reaction is learned and preserved into the output files.