amulet1989 Goto Github PK

autolabeling

custom_train

e_commerce-data-pipeline-for-eda

Data pipeline for an E-commerce site in Latam to analyze company data and to understand better their performance during the years 2016-2018. We focussed on two main areas Revenue and Delivery and completed an ELT(Extraction, Load, Transformation) and EDA (Exploratory Data Analysis) using SQLite, SQL, Pandas, Matplotlib, and Seaborn.

financial-advisor-chatbot-using-chat-conversational-react-description-agent

In this project, we created a chatbot acting as a financial advisor enable to answer questions related to public companies listed in NASDAQ. The bot uses a conversational Agent to coordinate chains of thoughts inserted in ChatGPT through its API, and has access to a database with around ~10.000 public financial documents.

fregate_nanoserver

gbmanalizer

This is an algorithm developed in Matlab to perform fully automatic segmentation of human Glioblastoma multiforme in Magnetic Resonance images. Pre-processed T1, T1C, T2 and Flair MRI modalities are required as described in the BraTS 2020 schedule.

gbmanalizerapp

Glioblastoma multiforme (GBM) is an aggressive brain tumor with a poor prognosis and high recurrence rate. GBManalizer, a Windows application, is introduced to automatically segment GBM and its components in magnetic resonance imaging (MRI) images. It utilizes preprocessing techniques and a neural network to achieve 83.9% accuracy.

groundingdino

Official implementation of the paper "Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection"

mapping_roi_service

microservices-architecture-using-docker-for-image-classification-with-cnn

This project develops a solution that can automatically classify images into over 1000 different categories using a Convolutional Neural Network (CNN) implemented in Tensorflow. Our solution consists of a Web UI and a Python Flask API that serves the CNN. The Web UI allows users to upload an image and receive the predicted class for that image.

mlops_wspycon2023

MLOps Workshop using Weights and Bias (Wandb) and Github Actions.

ms-former

MS-Former: Multi-Scale Self-Guided Transformer for Medical Image Segmentation (MIDL 2023)

redis_producer-consumer_example

You can see this code example to understand how to use Redis message queue and hash table to communicate two services.

re_identification_net

Un modelo Reidentification genera embeddings para identificar objetos capturados en diferentes escenas. El modelo es esencialmente un backbone de Deep Learning que toma imágenes recortadas de objetos como entrada y produce embeddings de características como salida.

sentiment-analysis-on-movies-reviews

This is a NLP project. Basically, this is a sentiment analysis problem, in this case, consisting of a classification problem, where the possible output labels are: `positive` and `negative`. Which indicates, if the review of a movie speaks positively or negatively.

tao-toolkit-triton-apps

Sample app code for deploying TAO Toolkit trained models to Triton

vehicle-classification-from-images

This is a Multi-class Classification task: we want to predict, given a picture of a vehicle, which of the possible 25 classes is the correct vehicle make-model.

vision_computer_utils