Artificial Intelligence

We collected programming problems and their solutions from previous studies. After applying some pre-processing steps, we queried advanced LLMs, such as GPT4, with the collected problems to produce machine-generated codes, while the original solutions were labeled as human-written codes. Finally, the entire collected dataset was divided into training, validation, and test sets, ensuring that there is no overlap among these sets, meaning no solutions in two different sets that solve the same programming problem.

Categories:
27 Views

Results of the Phase 1 - Competition - 11/11/2024

Competition phase of the first cut-off stage is now closed and we have gathered submissions from student, academic and industry teams from 10 countries. Submissions are evaluated and the results of the best performing teams are given below, with individual RMSSE metrics, overall average and final score. We would like to thank everyone for their participation and valuable contributions! The competition will return soon with the new cut-off stage.

Last Updated On: 
Fri, 11/15/2024 - 11:49

This dataset was developed in the context of the NANCY project and it is the output of the experiments involving cyberattacks against services that are running in a 5G coverage expansion scenario. The coverage expansion scenario involves a main operator and a micro-operator which extends the main operator’s coverage and can also provide additional services, such as Artificial Intelligence-based cyberattack detection.

Categories:
215 Views

FDA

We have created a dataset named FDA, which consists of diverse images of flexible objects collected from real-world scenarios or online sources. To the best of our knowledge, FDA is the first extensive, multi-category dataset specifically designed for the recognition of flexible objects. This dataset establishes a benchmark for evaluating the performance of models in flexible objects recognition tasks.

Categories:
56 Views

This dataset includes generated demand series, forecasts, forecast errors, and simulation results for the manuscript "Assessing the Effect of One-Step and Multi-Step Forecasting on Bullwhip Effect and Supply Chain Performance." Autoregressive demand series were generated using varying parameters, specifically based on an AR(1) demand process. Additionally, the M5 series, derived from Walmart, one of the world’s largest retailers, captures actual sales data from Walmart stores across the United States and is included as real-world data.

Categories:
102 Views

The project of paper "Massive Coordination of Distributed Energy Resource in VPP: A DRL based Multi-stage Optimization Approach"


Categories:
31 Views

Brushless DC (BLDC) motors depend on accurate rotor position detection via Hall sensors for optimal performance. Faults, such as sensor displacement, can disrupt commutation and lead to efficiency losses. This study utilizes deep learning to detect Hall sensor faults, focusing on a meticulously prepared dataset designed for this purpose. The dataset study consists of phase current measurements under various Hall sensor displacement conditions, categorized as No Delay, 0.0001 Delay, 0.005 Delay, and 0.01 Delay.

Categories:
104 Views

The Universal Networking Language (UNL) UDE Dictionary for Indian Cooking is a pioneering framework aimed at facilitating seamless communication and knowledge sharing across diverse languages and cultures, with a special focus on the rich culinary traditions of India. This dictionary provides a comprehensive and structured representation of essential culinary terms, ingredients, cooking techniques, and descriptors in Hindi, paired with their corresponding UNL equivalents. Each entry includes a UNL term, a definition and examples.

Categories:
35 Views

The Universal Networking Language (UNL) UDE Dictionary for French Cooking is a pioneering framework aimed at facilitating seamless communication and knowledge sharing across diverse languages and cultures, with a special focus on the rich culinary traditions of France. This dictionary provides a comprehensive and structured representation of essential culinary terms, ingredients, cooking techniques, and descriptors in French, paired with their corresponding UNL equivalents. Each entry includes a UNL term, a definition and examples.

Categories:
22 Views

The dataset includes five subclasses of white blood cells: eosinophils, basophils, neutrophils, monocytes, and lymphocytes. The types and positions of white blood cells in the images have been manually and accurately annotated using an annotation tool, and the annotation information files are stored in .txt, .xml, and .csv formats to meet the format requirements of different network model training for annotation files.

Categories:
136 Views

Pages