Deci, a deep learning company that uses artificial intelligence to develop AI, recently announced the launch of DataGradients. This open-source tool is designed to profile computer vision datasets and find essential insights.
In computer vision, the effectiveness of an AI model is deeply connected to the quality of the training data. Therefore, identifying any issues within the dataset is vital as it not only aids practitioners in avoiding training obstacles but also highlights potential reasons for a model’s underperformance, Deci explained.
According to the company, a comprehensive understanding of a dataset’s attributes can simplify crucial decisions such as choosing the most suitable model, the best loss function, and the optimal optimization method.
“DataGradients empowers data scientists to streamline their model development and training process, with crystal-clear visibility into their data. With DataGradients, we’ve made it that much easier to extract actionable insights from one’s datasets,” said Yonatan Geifman, CEO and co-founder of Deci. “DataGradients marks our third tool released as open source to the benefit of the wider AI community, following our launch of SuperGradients, our free, open-source training library for PyTorch-based deep learning models, and YOLO-NAS, our groundbreaking object detection foundation model.”
DataGradientsallows data scientists to quickly analyze the integrity of their data using just a single line of code. This tool can swiftly identify issues such as corrupted data, distribution shifts between training and testing sets, and duplicate annotations, among others.
Users are then provided with actionable insights on how to proactively address these issues, which helps to streamline their model design and training processes. This, in turn, ensures optimal performance and reliable results.
To begin profiling your data or start training models, you can visit DataGradients on Deci’s GitHub repository.