On December 3, qeexo Qishou released its new automl product today, which is a one click fully automated platform that allows users to quickly build machine learning solutions using sensor data on edge devices.
According to reports, automl will fill the market gap of fully automated machine learning platform for embedded edge devices.
At present, the vast majority of machine learning manufacturers focus on providing services for raspberry PI and above devices, while qeexo automl focuses on ARM cortex & trade; -M0 to M4 level MCU, it is understood that this product has supported st STM semiconductor SensorTile.box , a compact multi-sensor module including cortex-m4 MCU, will be expanded to more hardware platforms in the future.
"Countless companies are constantly collecting a lot of data from the edge. They want to be able to use machine learning, but they can't find the right tools or technical teams. With qeexo automl, they can only spend a small amount of time and resources compared with the normal situation, and they can generate production models through continuous iterations of prototypes and projects 。 Qeexo's preference for supporting arm based MCU products is due to arm's commitment to building a world-class ecosystem and its leading position in marginal markets. '
Qeexo automl automates data preprocessing, feature extraction, model selection, optimization of super parameters, validation results, and deployment of models, which require a lot of repetitive work in traditional machine learning processes. It will undoubtedly have a positive impact on solving the shortage of machine learning engineers. Not only that, for machine learning experts, it can save a lot of time and improve the efficiency of the core development work by handing over the tedious and error prone process to the platform.
Even for enterprises that are not familiar with machine learning but have massive sensor data, qeexoautoml can completely replace the role of machine learning Engineer in some specific tasks, so as to save the time and cost of establishing machine learning team and reduce the risk of enterprise trial and error.
According to the data, qeexo Qishou was born in Carnegie Mellon University and is the first company to provide automatic end-to-end machine learning services for embedded edge devices (cortex m0-m4). The company has acquired venture capital and has offices in mountain view, Pittsburgh, Shanghai and Beijing.