First AI Transfer Learning Kit (TAO Version)

Added to bookmarks
Bookmarks list

Bookmark has been removed
Bookmarks list

You can't add any more bookmarks

By registering as a member, you can increase the number of bookmarks you can save and organize them with labels.
Free membership registration

last updated：Jun 16, 2022

Official site

By using pre-trained models provided by Nvidia and applying transfer learning, the implementation period can be accelerated by approximately ten times.

Using TAO, you can either deploy NVIDIA's operational quality pre-trained models as they are or fine-tune the models to suit various use cases such as video AI and conversational AI. TAO is a core component of the TAO (Train, Adapt and Optimize) platform, which features a UI-based guided workflow for AI creation. The platform uses Ubuntu and Docker and is aimed at intermediate to advanced users. Additionally, installation services are available for beginners. TAO is a tool developed by NVIDIA, which stands for Train, Adapt and Optimize. It is an AI model adaptation platform that simplifies and accelerates the development of AI applications and services for businesses. By fine-tuning pre-trained models with custom data through a UI-based guided workflow, the creation of high-accuracy computer vision, speech, and language understanding models, which previously took months, can be completed in just a few hours. No extensive training or expertise in AI is required.

Related Link - https://spectrum-tech.co.jp/products/tao.html

Inquire About This Product

basic information

Features: (1) Significantly reduced development time using Nvidia pre-trained models By utilizing pre-trained models such as Resnet and Mobilenet, we train customer data using models like FasterRCNN, SSD, Yolo, and MaskRCNN for object detection and segmentation, creating engines that lead to fast tensorRT. This can shorten the development period by approximately 10%. Additionally, training time can be reduced by 120 hours. The datasets to be used are already downloaded. (2) Providing numerous practical examples beyond object detection. In addition to object detection, we offer practical examples such as pose detection, gaze detection, facial feature detection, and license plate detection, which can also be customized using Python. (3) Easy installation with Docker The platform uses Ubuntu and sets up the TAO toolkit in a Docker environment. Since it is a container environment, installation is straightforward. It is aimed at intermediate to advanced users but can be completed in half a day. We also provide installation services for beginners. (4) Reliable support We offer a comprehensive manual and 14 days of free support.