This article was co-written by @sean_mealin and @augustazhang.    SAS Visual Data Mining and Machine Learning makes it easy for users to build deep learning models that uncover insights from all types of data, including images, text, audio, and time series data. You may be familiar with the deepLearn action set, which supports training and deploying deep neural networks, recurrent neural networks, and convolutional neural networks in both SAS and open-source languages.    Now, users have access to even more deep learning capabilities, thanks to the new deep learning model zoo (dlModelZoo) action set available with the SAS Viya 2022.09 stable release.    Using the new action set, customers can take advantage of more powerful models as well as use models compiled to TorchScript. The new action set brings the power of PyTorch into the Viya environment, while hiding the messy mechanics of deep learning behind the scenes.    What does the new action set do?    Just like the deepLearn action set, dlModelZoo allows users to both train and score deep learning models. Customers can use models out-of-the-box, or they can use transfer learning to adapt a pretrained model to their own data.    Why should I use the new action set?    The dlModelZoo action set includes all the models you know and love from the deepLearn action set, plus some new models that are even more powerful. New models include YOLO V5, ENet, and more. Classic models such as ResNet, VGG, and UNet are also available.    Unlike other action sets, dlModelZoo uses PyTorch models, allowing third-party models to be imported and used. That means it is possible to build a model in PyTorch, convert it to a TorchScript representation, and import it to Viya for training and scoring.     What’s happening to the deepLearn action set?     The deepLearn action set is still supported. While it will receive critical updates and bug fixes, the newer models will be exclusively available using the dlModelZoo action set. Models built using the dlModelZoo action set cannot be used with other action sets, and vice versa.     What about DLPy?     Currently, DLPy only supports the deepLearn action set, and this support will be maintained for the foreseeable future. But don’t worry! Over the next few months, it will receive updates to work with the dlModelZoo action set, too. That means you’ll be able to access all the new deep learning capabilities from Jupyter notebooks.    Where can I learn more about dlModelZoo?     The dlModelZoo documentation is a good place to learn more and see some examples. Over the next couple of months, additional blog posts and other content will be coming out, which will take deep dives into the new action set. Stay tuned!     ... View more

The market for artificial intelligence is ballooning, and breakthrough innovations that once seemed abstract and alluring are now materializing at incredible speed. But even as these technologies grow increasingly sophisticated, we have the opportunity at SAS to make them more accessible than ever.   One of the key efforts of the Americas Artificial Intelligence team is to get these technologies into the hands of more SAS employees, enabling them to build AI applications to solve their customers’ unique business problems. We work to democratize the field of AI and to show that the technology doesn’t have to be intimidating, even for newer users.   Recently, we hosted two weeklong AI Hackathon events for associates in the SAS Customer Advisory Academies, allowing them to get hands-on with the technologies that SAS has to offer. For the first half of the week, members of our team provided instruction and training to associates, teaching them how to create applications with six different types of AI technology. Then, associates were challenged to team up, build their own applications in just one day, and present their work to a panel of judges.   The AI technologies and respective instructors were as follows: Optimization – @paugre  Natural Language Processing – @Sundaresh and me (@augustazhang in the communities) Computer Vision – @neela Anomaly Detection – @Capt_VA_SAS Recommendation Engines – @Capt_VA_SAS and @paugre Network Analysis and Network Optimization – @carpdr  Neela Vengateshwaran leads a hands-on exercise in creating an object detection model        Customer advisory associates collaborate with team members to build their own AI solutions in one day      Associates present their applications to colleagues and a panel of judges on the final day of the AI HackathonThough AI technologies can be used for a broad range of functions—sometimes involving sensitive issues like bias and employment, other times simply frivolous or downright silly—associates were challenged to meet specific criteria with their solutions.   They were tasked to create applications that weren’t only flashy and fun, but also demonstrated true business value to organizations. At the same time, they needed to keep their solutions relatively non-controversial and reasonable enough to complete within a single day. In other words, applications were to be tangible in value, innocuous in scope, and feasible in production.   The Customer Advisory associates’ efforts yielded some seriously impressive results. Here’s a look at the winning team projects from the last two AI Hackathons:   Customer Advisory Academy V: September 30 – October 4, 2019   First Place   Xiasi Liu, Ina Conrado, Andrew Gu, Isabella Hewitt AI Technologies: Text Analytics and Recommendation Engine Solution: Associates created a recommendation engine that could be used by a company selling candy subscription boxes. By analyzing customer reviews on different products, the solution learns client preferences and recommends which types of sweet treats should be included in their personalized assortments. Hackathon Highlights Monday article: SAS AI delivers Sly Upsells for SASsy Treats   Second Place   Neal Vaidya, Spain Niemer, Jordan Burns, Pauline Ranjan (not pictured) AI Technologies: Optimization Solution: This team’s AI application used demographic data to suggest the optimal way for presidential candidates to allocate campaign funds across different states. Hackathon Highlights Monday article: How we can use SAS AI to allocate political campaign funds more efficiently   Third Place   Xin Ru Lee, Yen Nguyen, Jordan Mowlai, Tyler Gillson AI Technologies: Text Analytics Solution: This team’s solution analyzed Yelp and Google reviews for local attractions, such as parks and restaurants, in order to identify the hottest potential spots for a hotel chain to expand with new locations. Hackathon Highlights Monday article: AI and the role of sentiment in location optimization   Customer Advisory Academy VI: November 4 – November 8, 2019   First Place   Ivan Nicolas Gutierrez Arias, Gonzalo Ulla, Busra Ozaydin, Johannes Pretsch, Antonie Berkel AI Technologies: Computer Vision Solution: This team created an AI-powered safari ranger. When a guest at the safari takes a photo of an animal, such as an elephant or a zebra, the application uses computer vision to identify its species and to provide useful facts about it. This data is then collected by the park to gain an understanding of which animals most interest their guests. Hackathon Highlights Monday article: How we built a safari park app with SAS and open source   Second Place   Left to right: Sam Amara, Kamilia Oukaci; Far right: Bartosz Koziarski AI Technologies: Text Analytics Solution: Associates created a solution to help a bank handle comments and concerns expressed by customers using their online system. The application uses text analytics to automatically analyze customer comments, identify sentiments, and offer appropriate solutions. Hackathon Highlights Monday article: Banks applying AI to transform customer complaints into business success   Third Place   Lukkas Hamann, Terence Timbang, Tahseen Badat, Elif Saracoglu AI Technologies: Optimization Solution: This team created a solution to optimize shipping logistics, automatically identifying the quantities of goods that should be transported from various warehouses to a number of retail locations, and the most efficient routes to do so. Hackathon Highlights Monday article: How we used SAS AI to Optimize Shipping Logistics   Hackathon Highlights Monday article series debuts   We are also excited to announce that six upcoming articles, written by the talented teams above, will be published in a new series called Hackathon Highlights Monday! Every Monday, one team will share more about their experiences during the AI Hackathon and provide us with details on the applications that they created.   Check back on Monday, January 20, 2020 for the first post! Even better, get email notifications when posts go live by subscribing to the Artificial Intelligence label in the SAS Communities Library:   Expand the "Find A Community" menu in the right nav. Select the SAS Communities Library at the bottom of the list. In the Labels box, click Artificial Intelligence. Select "Subscribe" from the Options menu. Don't forget to congratulate the teams and offer your thoughts in the comments of each post! ... View more