IBM today announced the launch of the new proletterof Deep Learning as a Service (DLaaS) for AI developers.
Με το Deep Learning as a Service (DLaaS), οι προγραμματιστές θα μπορούν να εκπαιδεύσουν νευρωνικά δίκτυα χρησιμοποιώντας δημοφιλή frameworks όπως το TensorFlow, το PyTorch και το Caffe χωρίς να αγοράζουν και να διατηρούν δαπανηρό material.
The service allows data scientists to train their models using only e.gconditions they need, paying only for the duration of GPU usage.
Each unit exwork cloud database is built for ease of use and is ready for programming deep learning networks without the need to manage the infrastructure users. In accordance with white paper published by IBM researchers:
Users can choose a set of supported deep learning frameworks, a neural network model, training data, and cost constraints. The service will then take care of the rest, providing them with an interactive, repetitive AI training experience.
For users to use them services θα πρέπει απλά να προετοιμάσουν τα δεδομένα τους, να τα ανεβάσουν, για να ξεκινήσουν την learning. They can then download the learning outcomes of their application.
It looks pretty simple and could possibly save weeks programming, according to TNW.
IBM is reportedly working to address the difficulty eqexcerciseς νευρωνικών δικτύων ή τουλάχιστον για να μειώσει τον χρόνο ανάπτυξης. Σύμφωνα με μια δημοσίευση στο blog the company's:
This deep learning as a service is an experimental learning environment which means that users do not have to worry about programming and problem management. The entire lifecycle of learning is automatically managed and results can be viewed in real time and reviewed later. Every educational start is automatically started, monitored and stopped after it is completed, saving time and money for users as they can only pay for the resources they use.
The new Deep Learning as a Service (DLaaS) works with the excellent platform of Watson. This means it was tested on one of the most advanced AI systems on the planet.
For more information, you can see it IBM blog.
