US 12,079,600 B2
Visual programming for deep learning
Haoxiang Lin, Redmond, WA (US); Mao Yang, Redmond, WA (US); Shuguang Liu, Redmond, WA (US); and Cheng Chen, Redmond, WA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Appl. No. 17/615,080
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
PCT Filed May 6, 2020, PCT No. PCT/US2020/031544
§ 371(c)(1), (2) Date Nov. 29, 2021,
PCT Pub. No. WO2020/263421, PCT Pub. Date Dec. 30, 2020.
Claims priority of application No. 201910578856.X (CN), filed on Jun. 28, 2019.
Prior Publication US 2022/0222049 A1, Jul. 14, 2022
Int. Cl. G06F 8/34 (2018.01); G06F 3/0486 (2013.01); G06N 3/082 (2023.01)
CPC G06F 8/34 (2013.01) [G06F 3/0486 (2013.01); G06N 3/082 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
presenting a visual representation of an artificial neural network, wherein the visual representation of the artificial neural network includes graphical elements representing layers of the artificial neural network;
in response to receiving a drag-and-drop operation on the graphical elements, modifying an intermediate representation of the artificial neural network, wherein the intermediate representation of the artificial neural network is independent of a deep learning framework and the drag-and-drop operation is configured to modify connections between the graphical elements including automatically connecting a first graphical element to a second graphical element of the graphical elements;
modifying, based on the intermediate representation of the artificial neural network, code of the artificial neural network for a target deep learning framework; and
in response to receiving an instruction for changing the target deep learning framework to a further target deep learning framework, determining code of the artificial neural network for the further target deep learning framework based on the intermediate representation of the artificial neural network.