Method of providing neural network model and electronic apparatus for performing the same

Abstract

Disclosed is a method of controlling an electronic apparatus. The method includes receiving a trained model based on a data set and information on a target device, compressing the trained model based on compression configuring information for compression of the trained model, and providing download data corresponding to the compressed trained model so that the compressed trained model may be deployed on the target device.

What is claimed is: 
     
         1 . A method for providing a neural network model that is performed by a computing device, comprising:
 receiving, at a processor of the computing device, a trained model that has been trained based on a data set and a target device identified in a device farm using information about the target device that has been inputted by a user;   compressing the trained model based on compression configuring information and latency information received from the device farm; and   providing download data corresponding to the compressed trained model so that the compressed trained model is deployed on the target device.   
     
     
         2 . The method of  claim 1 ,
 wherein the compression configuring information includes a first compression mode indicating that the trained model is compressed based on a model compression configuring value that is configured by the user, and   wherein, when the first compression mode is configured, compressing the trained model comprises:
 identifying a plurality of compressible target blocks among a plurality of blocks included in the trained model; 
 deriving a first set of compression parameters, including block compression configuring values for block compression applied to a respective one of the plurality of target blocks, based on both the model compression configuring value and a predefined algorithm; and 
 compressing the plurality of compressible target blocks based on the first set of compression parameters. 
   
     
     
         3 . The method of  claim 2 , wherein compressing the trained model further comprises:
 providing the first set of compression parameters to the user; and   when the block compression configuring values is modified by the user, compressing the plurality of target blocks based on a second set of compression parameters including the modified block compression configuring values.   
     
     
         4 . The method of  claim 1 ,
 wherein the compression configuring information includes a second compression mode indicating that information on a block included in the trained model is provided and the trained model is compressed based on block compression configuring values configured by the user, and   wherein, when the second compression mode is configured, compressing comprises:
 identifying a plurality of compressible target blocks among a plurality of blocks included in the trained model; 
 providing information on the plurality of target blocks to the user; 
 receiving a set of third compression parameters including the block compression configuring values applied to a respective one of the plurality of target blocks, where the block compression configuring values have been configured by the user for the compression of the plurality of target blocks; and 
 compressing the plurality of target blocks based on the set of third compression parameters. 
   
     
     
         5 . The method of  claim 4 , wherein the information on the block included in the training model includes at least one of identification information of the block, a latency corresponding to the block, or a quantity of channels included in the block. 
     
     
         6 . The method of  claim 5 , wherein compressing the trained model further comprises:
 receiving a plurality of latency data from the target device,   wherein each latency data of the plurality of latency data is associated with a respective one of the plurality of blocks,   wherein each latency data of the plurality of latency data is derived by executing an associated block of the plurality of blocks by the target device.   
     
     
         7 . The method of  claim 1 , wherein the compression configuring information includes at least one of compression methods, compression configuring values, or reference information for determining a compression target among a plurality of channels included in the trained model. 
     
     
         8 . The method of  claim 1 , further comprising:
 receiving, at the processor, a user command for retraining the compressed trained model;   generating a retrained model based on the compressed trained model, and   providing download data corresponding to the retrained model.   
     
     
         9 . The method of  claim 1 , further comprising:
 performing, at the processor, at least one quantization or calibration operation on the compressed trained model based on the information about the target device.   
     
     
         10 . An electronic apparatus for providing a neural network model, comprising:
 a communication interface, configured to send and receive data via a data network, including at least one communication circuit;   a memory configured to store at least one operation instruction; and   a processor,   wherein execution of the at least one operation instruction causes the processor to:
 receive a trained model that has been trained based on a data set and a target device identified in a device farm using information about the target device that has been inputted by a user; 
 compress the trained model based on compression configuring information and latency information received from the device farm; and 
 provide download data corresponding to the compressed trained model so that the compressed trained model is deployed on the target device. 
   
     
     
         11 . The method of  claim 10 ,
 wherein the compression configuring information includes a compression mode indicating that the trained model is compressed based on a model compression configuring value that is configured by the user, and   wherein, when the first compression mode is configured, the processor is further configured to:
 identify a plurality of compressible target blocks among a plurality of blocks included in the trained model; 
 derive a first set of compression parameters, including block compression configuring values for block compression applied to a respective one of the plurality of target blocks based on both the model compression configuring value and a predefined algorithm; and 
 compress the plurality of compressible target blocks based on the first set of compression parameters. 
   
     
     
         12 . The electronic apparatus of  claim 11 , wherein the processor is further configured to:
 provide the first set of compression parameters to the user, and   when at least one of the block compression configuring values is modified by the user, compress the plurality of target blocks based on a second set of compression parameters including the modified at least one of the block compression configuring values.   
     
     
         13 . The electronic apparatus of  claim 10 , wherein the compression configuring information includes a second compression mode indicating that information on a block included in the trained model is provided and the trained model is compressed based on block compression configuring values configured by the user, and
 wherein, when the second compression mode is configured, the processor is further configured to:
 identify a plurality of compressible target blocks among a plurality of blocks included in the trained model; 
 provide information on the plurality of target blocks to the user; 
 receive a set of third compression parameters including the block compression configuring values applied to a respective one of the plurality of target blocks, where the block compression configuring values have been configured by the user for the compression of the plurality of target blocks; and 
 compress the plurality of target blocks based on the set of third compression parameters. 
   
     
     
         14 . The electronic apparatus of  claim 13 , wherein the information on the block included in the training module includes at least one of identification information of the block, a latency corresponding to the block, or a quantity of channels included in the block. 
     
     
         15 . The electronic apparatus of  claim 14 , wherein the processor is further configured to:
 receive a plurality of latency data from the target device,   wherein each latency data of the plurality of latency data is associated with a respective one block of the plurality of blocks, and   wherein each latency data is derived by executing an associated block of the plurality of blocks by the target device.   
     
     
         16 . The electronic apparatus of  claim 10 , wherein the compression configuring information includes at least one of a compression method, a compression configuring values, or reference information for determining a compression target among a plurality of channels included in the trained model. 
     
     
         17 . The electronic apparatus of  claim 10 , wherein the processor is further configured to:
 receive a user command for retraining the compressed trained model,   generate a retrained model based on the compressed trained model, and   provide download data corresponding to the retrained model.   
     
     
         18 . The electronic apparatus of  claim 10 , wherein the processor is further configured to quantize or calibrate the compressed trained model based on the information about the target device. 
     
     
         19 . The electronic apparatus of  claim 10 , wherein the processor is further configured to determine a compression configuring value of the trained model based on the latency information. 
     
     
         20 . A computer-readable recording medium on which is recorded a program that causes a computing device to execute the method of  claim 1 .