Neural network compression method, apparatus and device, and storage medium

1. A neural network compression method performed by a neural network compression device and used for implementing fast neural network compression with reduced computation load on the neural network compression device, comprising:

performing forward inference on target data by using a pre-trained target parameter sharing network to obtain an output feature map of a last convolutional module of the pre-trained target parameter sharing network;

extracting a channel related feature from the output feature map of the last convolutional module of the pre-trained target parameter sharing network;

inputting the extracted channel related feature and a target constraint condition into a target meta-generative network of a pre-trained target weakly supervised meta-learning framework; and

predicting an optimal network architecture under the target constraint condition by using the target meta-generative network to obtain a compressed neural network model.

2. The method according to claim 1 , wherein the pre-trained target weakly supervised meta-learning framework comprises the target meta-generative network and a target meta-evaluation network connected with the target meta-generative network; and supervised information of the target meta-generative network is from gradient information of the target meta-evaluation network.

3. The method according to claim 2 , wherein the pre-trained target parameter sharing network and the pre-trained target weakly supervised meta-learning framework are obtained by the following operations:

determining a target neural network model and an initial weakly supervised meta-learning framework, wherein the initial weakly supervised meta-learning framework comprises an initial meta-evaluation network and an initial meta-generative network;

controlling the target neural network model to perform learning at a training stage;

controlling the initial meta-evaluation network and the initial meta-generative network to perform learning at a validation stage; and

repeatedly performing the operations of controlling the target neural network model to perform learning at the training stage and controlling the initial meta-evaluation network and the initial meta-generative network to perform learning at the validation stage until a set first end condition is satisfied, so as to obtain the pre-trained target parameter sharing network and the pre-trained target weakly supervised meta-learning framework.

4. The method according to claim 2 , wherein the pre-trained target parameter sharing network and the pre-trained target weakly supervised meta-learning framework are obtained by the following operations:

determining a target neural network model and an initial weakly supervised meta-learning framework, wherein the initial weakly supervised meta-learning framework comprises an initial meta-evaluation network and an initial meta-generative network;

performing parameter sharing training on the target neural network model to obtain the pre-trained target parameter sharing network;

controlling the initial meta-evaluation network and the initial meta-generative network to perform learning at a validation stage; and

repeatedly performing the operation of controlling the initial meta-evaluation network and the initial meta-generative network to perform learning at the validation stage until a set second end condition is satisfied, so as to obtain the pre-trained target weakly supervised meta-learning framework.

5. The method according to claim 3 , wherein the initial meta-evaluation network is controlled to perform learning at the validation stage by the following operations:

generating a set of initial neural network architecture;

predicting a weight parameter of a last convolutional module of the target neural network model by using the initial meta-evaluation network according to the initial neural network architecture;

constructing a replacement convolutional module for the last convolutional module of the target neural network model by using the initial meta-evaluation network, wherein the replacement convolutional module takes a weight parameter predicted by the initial meta-evaluation network as a weight and takes input data of the last convolutional module of the target neural network model as an input;

determining a loss function using an output feature map of the replacement convolutional module; and

calculating a gradient according to the loss function by using the initial meta-evaluation network, and performing parameter update.

6. The method according to claim 5 , wherein determining the loss function using the output feature map of the replacement convolutional module comprises:

inputting the output feature map of the replacement convolutional module into a classifier of the target neural network model to obtain a classification error;

calculating a mean square error between the output feature map of the replacement convolutional module and an output feature map of the last convolutional module of the target neural network model; and

determining the loss function according to the classification error and the mean square error.

7. The method according to claim 3 , wherein the initial meta-generative network is controlled to perform learning at the validation stage by the following operations:

performing forward inference by using the target neural network model to obtain an output feature map of a last convolutional module of the target neural network model;

extracting a channel related feature from the output feature map of the last convolutional module of the target neural network model;

inputting the extracted channel related feature and a current constraint condition into the initial meta-generative network;

predicting an optimal network architecture under the current constraint condition by using the initial meta-generative network, and inputting the optimal network architecture into the initial meta-evaluation network; and

acquiring a loss function of the optimal network architecture under the current constraint condition by using the initial meta-evaluation network and backward transferring gradient information so that the initial meta-generative network performs gradient computation and parameter update on parameters of the initial meta-generative network based on the gradient information.

8. The method according to claim 2 , wherein a network architecture of each of the target meta-evaluation network and the target meta-generative network contains two fully-connected layers, and an input layer of the target meta-generative network and an output layer of the target meta-evaluation network adopt a parameter sharing mechanism.

9. The method according to claim 4 , wherein the initial meta-evaluation network is controlled to perform learning at the validation stage by the following operations:

generating a set of initial neural network architecture;

predicting a weight parameter of a last convolutional module of the target neural network model by using the initial meta-evaluation network according to the initial neural network architecture;

constructing a replacement convolutional module for the last convolutional module of the target neural network model by using the initial meta-evaluation network, wherein the replacement convolutional module takes a weight parameter predicted by the initial meta-evaluation network as a weight and takes input data of the last convolutional module of the target neural network model as an input;

determining a loss function using an output feature map of the replacement convolutional module; and

calculating a gradient according to the loss function by using the initial meta-evaluation network, and performing parameter update.

10. The method according to claim 9 , wherein determining the loss function using the output feature map of the replacement convolutional module comprises:

inputting the output feature map of the replacement convolutional module into a classifier of the target neural network model to obtain a classification error;

calculating a mean square error between the output feature map of the replacement convolutional module and an output feature map of the last convolutional module of the target neural network model; and

determining the loss function according to the classification error and the mean square error.

11. The method according to claim 4 , wherein the initial meta-generative network is controlled to perform learning at the validation stage by the following operations:

performing forward inference by using the target neural network model to obtain an output feature map of a last convolutional module of the target neural network model;

extracting a channel related feature from the output feature map of the last convolutional module of the target neural network model;

inputting the extracted channel related feature and a current constraint condition into the initial meta-generative network;

predicting an optimal network architecture under the current constraint condition by using the initial meta-generative network, and inputting the optimal network architecture into the initial meta-evaluation network; and

acquiring a loss function of the optimal network architecture under the current constraint condition by using the initial meta-evaluation network, and backward transferring gradient information, so that the initial meta-generative network performs gradient computation and parameter update on parameters of the initial meta-generative network based on the gradient information.

12. The method according to claim 1 , wherein the channel related feature is a maximum value of the output feature map on each channel, wherein a maximum value of each feature submap after a feature map tensor of N input data is split along channel C to finally form an N*C*1-dimensional feature tensor.

13. The method according to claim 1 , wherein the target constraint condition is upper and lower limits of a channel compression ratio of each layer corresponding to a Floating Point Operations Per second (FLOPs) limit or a latency limit.

14. The method according to claim 3 , wherein the initial meta-generative network and the initial meta-evaluation network are trained alternately, and the training of the initial meta-generative network depends on the initial meta-evaluation network; and the training of the initial weakly supervised meta-learning framework is performed synchronously or asynchronously with the training of the target neural network model.

15. The method according to claim 4 , wherein the initial meta-generative network and the initial meta-evaluation network are trained alternately, and the training of the initial meta-generative network depends on the initial meta-evaluation network; and the training of the initial weakly supervised meta-learning framework is performed synchronously or asynchronously with the training of the target neural network model.

16. A neural network compression device used for implementing fast neural network compression with reduced computation load on the neural network compression device, the neural network compression device comprising:

a memory, configured to store a computer program; and

a processor, when executing the computer program, configured to:

perform forward inference on target data by using a pre-trained target parameter sharing network to obtain an output feature map of a last convolutional module of the pre-trained target parameter sharing network;

extract a channel related feature from the output feature map of the last convolutional module of the pre-trained target parameter sharing network;

input the extracted channel related feature and a target constraint condition into a target meta-generative network of a pre-trained target weakly supervised meta-learning framework; and

predict an optimal network architecture under the target constraint condition by using the target meta-generative network to obtain a compressed neural network model.

17. The device according to claim 16 , wherein the pre-trained target weakly supervised meta-learning framework comprises the target meta-generative network and a target meta-evaluation network connected with the target meta-generative network; and supervised information of the target meta-generative network is from gradient information of the target meta-evaluation network.

18. The device according to claim 17 , wherein the processor is configured to obtain the pre-trained target parameter sharing network and the pre-trained target weakly supervised meta-learning framework by the following operations:

determining a target neural network model and an initial weakly supervised meta-learning framework, wherein the initial weakly supervised meta-learning framework comprises an initial meta-evaluation network and an initial meta-generative network;

controlling the target neural network model to perform learning at a training stage;

controlling the initial meta-evaluation network and the initial meta-generative network to perform learning at a validation stage; and

repeatedly performing the operations of controlling the target neural network model to perform learning at the training stage and controlling the initial meta-evaluation network and the initial meta-generative network to perform learning at the validation stage until a set first end condition is satisfied, so as to obtain the pre-trained target parameter sharing network and the pre-trained target weakly supervised meta-learning framework.

19. The device according to claim 17 , wherein the processor is configured to obtain the pre-trained target parameter sharing network and the pre-trained target weakly supervised meta-learning framework by the following operations:

determining a target neural network model and an initial weakly supervised meta-learning framework, wherein the initial weakly supervised meta-learning framework comprises an initial meta-evaluation network and an initial meta-generative network;

performing parameter sharing training on the target neural network model to obtain the pre-trained target parameter sharing network;

controlling the initial meta-evaluation network and the initial meta-generative network to perform learning at a validation stage; and

repeatedly performing the operation of controlling the initial meta-evaluation network and the initial meta-generative network to perform learning at the validation stage until a set second end condition is satisfied, so as to obtain the pre-trained target weakly supervised meta-learning framework.

20. A non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores a computer program, and the computer program, when executed by a processor in a neural network compression device, implements the following operations used for implementing fast neural network compression with reduced computation load on the neural network compression device:

performing forward inference on target data by using a pre-trained target parameter sharing network to obtain an output feature map of a last convolutional module of the pre-trained target parameter sharing network;

extracting a channel related feature from the output feature map of the last convolutional module of the pre-trained target parameter sharing network;

inputting the extracted channel related feature and a target constraint condition into a target meta-generative network of a pre-trained target weakly supervised meta-learning framework; and

predicting an optimal network architecture under the target constraint condition by using the target meta-generative network to obtain a compressed neural network model.