Methods and systems for post-training quantization of neural networks
OwnerYuqiao WEN, Yanshuai CAO +1inventors
InventorsYuqiao WEN, Yanshuai CAO, Lili MOU
Life of this patent
Estimating the term…
Abstract
Disclosed are methods and systems for post-training quantization of a neural network. The systems comprise computer processing and memory means for carrying out the methods. The methods comprise sequentially selecting one of a plurality of layers in the neural network. The methods also comprise, for each selected layer, sampling a transformation associated with the selected layer, quantizing the neural network using the sampled transformation and calculating a loss associated with the quantization. The methods also comprise determining the acceptability of the calculated loss using a discrete search algorithm and accepting the sampled transformation if the loss is determined to be acceptable.
Claims19
A method of post-training quantization of a neural network, the method comprising:
- sequentially selecting one of a plurality of layers in the neural network;
- and
- for each selected layer:
- sampling a transformation associated with the selected layer;
- quantizing the neural network using the sampled transformation;
- calculating a loss associated with the quantization;
- determining the acceptability of the calculated loss using a discrete search algorithm;
- and
- accepting the sampled transformation if the loss is determined to be acceptable.
A system for post-training quantization of a neural network, the system comprising:
- at least one database having stored thereon a neural network;
- at least one processor communicatively coupled with the at least one database;
- and
- at least one memory communicatively coupled to the at least one processor, the at least one memory having stored thereon computer program code that is executable by the at least one processor and that, when executed by the at least one processor, causes the at least one processor to perform a method comprising:
- sequentially selecting one of a plurality of layers in the neural network;
- and
- for each selected layer:
- sampling a transformation associated with the selected layer;
- quantizing the neural network using the sampled transformation;
- calculating a loss associated with the quantization;
- determining the acceptability of the calculated loss using a discrete search algorithm;
- and
- accepting the sampled transformation if the loss is determined to be acceptable.
At least one non-transitory computer readable medium having stored thereon computer program code that is executable by at least one processor and that, when executed by the at least one processor, causes the at least one processor to perform a method comprising:
- sequentially selecting one of a plurality of layers in the neural network;
- and
- for each selected layer:
- sampling a transformation associated with the selected layer;
- quantizing the neural network using the sampled transformation;
- calculating a loss associated with the quantization;
- determining the acceptability of the calculated loss using a discrete search algorithm;
- and
- accepting the sampled transformation if the loss is determined to be acceptable.
Dependencies are inferred from the claim text.