METHODS AND SYSTEMS FOR REDUCING MEMORY FOOTPRINTS ASSOCIATED WITH CLASSIFIERS
First Claim
1. A method for reducing the required footprints associated with classifiers, said method comprising:
- employing classifier weights from a training cycle to rank features utilizing particular criteria;
selecting at least one top feature among said features; and
repeating said training cycle using only said at least one top feature in order to thereafter determine if additional features among said features should be included or excluded to reduce said footprints associated with said classifiers.
4 Assignments
0 Petitions
Accused Products
Abstract
Methods and systems for reducing the required footprint of SNoW-based classifiers via optimization of classifier features. A compression technique involves two training cycles. The first cycle proceeds normally and the classifier weights from this cycle are used to rank the Successive Mean Quantization Transform (SMQT) features using several criteria. The top N (out of 512 features) are then chosen and the training cycle is repeated using only the top N features. It has been found that OCR accuracy is maintained using only 60 out of 512 features leading to an 88% reduction in RAM utilization at runtime. This coupled with a packing of the weights from doubles to single byte integers added a further 8× reduction in RAM footprint or a reduction of 68× over the baseline SNoW method.
-
Citations
20 Claims
-
1. A method for reducing the required footprints associated with classifiers, said method comprising:
-
employing classifier weights from a training cycle to rank features utilizing particular criteria; selecting at least one top feature among said features; and repeating said training cycle using only said at least one top feature in order to thereafter determine if additional features among said features should be included or excluded to reduce said footprints associated with said classifiers. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 18)
-
-
9. A system for reducing the required footprints associated with classifiers, said system comprising:
-
a processor; a data bus coupled to said processor; and a computer-usable medium embodying computer program code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said processor and configured for; employing classifier weights from a training cycle to rank features utilizing particular criteria; selecting at least one top feature among said features; and repeating said training cycle using only said at least one top feature in order to thereafter determine if additional features among said features should be included or excluded to reduce said footprints associated with said classifiers. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16)
-
-
17. A processor-readable medium storing code representing instructions to cause a process for reducing the required footprints associated with classifiers, said code comprising code to:
-
employ classifier weights from a training cycle to rank features utilizing particular criteria; select at least one top feature among said features; and repeat said training cycle using only said at least one top feature in order to thereafter determine if additional features among said features should be included or excluded to reduce said footprints associated with said classifiers. - View Dependent Claims (19, 20)
-
Specification