Deep learning training method, device and equipment and readable storage medium
A deep learning and training set technology, applied in the field of deep learning, can solve the problems of difficult sample collection, difficult to obtain label samples, and high cost of label analysis, and achieve the effect of enriching training data and good effect.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0037] figure 1 It is a flow chart of the deep learning training method provided by Embodiment 1 of the present invention; figure 2 It is a schematic diagram of the overall flow of the deep learning training method provided by Embodiment 1 of the present invention. The embodiment of the present invention aims at the problem that in some fields, due to the difficulty of sample collection and the high cost of label analysis, it is usually difficult to obtain label samples, the lack of label samples, and the serious problem of small samples, resulting in the poor effect of the trained deep model. Provide deep learning training method.
[0038] like figure 1 and figure 2 As shown, the specific steps of the method are as follows:
[0039] Step S101. Perform model training on multiple small-sample training sets randomly selected from the training data set to obtain model parameters of each small-sample training set.
[0040] Firstly, the source domain training data set is obt...
Embodiment 2
[0054] image 3 It is a flow chart of the deep learning training method provided by Embodiment 2 of the present invention. On the basis of the first embodiment above, in this embodiment, a plurality of small sample training sets randomly selected from the training data set include: grouping the source domain training data set to obtain multiple training data sets; The process of extracting a preset number of training data from multiple training data groups respectively to obtain multiple small-sample training sets. like image 3 As shown, the specific steps of the method are as follows:
[0055] Step S201, grouping the source domain training data sets to obtain multiple training data groups.
[0056] Firstly, the source domain training data set is obtained, and data preprocessing is performed on the source domain training data set to obtain the preprocessed source domain training data set. In order to be suitable for deep learning training with a learning strategy, the pre...
Embodiment 3
[0105] Figure 5 It is a schematic structural diagram of the deep learning training device provided by Embodiment 3 of the present invention. The deep learning training device provided in the embodiment of the present invention can execute the processing flow provided in the embodiment of the deep learning training method. like Figure 5 As shown, the deep learning training device 30 includes: a training module 301 , a parameter updating module 302 and a verification module 303 .
[0106] Specifically, the training module 301 is used to perform model training on a plurality of small sample training sets randomly selected from the training data set to obtain model parameters of each small sample training set.
[0107] The parameter update module 302 is used to update the initial parameters of the source domain model according to the model parameters of each small sample training set.
[0108] Verification module 303 is used for:
[0109] Verify whether the model obtained ac...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More