VALIDATION CRITERION MET DURING TRAINING OF VGG19

Question

Deepika B 2020-2-25

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/507351-validation-criterion-met-during-training-of-vgg19

评论： Srivardhan Gadila 2020-4-23

Is this network is overfitting? There is condition called "validation criterion met " arises and stop the training process why? what is the solution for that? Is this is possible to train a neural network using transfer learning(vgg19) by having single 8gb ram cpu ? Is there any chance that this 8 gb ram will lead to these kind of errors ?

Thank You in advance..

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Srivardhan Gadila 2020-2-28

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/507351-validation-criterion-met-during-training-of-vgg19#answer_417724

I think you have mentioned some function for the 'OutputFcn' Name-Value pair argument in trainingOptions and this function stopped the training after seeing that the validation has loss increased from previous step.

You can used pretrained VGG19 network for transfer learning and you can refer to the following workflows:

Make use of the imageDatastore(also refer to Datastores for Deep Learning) for training the network on the mentioned hardware. Specify the appropriate Name-Value pair arguments for 'ExecutionEnvironment' & 'MniBatchSize' in the trainingOptions.

2 个评论
显示无隐藏无

Deepika B 2020-2-29

Thank you so much, for your kind response sir.

what are all the information i got from this confusion matrix ? Based on this confusion matrix, Is my model is a good one or bad one ?

how do i test my model using test data ?

thank you in advance sir.

Srivardhan Gadila 2020-4-23

The rows correspond to the predicted class (Output Class) and the columns correspond to the true class (Target Class). The diagonal cells correspond to observations that are correctly classified. The off-diagonal cells correspond to incorrectly classified observations. Both the number of observations and the percentage of the total number of observations are shown in each cell.

Refer to Examples of plotconfusion function.

请先登录，再进行评论。