04 April 2018 2 510 Report

Hi I used the below program for fine tuning VGG16 for 5 classes (11600 for each class) following the example in https://keras.io/applications/ . It gives a good validation accuracy, which is around 93%. However, I used the same architecture to solve 2 classes selected from the same 5 classes (binary classifier), the validation accuracy always stays around 42%. I expected to get better results. Is the problem from the model or the data? Any advice?

Thanks

More Kamal M Othman's questions See All
Is there any doorway indoor images dataset?

Hi Everyone, I am looking for indoor doorway images dataset (open doors), e.g. in houses or offices, for classification project. Can anyone help? I found only doorway outdoor dataset in Places...

03 April 2018 2,863 3 View

Similar questions and discussions
What's the best way to measure growth rates in House sparrow chicks from day 2 to day 10?

What's the best way to measure growth rates in House sparrow chicks from day 2 to day 10? Since, the growth curve from day 2 to 10 won't be like the "Logistic curve" it might not follow logistic...

03 March 2021 1,401 3 View

Can NFL theorem be valid in infinite search space in ML?

NFL theorem is valid for algorithms training in fixed training set. However, the general characteristic of algorithms in expanded or open dataset has not been proved yet. Could you show your...

01 March 2021 1,189 3 View

Is there any research regarding how training and development is going to change in the era of Fourth Industrial Revolution?

As the organizations getting prepared for the fourth industrial revolution, to enhance the skills of their workforce what type of changes are necessary and going to occur in the training and...

01 March 2021 3,199 3 View

Which group formation theory do you recommend for projects that aim to bring together entrepreneurs for training for regional development?

I would like to have ideas about which theories of group formation are suitable for entrepreneurial training projects, with the aim of creating bonds, building trust, and collaboration for...

28 February 2021 4,223 2 View

Satellite orbit estimation using Recurrent Neural Network - what should I use?

Hello, I am a master's student studying in Yonsei Univeristy, Korea. I am trying to estimate the state of satellite, using Neural Network. Below is a simple flow of my study. 1. Train (t0 ~ t1)...

23 February 2021 6,383 3 View

Can the problems of overfitting be solved without cross validation?

For an overfit data set the cross validation error will be much bigger than the training error. Is it true or false?

23 February 2021 3,616 2 View

How to assure whether a regression tree overfit or not by seeing bias-variance value of the model?

Lets consider a regression tree in which variance is 1.1065*e-10 and bias is 2.962e-13. Also the model RMSE on training set is 1.5e-5 and on training set is 1.2950e-5. Is the model is overfit or...

22 February 2021 7,755 2 View

Is it possible to add a source term to the momentum equation in ANSYS CFX?

I want to add a source term as a body force to the momentum equation in Ansys CFX for modeling the VGs and I don't know how do it !!

21 February 2021 2,971 4 View

In Machine Learning, isn't the problem Out-of-Distribution (OOD) instances really a form of over training?

If (1) the over training manifests itself in failure against instances that are new or excluded from the training set, and if (2) OOD instances are problematic because they are new and by...

18 February 2021 4,839 2 View

Any methods to train a neural network with random (stochastic) and non-measurable variable?

Dear scholars, I was wondering if there are any scientific methods to train a neural network or ANFIS when one of the input variables is stochastic (random) and more importantly non-measurable?...

17 February 2021 1,909 3 View

Our partners will collect data and use cookies for ad personalization and measurement. Learn how we and our ad partner Google, collect and use data. Agree & close