Classifier

Dear Client I am an experienced Machine Learning Engineer and data scientist. I have reviewed your job and here are few questions; > If your training is done, you must already have a classifier already? or am i wrong..... > Did you have trained the network using images, mean is it a CNN trained? > If it is a CNN, did you have already trained the dense layers as you have used transfer learning, i assume you might have used CNN layers from the previous architecture and created few new layers? > is your current code in notebook? and you want to use the model for predictions in the same? Will be waiting for your response. Warm Regards Z

Can I ask what is actually your intention? If you did train model by using transfer learning, then you probably have classifier already. What I can imagine is that you want to use some other classifier, like one that classifies your* dataset, not dataset originally trained on. If that's true, I can do it. I have done it already in couple of projects. I did several similar jobs like a model that classifies is-a-bird/is-it-something-else, is-a-man/is-a-wman, fat-face/narrow-face and many more networks including RNN for extracting numbers from text, LSTM for toy-project of weather prediction, image compression by using auto-encoder, finding similar images (out of 100.000 dataset) by using FAISS and little bit more. Best, Vedran

I have a background in the neural network. Recently, for my research, I generated classifier to remove bad sample and also generated surrogate models to save computational cost in my project. I am using this technique in my research since 2017.