I’m designing a neural network (pytorch) that accomplishes two different, but entangled, tasks. One is very difficult, one is very easy. While training two different models (two trainings, two set of parameters,…) can crack both of the problems independently, when combined, the hard task fails, no matter what I do. I would like to understand… Read More AI good practices in combining tasks?
Hi all! Can you help me? I have a question: "How can I retrain a PyTorch model with just a .pt file ?” I’ve looked at many guides but haven’t found the answer. Everything there has a model class. I tried to import using torch.load() and others. But it doesn’t work. When I load the… Read More Load and Retrain PyTorch model…
I made the ML Model using the Resnet9 algorithm. In this model, I have 38 classes. After the prediction, It returns the tuple like this: It is returning negative values. In the above case, the highest value is -204.0966 means it has the highest chance of being the correct one. I want to make it… Read More How can I calculate confidence level of my ML Model for each Image of ResNet9
So essentially I don’t have enough RAM to train the model I want from scratch using the 2000 classes all at once. Because of that I was wondering if I could use an output layer of 200 neurons and save the weights after training the model with those 200 classes and then load those same… Read More Is there a difference in the weights of hidden layers if I train a model with an output layer of 10 neurons 10 times or only one time with 100 neurons
so i downloaded a library that hasnt been in use in years, for Neural Evolutionary Augmenting Topologies. Basically, a neural network that evolves. It came with many, MANY errors out of the box (somewhere around 20-30) and i managed to fix them all, except for these: Error C3694 a structured binding declaration can contain no… Read More Group of structured binding errors, pertains to neural networking
I am trying to develop a softmax function in python for my backpropagation and gradient descent program. I am calling the softmax function after I get my outputs of the output layer (2 outputs), the outputs are in a vector-like so [0.844521, 0.147048], and my current softmax function I have implemented is like this: import… Read More Softmax activation function in python using math library
I am using shape_predictor_81_face_landmarks.dat in order to obtain facial landmarks. Here is the result: As you you can see this is the points are actually 2D numpy arrays: [[201 122] [201 140] [202 157] [204 174] [209 190] [219 205] [230 219] [244 231] [261 236] [278 233] [292 222] [304 209] [314 195] [320… Read More Which model should I approach to classify people based on their face?