New answers tagged

1

some questions will help give better answers: When you say underfitting, I assume you mean that the low accuracy is on the train set, correct? I'm asking also because with that amount of parameters for such a small training set I would be far more concerned with overfitting 530 images is very small dataset, I would consider going with a pretrained ...


0

How many images do you have as input? Maybe you haven't load the images as expected (seems about 30k). Check if the download process has been done properly by getting the images'quantity.


0

The thing about neural networks is that they are uninterpretable. I bet there might be some techniques that tackle this issue. You can try google scholar for that. But yeah, neural networks are basically black boxes. You can extract the convolutional filters and see the results of each layer on the image, but you will just come up with ,,When this shape is ...


0

Problem solved. It was a dumb and silly mistake after all. I was being naive - maybe I need to sleep, I don't know. The problem was just the last layer of the network: model.add(tf.keras.layers.Dense(10, activation = 'softmax')) It was supposed to be model.add(tf.keras.layers.Dense(num_classes, activation = 'softmax')) I could not build a network with an ...


0

First, you have a huge number of nominal categories (1000 classes). If you need a simple answer regardless of dimensionality and complexity, you just need to use one-hot encoding and sigmoid activation function in the last layer for the 1000 neurons there. But you will end up with a huge sparse output matrix. If you look for an optimized solution and your ...


3

First of all, here is the answer to your question: YES You have a beginning of CNN (Convolutional Neural Network) in your code. Just don't forget to add activation functions (usually ReLU) after convolutional layers. Another unconventional thing is you use 2 outputs for your binary classification, which is not the way to go, we would rather use 1 output that ...


0

The similarity between datasets cannot explain the test error of neural networks, the optimization can go very differently for slightly different datasets


Top 50 recent answers are included