New answers tagged

0

Doing some further searching, I found a nice post by Jason Brownlee that has helped me better understand the problem and one potential solution: a multi-input model with an LSTM input for the historical data and a vector input for expected conditions. Furthermore, Keras' Functional API (https://keras.io/guides/functional_api/) will help me build such a ...


1

This is correct if one did not include biases. By including biases ($b_o$ and $b_h$). Number of parameters in $b_o$ is equal to number of outputs (k) and number of parameters in $b_h$ is equal to number of hidden layers (n). Hence the final value is: $n^2 + n + mn + kn + k$


0

It all depends on the nature of the data in relation to the labels. For instance, if all that you need to appropriately classify the input sequence is to know the values at certain fixed points, then a mere multi-layer perceptron (MLP) could do. However, if in order to properly classify it was needed to take a look at the trends, maybe the MLP would not ...


0

A finite impulse response (FIR) RNN can be expressed as a directed acyclic graph (DAG), hence it can be represented as a FFNN. So, you could theoretically make an equivalent FFNN to an RNN in this situation. In terms of performance between non-equivalent models, an RNN would probably be better because, as they are inherently designed for sequential data. In ...


1

I think this will help. My understanding is that one-to-many and many-to-many(like in 4th case of your pictorial) are in a way similar to autoregressive networks, where you utilize the prediction you've made and use it predict further ahead.


0

You are describing named-entity recognition (NER) which seeks to locate and classify named entities mentioned in text into pre-defined categories. In this case, the pre-defined categories are music genres. There are many algorithms for NER. In addition to Recurrent Neural Networks (RNN), Hidden Markov Models (HMMs) and CRF (Conditional Random Field) are ...


Top 50 recent answers are included