Questions tagged [lstm]

LSTM stands for Long Short-Term Memory. When we use this term most of the time we refer to a recurrent neural network or a block (part) of a bigger network.

Filter by
Sorted by
Tagged with
1
vote
1answer
24 views

LSTM Time-series classification - derived feature

I have a time-series dataset and I want to derive a new feature based on a date column which I believe might improve my predictive model. The feature is if it's weekday or weekend. I am not sure how ...
0
votes
0answers
19 views

NLP Word ordering problem in Keras

I'm trying to solve the word ordering task: given a syntactically unordered sentence, recover the right order of the words. The adopted approach is to transform each sentence in a dependency tree and ...
0
votes
0answers
19 views

How do I multiple individual time-series data to train a LSTM model?

I have 100 univariate time-series from individual patients measuring glucose levels. Each time-series is ~20k. I need to train my LSTM model using all of the data. I don't think concatenating all ...
0
votes
0answers
24 views

Multivariate supervised LSTM for for regression only (not for prediction)

I am a beginner with machine learning. I want to get a multivariate regression model with LSTM. I followed this tutorial by Jason Brownlee: https://machinelearningmastery.com/multivariate-time-series-...
1
vote
0answers
22 views

Keras Bidirectional LSTM: low training and validation loss but very bad predictions

I'm training a Bidirectional LSTM using Keras. My task is to predict the words order in a sentence, so, given a sentence, ...
0
votes
0answers
7 views

seq2seq - Inference model and train model produce far too different results on the same validation set

I am working on a timeseries seq2seq problem. For my approach, I am using LSTM seq2seq RNN's with Teacher Forcing. As you already know, for the purpose of the task a model should be trained, and then ...
0
votes
0answers
39 views

Regression by PhasedLSTM with a gradient explosion

I found PhasedLSTM inspirational, and used it (PLSTM: Phased LSTM in Keras) to perform the regression (to find the correlation between an input sequence and an output sequence), with Adam optimizer <...
0
votes
0answers
47 views

Neural Network Architecture to approximate simple function

What neural network architecture would approximate this function? This is monotonically decreasing non negative time series, that has non-stationary mean. $$ q(t) = \frac{qi}{(1+b*D*t)^(\frac{1}{b})}...
1
vote
1answer
39 views

Noramlization Time Series to Predict Stocks exact Price

I am trying to made a model for neural network that tries to predict prices of the stocks point by point using a LSTM (yeah I know that probably did not get anything and I should predict up/down or ...
0
votes
0answers
26 views

Validation Loss does not decrease but validation average precision improves

I am training three neural networks (2CNNs, 1LSTM) on an EHR dataset, which is to predict diabetes according to (100 labs * 360 days) data. However, when I trained them, there is a common phenomenon ...
1
vote
0answers
19 views

Multi-feature padding for LSTM

I am trying to train a LSTM on an NER dataset which contains multiple features. But I'm having trouble understanding how to pad multiple features. The dataset contains the following 3 features per row:...
0
votes
0answers
11 views

LSTM model for time series forecasting giving very poor performance even on training data

I am trying to perform timeseries forecasting with LSTM. I trained a model which is giving very less loss on training but when I try to predict the training data itself, it gives values far off from ...
0
votes
0answers
19 views

Forecasting binary time series

I am working on the next event occurrence prediction task and the data is binary time series with 1 if the event occured and 0 if not. I want to predict whether the event will occur or not on the day ...
1
vote
0answers
83 views

PyTorch time series prediction beyond test data

I am currently playing around with pytorch models for time series prediction. I have managed to successfully run a model to predict test data. I was wondering how can I use it to predict beyond test ...
0
votes
1answer
47 views

How to ignore vectors of zeros (i.e. paddings) in Keras?

I'm implementing a LSTM model with Keras. My dataset is composed by words and each word is an 837 long vector. I grouped the words in groups of 20 and to do this I ...
0
votes
1answer
25 views

How to train NER LSTM on single sentence level

My documents are only a single sentence long, containing one annotation. Sentences with the same named entity of course are similar, but not context-wise. NER ...
0
votes
0answers
13 views

Multiple Outputs LSTM

I am trying to create a neural network capable of classifying the type of music that a user normally listens to.The idea is that the neural network will receive a 2D input matrix. These matrix ...
0
votes
0answers
19 views

Binary classification problem with time series and static data

I am trying to solve a classification problem on a dataset with company information. My dataset: Multiple companies (around 16,000 in training set) For which I have time series info about salary, ...
0
votes
0answers
12 views

Creating ANN's (LSTM) for multiple datasets

I have three datasets, each representing time-series water quality data from three different regions (upper, middle, lower regions) of the same geographic area. I want to create different types of ...
0
votes
0answers
22 views

What type of Machine Learning to use?

I have a beam loaded with random loads, for each timestep I obtain accelerations and maximum moments on the beam. My prediction should receive a timeseries of accelerometer data and output the ...
0
votes
0answers
9 views

Truncated Backpropagation Through Time (TBBTT) in Reinforcement Learning

I am currently looking at the OpenAI Five paper from OpenAI. For backpropagation they write: ...
0
votes
1answer
50 views

Prepare data for an LSTM

-I want to make a python program using the LSTM model to predict an output value that is 1 or 0. -My data is stored in a .csv file of the form: (Example of line) Date time ...
1
vote
0answers
12 views

Oversampling for regression for data grouped in clusters

I am dealing with a regression problem in which I want to predict the upcoming value of a time-dependent variable by using the previous values of other variables (not including the output variable ...
0
votes
1answer
36 views

How to implement LSTM using Doc2Vec vectors to get representation?

Hi all. I'm a newbie in ML. I read and found a paper about A Multi-Level Plagiarism Detection System Based on Deep Learning Algorithms and want to implement this model . But I can't find more about ...
0
votes
1answer
76 views

LSTM evaluation metric MAE explanation

I have a hard time understanding the LSTM model performance as I summarize my model as follow: ...
1
vote
0answers
10 views

Keras LSTM using timesteps = 1 and train_on_batch

I'm using an LSTM to achieve a classification problem. I have a dataset composed by sentences, each sentence is composed by a variable number of words and I have to ...
0
votes
0answers
17 views

Tensorflow data API: Building continuous streams of data from a Dataset of Datasets

I'm trying to build a language model with LSTMs (like ELMO). I've got a lot of documents and want to split them into words as input, but keep their order. So it should get all words of the first ...
1
vote
0answers
40 views

Data leakage in bidirectional LSTM timeseries data

Does it cause data leakage to train a bidirectional LSTM on data where a user can be a sample in the training data multiple times? Each row is a snapshot at a different point in time for a given ...
1
vote
0answers
16 views

LSTM features classification output

I am very new at this, so I might be wrong about my choice of model, but my problem is the following. I am trying to generate music, hence the reason I am using an LSTM. I have the following sequence ...
2
votes
1answer
163 views

LSTM: many to one and many to many in time-series prediction

I am trying to predict the trajectory of an object over time using LSTM. I have three different configurations of training and predicting values in my mind and I would like to know what the best ...
0
votes
0answers
19 views

How to deal with discrete variables in Multivariable Time Series forecasting?

I am tackling this time series forecasting problem to basically predict number of sales in the future training dataset looks like this: ...
0
votes
0answers
14 views

Time series problem with LSTM is not predicting correctly

. My Initial dataset looks like this: Text sample of Dataset: ...
0
votes
0answers
9 views

How to feed multiple asymmetric inputs to LSTM layer?

I'm trying to create an encoder-decoder architecture with an LSTM encoder. The intention is to use both the input image as well as the class label as inputs to the encoder, and to have them share the ...
0
votes
0answers
23 views

What are the advantages of combining BiLSTM and CRF?

BiLSTM-CRF is a common model for sequence tagging (POS tagging, NER, ect.). What are the advantages of combining BiLSTM and CRF? What is the role of each one of the parts in this combination?
1
vote
0answers
54 views

State of the Art/Research 2020 of Time Series Forecasting/Prediction

Im looking for the state of the art/research of time series data for forcasting/prediction. As far as im aware it is Extrem Gradient Boosting (XGBoost) or LSTM (neuronal networks) or are there other ...
0
votes
2answers
23 views

Keras fit_generator() for long signals

I want to make an LSTM network and I have quite a long signal that I want to use as my training data My X_train is a CSV-file which contains 12 signals with a length of 54 837 488 My y_train is an ...
0
votes
0answers
48 views

How to implement a LSTM for multilabel classification problem?

I would like to develop an LSTM because I have a variable input matrix. I am zero-padding to a specific length of 800. However, I am not sure of how to classify a certain situation when each input ...
0
votes
0answers
12 views

Does someone have a low level inplemenation of tensorflow's ConvLSTM2D

Right now I am using the ConvLSTM2D module of tensorflow and it works. But now I would like to feed an additional input between the convolutional stage and the lstm module. Therefore I would need a ...
0
votes
0answers
20 views

MAPE over 100% after normalization of dataset

I try to forecast power demand for next 24 hours. Years 2017 and 2018 are my training set, 2019 is test set. I use multistep vanilla LSTM . In first step I used original data with any preparation and ...
0
votes
0answers
11 views

time series forecasting of time to leave for multiple customers using one model

I am a beginner in the domain of forecasting and I was wondering if such a problem could be solved with time series analysis : given customer historical data of taxi pickups,along with the weather ...
0
votes
0answers
8 views

Is it possible to train a model on a more complicated data set and then retrain on a simpler dataset that has continuous frames for LSTM model?

I'm attempting to do lane detection using CNN-LSTM architectures on the TuSimple dataset. However the TuSimple data set isn't very difficult and the results are sometimes poor (visually). I want to ...
0
votes
0answers
18 views

Is there any multivariate time series prediction tutorial/code using RNN LSTM for R?

I have been searching and reading many articles, tutorials and academic journals including this forum. All I find is predicting a univariate time series. (except for François Chollet's tutorial, I ...
0
votes
0answers
20 views

Multiple time series sequence prediction for multiple multivariate time series

My question is somehow similar to this question, but not satisfied with the answer. I have 100 samples, each sample consists of ...
0
votes
0answers
16 views

Can we use RNN or LSTM for prediction and not forecast

I know RNN and LSTM learn from past data, and can forecast next data. In my situation, I have a learning data-set that hide other information I wish to discover or approximate.(This seems rather an ...
0
votes
0answers
16 views

Attention model with seq2seq over sequence

On the official tensorflow page there is one exmple of a decoder (https://www.tensorflow.org/tutorials/text/nmt_with_attention#next_steps): ...
2
votes
0answers
45 views

Remove subwords from BERT output

I'm trying to build a multilingual WSD system with BERT on top as the embedding layer. In order to have better performances, after BERT finishes its job (and performs Transfer Learning), I need to ...
2
votes
1answer
31 views

How do you think about neural networks and ways to design new models?

I'm currently learning about neural networks, and it seems to me that there usually is no good theoretical explanation given for why certain architectures work; there is most of the times, no formal ...
1
vote
1answer
64 views

How to code a simple forward propagation of recurrent neural networks?

I know the theory behind recurrent neural networks or RNN but I am confused about its implementation. This is an rnn equation I got from the web, I tried to code the forward propagation alone in ...
0
votes
1answer
137 views

keras CNN lstm add model depth

Would anyone have any advice on how to add model depth? This works below but I was hoping to experiment with adding in additional non-TimeDistributed layers. ...
0
votes
0answers
26 views

How to train LSTM for multiple time series with multiple variable and diferent size of time series?

I have a dataset of aircraft messages wich have an column that identify each aircraft example: idaircraft=1 , timestamp=340503404, altitude=xxxxxx,longitude = xxxxx, latitude = xxxxx, Touchdown = ...

1 2
3
4 5
17