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.
1
vote
0answers
11 views
How smaller does the input data get reduced in a LSTM autoencoder
Question
In a LSTM autoencder, how smaller does my input data(59 features) get reduced in a latent vector, which is usually located in the middle between an encoder and a decoder?
Why did the ...
1
vote
0answers
18 views
How to predict value in every 120 minutes using LSTM in python
I want to predict value in every 120 minutes continuous using LSTM model. Here I wrote the code for predction. But I'm not getting proper prediction values . Here from start time I need to predict ...
1
vote
1answer
15 views
How to put data into a 1-dimensional ConvLSTM2D with keras?
I am attempting to adapt the frame prediction model from the keras examples to work with a set of 1-d sensors. I have android wearable sensor data and am designing an algorithm that can hopefully ...
3
votes
4answers
87 views
What does it mean when my test data accuracy is higher than my training data?
I'm using four years of data, training on the first 3 and testing on the fourth. Using LSTM w/ Keras. My test data set (which has no overlap at all with the training) is consistently performing better ...
1
vote
0answers
19 views
In Keras, how to get 3D input and 3D output for LSTM layers
In my original setting, I got
X1 = (1200,40,1)
y1 = (1200,10)
Then, I work perfectly with my codes:
...
0
votes
0answers
18 views
Input of the decoder of a LSTM seq2seq neural network?
I have been trying to setup a seq2seq network with LSTM layers where the output of the network should be the seven day ahead forecast, given the previous fourteen days as input. I started with a ...
0
votes
2answers
26 views
Why LSTM models do not require labels for each step?
For time related problems like, for example, stock prediction:
Let's say we have 300 days of data, 10 features, and one target: the price.
Why, for the training, we only need the price of the 300th ...
0
votes
0answers
15 views
very low val_acc in LSTM for predicting numbers in sequence
I have problem with my LSTM network or my data. I woudl like to create LSTM network to predict number based on previous N numbers. This numbers will be encoded piano notes.
About my dataset:
1. I ...
1
vote
0answers
12 views
LSTM: how to use just context?
I would like to use a model with a bidirectional LSTM layer to predict a sequence of outputs from a sequence of inputs (of length $l$).
To compute the output $o_i$ I would like to use the inputs $...
1
vote
0answers
33 views
LSTM prediction for many multivariate time series
Let's say we have 8,000 different time series where each of them has 10,000 samples and 25 features. The goal is to have an LSTM sequence to sequence model (using Keras) where one can use a sequence ...
1
vote
0answers
24 views
Generate sentences using given data [closed]
I am working on an automated insights generation use case where I want to generate meaningful sentences from given aggregated data.
For example,
Data:
...
1
vote
2answers
42 views
Training LSTM with different sequence lengths in Keras functional api
I am trying to train an LSTM model using Keras functional API. My training data is of shape:
>>> data.shape()
(100000,variable_sequence_lengths,295)
...
0
votes
0answers
13 views
Buy/Trade strategy logic for Stock Market Prediction [closed]
I need help in formulating/ brainstorming a strategy for trading ie. logic for generating buy/sell/hold signals for my Stock Market Prediction Model.
I am using Open, High, Low, Close, Volume indices ...
0
votes
0answers
14 views
0
votes
0answers
8 views
Train LSTM RNN with multiple different sets of time series data in Keras
I am trying to set up a program where an airplane is taking off from one city and flying to another. Depending on a number of factors it can take different routes to get to the city. Since some ...
0
votes
1answer
20 views
How to structure time series for binary classification in Keras?
I'm trying to classify churn (1 or 0) for a user based on day-to-day time series and activity levels, something like this:
...
0
votes
0answers
11 views
What will go wrong if we apply linear or other types of regression to translate sentences between two languages?
Disclaimer: I asked the question at https://stats.stackexchange.com/questions/408463/what-will-go-wrong-if-we-apply-linear-or-other-types-of-regression-to-translate, but didn't get any response, so I'...
0
votes
0answers
10 views
Long-term Recurrent Convolutional Networks (Keras) for Game Bot
I want to use CNN and LSTM to make a Game Bot. Basic idea is to capture 32 frames of gameplay, and then, for that sequence, predict an output. My gamebot is more like a self driving car. Capturing the ...
0
votes
0answers
7 views
Classify activity person go into and go out the car (behavior detection)
I'm working on the problem classify activity get out the car and get in. Also need to classify if upload and download activity going near the car
Need advice how to fix problem of overfitting model ...
1
vote
0answers
17 views
Time Series Forecasting for Multiple Customers using one RNN
I have a product which has univariate and also multivariate time series data from multiple customers. I have variable amount of data available. Ranging between couple of years to couple of months.
...
1
vote
0answers
10 views
How to download BibTex raw dataset for deep learning (LSTM )?
I am trying to test my model on the benchmark text classification dataset "Bibtex". The Bibtex dataset is available in tf-idf vectors but LSTM works on sequences. Bag of Words is a vectorized ...
5
votes
0answers
38 views
How to arrange the dataset/images for CNN+LSTM
I am working on an image classification problem using Transfer Learning with Resnet50 as base model (in Keras) (For example Class A and Class B).
There is a time factor involved in this ...
1
vote
1answer
22 views
Problems finding an LSTM model for classification
I am doing a study for the classification of musical genres using deep learning techniques. The work consists of making a classification using an LSTM model.
I am using GTZAN as a data set, and ...
1
vote
1answer
28 views
[Keras][LSTM] error due to shape mismatch
I have following data. Where I have 2 samples. Each sample I have 3 time steps each with 2 features. I intend to have 2 batches (to updates weights after every sample)
...
0
votes
0answers
24 views
Simple Keras LSTM for Time Series Classification
I've tried reading through a few tutorials but they all seem to involve more complex tasks that involve steps I don't quite understand and my not be necessary for my use, so I'm having trouble ...
0
votes
0answers
16 views
3
votes
0answers
16 views
How do I implement masking in TensorFlow eager execution?
I am training a stateful RNN on variable length sequences (optional: see my previous question for more details).
I padded the sequences to a fixed length with the value -1.
The when batches are ...
1
vote
0answers
12 views
Using TF Dataset API to process sequences for stateful RNN
I am trying to use the TensorFlow (v1.13) Dataset API to save and load long sequences for a stateful RNN.
Basically, lets say I have n_seq sequences, each fixed ...
0
votes
0answers
10 views
Predict future event using text and class
I have a dataset containing reports of events in a company when something occurred. The report have:
Text description of the event
Code classification of the event
Date of the event
The code ...
0
votes
0answers
23 views
Time series output of LSTM network has a much lower scale than the input scale
I'm trying to use an LSTM network to predict a sequence of a time series variable. I'm trying to predict a sequence of 3 elements based on the sequence of the previous 6 elements. The Keras code that ...
0
votes
0answers
19 views
What's the difference between hidden layer size and sequence length in RNN and LSTM?
I have been exploring RNNs in keras implementations. In the LSTM layer we have to provide a hidden layer size and also a sequence length. My question is, what does hidden layer size correspond to and ...
0
votes
0answers
28 views
What method will be appropriate for fitting this kind of data?
My data set has 15 independent variables and 2 predictors or response. The problem is to find a mapping between input and output variables. Basically, given a feature vector as input, the trained ...
1
vote
1answer
19 views
Smaller network width than output size?
I am trying to figure out if it makes sense that the width of the network could be smaller than the input/output size?
So for example, I am giving the Neural Network 2048 numbers, and I am expecting ...
0
votes
0answers
13 views
Drawing Trees, finding Probabilities and Predicting
My first post here :)
I have some transition states. Like below, where each row reflect to a specific process:
...
2
votes
1answer
44 views
Best way to classify plots which are overlapping?
I have an experiment in which it was done under two conditions. For each condition, the experiment was performed 26 times. The output of the experiment is a plot with 70 time indices. I would like to ...
1
vote
0answers
36 views
How to apply an RNN to forecast non-stationary time series?
Is it possible to predict a time series which is non-stationary, in the sense that, the dependent variable Y have an increasing trend? Therefore, the highest value of $Y$ in the training set may be ...
0
votes
0answers
27 views
How to get the right keras model.predict label (NLP problem)
I implemented a small LSTM neural network to predict the notes for a movie. But I have an interpretation problem to convert the prob_result that ...
0
votes
0answers
9 views
I am building a gesture recognition system using video data from ConGD Dataset and am not able to create the 5D array,as input to the network
I need a 5D input for my network using ConvLSTM and 3D-CNN. Converting the videos to an array of videos would contain a 5D array (Num of examples, Num of frames, FrameWidth, FrameHeight, NoOfChannels),...
1
vote
2answers
43 views
Loss value going down while accuracy remains constant?
While I am training, it seems like my loss is going down, but my accuracy remains constant throughout training. It always seems to go towards 0.0023 no matter how I tweak my network, input data length,...
1
vote
1answer
44 views
Using LSTM to predict binary classification - accuracy stuck at 50% - how to use statefulness
I am trying to use an LSTM model to make binary classifications; however when I train the model the loss stays around 0.69 (ie. -$\ln(0.5)$) and the accuracy at 0.5, which suggests to me the model is ...
0
votes
0answers
12 views
Hybrid Model : RNN + MLP + RNN
I am trying to develop a model, as follow: an RNN with three LSTM takes in the input (5|1|54), the 5 previous days and 54 feature. In the end of the first RNN I would like to take the mean and std of ...
0
votes
1answer
32 views
LSTM input and output for sentiment analysis
I'm studying this LSTM network:
https://www.kaggle.com/paoloripamonti/twitter-sentiment-analysis
...
2
votes
1answer
73 views
How to design batches in a stateful RNN
I am using TF Eager to train a stateful RNN (GRU).
I have several variable length time sequences about 1 minute long which I split into windows of length 1s.
In TF Eager, like in Keras, if ...
0
votes
0answers
21 views
clarify convLSTM usage for regression
I am trying to use keras and convLSTM layer to predict future weather data based on previous weather radar pictures.
I use i timesteps i.e. i radar images as input data. In my diagram, I assume i=3.
...
0
votes
0answers
60 views
LSTM Predict values out of test
I'm trying to predic stock values from a dataset, for example: Google stock.
I have this easy model.
...
1
vote
0answers
27 views
Measuring uncertainty in an LSTM network using dropout in keras/tensorflow
I've created a simple LSTM network for testing
...
0
votes
0answers
14 views
ValueError: Dimensions must be equal, but are 256 and 12 for 'attention_layer/MatMul_1' (op: 'MatMul') with input shapes: [?,256], [12,256]
I'm working on a sequence-to-sequence approach using LSTM and a VAE with an attention mechanism.
...
0
votes
0answers
25 views
ValueError: Cannot convert a partially known TensorShape to a Tensor: (?, 256)
I'm working on a sequence to sequence approach using LSTM and a VAE with an attention mechanism.
...
0
votes
0answers
22 views
Duplicate QUORA question detection:Kaggle Dataset
I have tried to use 2 BILSTMs along with the attention layer but the validation accuracy is not improving at all. Could anyone suggest an alternative to increase the accuracy?
Layer structuring:
<...
0
votes
1answer
37 views
LSTM with target outside the timeseries
Say I want to predict the final size of a flower depending on the raining and temperature during 200 days. The flowers are in differents towns, so each flower has its own conditions of rain and ...
