6,119 questions
-1
votes
1
answer
35
views
defining the input shape of LSTM in Kears [duplicate]
I'm trying to build LSTM using keras. but it doesn't accept to pass the input shape or any other other parameters in the LSTM layer.
This is my code
def LSTM(structure):
model = Sequential()
...
- 19
0
votes
2
answers
417
views
How do I concatenate outputs of two different models if the shapes are completely different?
I am having two pretrained models in pytorch which use different type of input data. Both of them I am using for feature extraction and want to concatenate their outputs at the end and put them into ...
- 21
0
votes
2
answers
327
views
why is my predicted LSTM model shifted left as well as the accuracy 0
I have seen some other posts with shifted LSTM models being shifted however looking into them I didn't find a solution.Is something wrong with how im slicing the data in the Sequential_Input function? ...
- 95
0
votes
1
answer
237
views
Error using plot in MATLAB. Invalid data argument
I am using MATLAB 2021a. I have regression problem. I facing error in using buffer.
Problem is here actually.
Q_signal is 1000x1 double
and
estimated_Q_signals is 991x1 cell
and every cell is 1x10 row ...
- 1
1
vote
0
answers
118
views
How to add multiple embeddings (layers) to LSTM layer
The similar question was asked before here How to Merge Numerical and Embedding Sequential Models to treat categories in RNN, but I didn't understand a clear answer. I will use the author’s @GRS ...
0
votes
0
answers
203
views
How to train GNN on huge datasets?
I have a dataset having a shape of (783126,3) where in my dataset there are three columns R1, R2 and Score representing Residue 1, Residue 2 and their Score. I want to train a Graph Neural Network on ...
1
vote
1
answer
102
views
LSTM model has a concerning loss graph and returns a constant value on predictions
My dataset is 597515 rows, 31 columns measuring one variable. So input shape (31,1).
import pandas as pd
import tensorflow as tf
gpu_devices = tf.config.experimental.list_physical_devices("GPU&...
- 71
0
votes
1
answer
154
views
Input 0 of layer "bidirectional" is incompatible with the layer: expected ndim=3, found ndim=2. shape received: (None, 32)
Here is the snippet of code:
vectorizer = TextVectorization(max_tokens=MAX_FEATURES,
output_sequence_length=1800,
output_mode='int')
...
- 43
1
vote
0
answers
60
views
Low Tensorflow Model Test Accuracy
My model takes binary input which is from binary encoded text (not one-hot encoded). I achieve a binary accuracy of 99.5% and accuracy of 85%. Immediately after training, I achieve a test result of ...
- 21
9
votes
5
answers
63k
views
No module named 'keras.wrappers'
I have this code on google colab which allows me to optimise an LSTM model using gridsearchCV, but recently an error message has appeared:
ModuleNotFoundError: No module named 'keras.wrappers'.
is ...
- 125
0
votes
0
answers
179
views
ValueError: Exception encountered when calling layer 'reshape' (type Reshape)
I am working on image captioning model using satellite images. I have a .json file through which I am training the model for generating captions. Below is my code-
import numpy as np
import pandas as ...
2
votes
0
answers
75
views
Keras LSTM for continuous output and with EarlyStopping
I wrote an LSTM regression model. It is a model with batch_size=1 and return_sequences = True for the last LSTM layer. I also set validation_data and patience for training. But there seem to be ...
- 21
0
votes
0
answers
94
views
Issues with LSTM-based Time Series Prediction in TensorFlow
I'm trying to predict time-series stock market data using TensorFlow and a Recurrent Neural Network (RNN) with LSTM layers. However, I encountered an input data shape-related error. So, I need ...
0
votes
0
answers
90
views
Keras error: LSTM.__init__() missing 1 required positional argument: 'units'
The code in TT_RNN is :
def __init__(self,
tt_input_shape, tt_output_shape, tt_ranks,
activation='tanh',
recurrent_activation='hard_sigmoid',
...
- 1
0
votes
0
answers
155
views
Very low accuracy and high loss with LSTM model
I try to learn the inverse kinematics of a robotic manipulator. To do that I have a simulator with which I acquired data.
My dataset is composed of positions in X, Y and Z and actuator variables (6 of ...
0
votes
1
answer
719
views
LSTM Time Series forecasting
I've been working on a time series forecasting project using LSTM (Long Short-Term Memory) by following a YouTube tutorial. My dataset consists of NDVI (Normalized Difference Vegetation Index) values ...
- 83
0
votes
0
answers
271
views
Exception encountered when calling layer 'lstm' (type LSTM)
I am trying to build a time series model using LSTM to predict money printing values. Dataset contains monthly money printing records, so have used the sequence_order as 12. Following is the code ...
0
votes
1
answer
80
views
My LSTM network doesn't work while doing inference?
I'm trying to built a Conv-LSTM network using PyTorch, model is pretty much like an image caption generator, the model learns to predict words pretty good while training but doesn't predict anything ...
- 7
-2
votes
1
answer
223
views
How can I correct my Time Series LSTM RNN for Binary Classification favoring Class 0?
I am attempting to predict a binary outcome based on 15 continuous sequences (except one which isn't a continuous line, but still a sequence). The dataset contains 933k datapoints for all 15 features ...
- 1
0
votes
0
answers
1k
views
SHAP DeepExplainer for LSTM time series data
I have a PyTorch LSTM model that takes as input a sequence of 12 time series values, and I also include 3 static features. The output is a sequence of 6 values. I want to use SHAP (SHapley Additive ...
- 95
0
votes
2
answers
175
views
LSTNet RuntimeError: Input and parameter tensors are not at the same device
I cloned a github repo which runs a pytorch deep learning module, I customized this part to send the module and the data into GPU.
train_dataset = MarketDataset(train_data, history_len=history_len)
...
- 191
1
vote
2
answers
6k
views
How to perform hyperparameter tuning of LSTM using GridSearchCV?
I am new to deep learning, and I started implementing hyperparameter tuning for LSTM using GridSearchCV. My dataset contains 15551 rows and 21 columns and all values are of type float. Here is my code:...
- 113
0
votes
1
answer
196
views
TensorFlow training issue: Failed to convert a NumPy array to a Tensor (Unsupported object type numpy.ndarray)
I'm trying to train an LSTM network with some sequential data with vehicle trajectories, in order to get driver's intention. Here is how I have structures my data:
3 positional features per vehicle, ...
1
vote
1
answer
312
views
Using an LSTM model with 2 simultaneous signals to denoise one of the signals
in the medical imaging field, it is common to have two signals, where artifact in one signal bleeds into the other. I'm exploring whether a deep learning model can be used with these signals to ...
- 182
0
votes
0
answers
33
views
input shape in LSTM-CNN model
How can I give this data to LSTM-CNN model as an input shape?
I put my own model code. My data is a satellite image.
print(train_x.shape,train_y.shape,test_x.shape,test_y.shape)
(9219, 7, 7, 12) (...
0
votes
2
answers
2k
views
SHAP TypeError: 'NoneType' object is not callable
I have a Keras Sequential Model with LSTM and want to analyse it with SHAP. The model runs, but when it comes to the SHAP block it terminates with a type error:
class ANN: ...
- 103
0
votes
1
answer
71
views
I am trying to run a CNN-LSTM model on a series of images with dimensions (9219, 7, 7, 12) but I get the following error in fit model
model = Sequential()
model.add(Conv1D(32, 4, activation='relu', padding='same', input_shape=(train_x.shape[1], train_x.shape[2] * train_x.shape[3])))
model.add(LSTM(32, return_sequences=True))
model....
0
votes
1
answer
122
views
ValueError: Input 0 of layer "sequential_21" is incompatible with the layer: expected shape=(None, 4, 1), found shape=(None, 7, 7, 12)
I am trying to run a CNN-LSTM model on a series of images with dimensions (9219, 7, 7, 12) but I get the following error :
model = Sequential()
model.add(Conv1D(32, 4, activation='relu', padding='...
0
votes
1
answer
329
views
Subclassing keras.model to create a custom autoregressive LSTM model with multi-column input
I'm trying to create a model to forecast energy grid load (net electricity consumed by the grid) based on weather data. In production, we won't have load data to do a standard batch prediction. We're ...
- 1
1
vote
0
answers
141
views
keras seq2seq with GRU instead of LSTM
I am trying to modify the code in https://keras.io/examples/nlp/lstm_seq2seq/ so it uses GRU instead of LSTM. I have managed to get it to train properly and have constructed the encoder-decoder model ...
- 142
1
vote
0
answers
60
views
CNN - RNN model data preparation
I have a dataset that has images for 2000,2010 and 2020. I wrote them in 3 different tfrecord files named combined_2000.tfrecord, combined_2010.tfrecord and combined_2020.tfrecord. I want to use a CNN ...
- 307
-2
votes
1
answer
84
views
Issue with LSTM/GRU model acurracy in a protocol prediction implementation
I have a specific model to create.I will go into further details below
I have a wireshark cap file,from which i have generated a csv file of 1.4 rows(not the entire cap file but just a sample of it)
...
- 37
0
votes
2
answers
98
views
Neural Network Architecture for Time Series as Inputs and Outputs with Variable-Length Inputs
I'm currently learning and working on implementing a time series prediction using different packages TensorFlow, Pytorch Neural Network. My goal is to feed a combination of time series data (GM Data) ...
- 3
0
votes
2
answers
596
views
Tesseract Training - Error reading radical code table data/langdata/radical-stroke.txt
I've tried to train Tesseract OCR on specific font, based on polish language model (pol) and my own "ground truth" text - it may be important, that the one generated by me does not contain ...
- 1
0
votes
0
answers
52
views
Why LSTM provides different prediction with different batch_size?
I am trying to experiment with a stateless LSTM model on a large dataset and trying to understand the effect of batch_size on the prediction.
What I understand is that in model.predict, the batch_size ...
- 57
1
vote
0
answers
138
views
Integration of Custom Data Loader with LSTM
I am trying to integrate Custom Data Loader with LSTM
num_epochs = 10
for epoch in range(num_epochs):
model.train()
for batch_x, batch_y in dl:
optimizer.zero_grad()
outputs = ...
0
votes
0
answers
43
views
LSTM network for multiclass classification of timeseries not obtaining good accuracy
I have developed a LSTM network for multi-class classification of timeseries. The timeseries contains 15 columns (features) and 6 classes (labels) [0, 1, 2, 3, 4, 5]. Features include the 3-phase ...
-2
votes
1
answer
716
views
Why does my Keras LSTM model not learn patterns on the training set and even gets worse over time?
I am trying to fit a sequential model (LSTM) to my data. For this I am using the Keras library.
Simplified, my target is a counter that counts people every day in a location. The features are weather ...
- 361
1
vote
1
answer
144
views
Multistep LSTM: Why am i getting the same curve/pattern for each prediction?
I'm new to tensorflow and LSTM models (and coding in general) and would really appreciate some help. I'm getting somewhere, but no matter what i try i seem to always get the same pattern in each ...
0
votes
0
answers
24
views
Stacked Bidirectional LSTM model
The following is the stacked bidirectional LSTM model using TF2.
class ForexPredictionModel(tf.keras.Model):
def __init__(self, seg_len, **kwargs):
super(ForexPredictionModel, self).__init__(**...
- 7,346
0
votes
0
answers
69
views
Stock Price Prediction: Underperforming LSTM Model Compared to Regular Regression - Seeking Guidance
My dissertation is about Predicting Stock prices. In my case I am using an LSTM model and a various technical indicators. the problem is that now a normal regression model is outperforming my LSTM ...
1
vote
1
answer
75
views
Predicting new timeseries based on related timeseries?
Let's say I have multiple timeseries, representing different features, all of length n, and I want to predict a new timeseries which represents another feature, without any past history for that ...
- 623
1
vote
0
answers
52
views
How to make particular recurrent connection in my Keras/tensorflow neural network model?
I have a LSTM feed-forward neural network as written below. For some reasons, I need to add a backward (recurrent) connection from layer3 to layer1 as well, which results in a loop in my model's ...
- 183
0
votes
0
answers
58
views
How to free memory in lstm
for i in range(0,len(dictStats)):
pkb_ips = dictStats[i:i + 1].index.map(str)[0]
model_path = file_url + model_url + '/model/' + pkb_ips + '.h5'
sc_path = file_url + model_url ...
- 1
0
votes
0
answers
27
views
LSTM model predicting some column correct while rest giving random number
I am working on a AI model trying to predict inahalable particle number among 7 people.
I have 7 columns indicating 7 occupants inhalable particle number over 30 mins. I have multiple cases with ...
- 1
1
vote
0
answers
66
views
In LSTM, the predicted value is shifted to the right by one compared to the actual value
I am doing a time series forecasting with LSTM. As an input feature, the air pollution level in the last 5 hours and the target is to predict the pollution level in the next hour.
It seems to work ...
- 11
0
votes
1
answer
801
views
Predicting future values in a multivariate time forecasting LSTM model that predict future 10 days price on the basis of previous data
I am confused on how to predict future results with a time series multivariate LSTM model.
I am trying to build a model for a stock market prediction and I have the following data features
Date High ...
-1
votes
1
answer
503
views
CNN+LSTM for Video Classification
I am attempting to produce a model that will accept multiple video frames as input and provide a label as output (a.k.a. video classification). I have seen code similar to the below in several ...
0
votes
1
answer
74
views
Tensorflow: Graph of the iterator is different from the graph the dataset was created in
While using the Keras_tuner library I got the error in the title. I have no clue how to fix it and could not find any solutions for some time.
I am using Tensorflow BatchDatasets and passing them for ...
- 1
0
votes
0
answers
226
views
How to use Keras' LSTM to predict at a single timestamp
I currently have a trained LSTM model that I'm using to model a column in Dataframe A. I have all of the training and validation data at my disposal, but I'm writing a method that iterates over ...
- 33