Flexible LSTM Evaluator Wrapper
A simple wrapper to flexibly create a Sequence Evaluator Machine learning using LSTM Networks in TfLearn.
It can be use to evaluate any sequences just by tuning the parameters and training the net.
The trained model will be automatically saved each run. Batch size, number of epochs and temperature for generations can be passed as commandline parameters, as the number of hidden lstm layers.
usage: lstm_DNN.py [-h] --dataset DATASET [--batch_size BATCH_SIZE]
[--epochs EPOCHS]
[--model_file MODEL_FILE]
[--hidden_layer_size HIDDEN_LAYER_SIZE]
[--lstm_node_size LSTM_NODE_SIZE]
[--max_sequence_lenght MAX_SEQUENCE_LENGHT]
optional arguments:
-h, --help show this help message and exit
--dataset DATASET Path to the dataset file
--batch_size BATCH_SIZE
How many string train on at a time
--epochs EPOCHS How many epochs to train
--model_file MODEL_FILE
Path to save the model file, will be loaded if present
or created
--hidden_layer_size HIDDEN_LAYER_SIZE
Number of hidden lstm layers
--lstm_node_size LSTM_NODE_SIZE
Number of nodes in each lstm layers
--max_sequence_lenght MAX_SEQUENCE_LENGHT
Max lenght of sequence sample. if 0 is set to the
maximum sequence lenght from dataset (Warning:
requires serious amount of Memory)