Paper under review
usage: train.py [-h] [--datafolder DATAFOLDER] [--modelfile MODELFILE]
[--cuda 1 or 0] [--train TRAIN] [--predict_size PREDICT_SIZE]
[--minibatches_per_step MINIBATCHES_PER_STEP]
[--minibatch_size MINIBATCH_SIZE] [--epoch EPOCH]
[--learning_rate LEARNING_RATE] [--data_split DATA_SPLIT]
[--data_kept DATA_KEPT] [--source_scale SOURCE_SCALE]
[--source_bias SOURCE_BIAS] [--downsampling DOWNSAMPLING]
[--noiselevel NOISELEVEL] [--pdfheader PDFHEADER]
SMFS Event Detect
optional arguments:
-h, --help show this help message and exit
--datafolder DATAFOLDER
folder for data (.np) files
--modelfile MODELFILE
folder for model (.pt) files
--cuda 1 or 0 use cuda
--train TRAIN set 1 to train the model, set 0 to test the trained
model
--predict_size PREDICT_SIZE
predict_size for predicting
--minibatches_per_step MINIBATCHES_PER_STEP
minibatches_per_step for training
--minibatch_size MINIBATCH_SIZE
minibatch_size for training
--epoch EPOCH epochs for training
--learning_rate LEARNING_RATE
learning_rate for training
--data_split DATA_SPLIT
data_split for truncating dataset
--data_kept DATA_KEPT
data_kept for truncating dataset
--source_scale SOURCE_SCALE
source_scale in nm and pN for transforming input
signals in the dataset
--source_bias SOURCE_BIAS
source_bias after applying source_scale for
transforming input signals in the dataset
--downsampling DOWNSAMPLING
perform downsampling using averaging filter on input
data
--noiselevel NOISELEVEL
add extra Gaussian noise (level in nm and pN) into
input dataset
--report REPORT folder for saving repots
--report_note REPORT_NOTE
add prefix to each report file