#!/usr/bin/env python3
"""Module containing the PredictNeuralNetwork class and the command line interface."""
import argparse
import h5py
import json
import csv
import numpy as np
import pandas as pd
from biobb_common.generic.biobb_object import BiobbObject
from tensorflow.python.keras.saving import hdf5_format
from sklearn.preprocessing import scale
from biobb_common.configuration import settings
from biobb_common.tools import file_utils as fu
from biobb_common.tools.file_utils import launchlogger
from biobb_ml.neural_networks.common import check_input_path, check_output_path, getHeader, getTargetValue, get_list_of_predictors, get_keys_of_predictors, get_num_cols
[docs]class PredictNeuralNetwork(BiobbObject):
"""
| biobb_ml PredictNeuralNetwork
| Makes predictions from an input dataset and a given model.
| Makes predictions from an input dataset (provided either as a file or as a dictionary property) and a given model trained with `TensorFlow Keras Sequential <https://www.tensorflow.org/api_docs/python/tf/keras/Sequential>`_ and `TensorFlow Keras LSTM <https://www.tensorflow.org/api_docs/python/tf/keras/layers/LSTM>`_
Args:
input_model_path (str): Path to the input model. File type: input. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/data/neural_networks/input_model_predict.h5>`_. Accepted formats: h5 (edam:format_3590).
input_dataset_path (str) (Optional): Path to the dataset to predict. File type: input. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/data/neural_networks/dataset_predict.csv>`_. Accepted formats: csv (edam:format_3752).
output_results_path (str): Path to the output results file. File type: output. `Sample file <https://github.com/bioexcel/biobb_ml/raw/master/biobb_ml/test/reference/neural_networks/ref_output_predict.csv>`_. Accepted formats: csv (edam:format_3752).
properties (dic - Python dictionary object containing the tool parameters, not input/output files):
* **predictions** (*list*) - (None) List of dictionaries with all values you want to predict targets. It will be taken into account only in case **input_dataset_path** is not provided. Format: [{ 'var1': 1.0, 'var2': 2.0 }, { 'var1': 4.0, 'var2': 2.7 }] for datasets with headers and [[ 1.0, 2.0 ], [ 4.0, 2.7 ]] for datasets without headers.
* **remove_tmp** (*bool*) - (True) [WF property] Remove temporal files.
* **restart** (*bool*) - (False) [WF property] Do not execute if output files exist.
Examples:
This is a use example of how to use the building block from Python::
from biobb_ml.neural_networks.neural_network_predict import neural_network_predict
prop = {
'predictions': [
{
'var1': 1.0,
'var2': 2.0
},
{
'var1': 4.0,
'var2': 2.7
}
]
}
neural_network_predict(input_model_path='/path/to/myModel.h5',
input_dataset_path='/path/to/myDataset.csv',
output_results_path='/path/to/newPredictedResults.csv',
properties=prop)
Info:
* wrapped_software:
* name: TensorFlow
* version: >2.1.0
* license: MIT
* ontology:
* name: EDAM
* schema: http://edamontology.org/EDAM.owl
"""
def __init__(self, input_model_path, output_results_path,
input_dataset_path=None, properties=None, **kwargs) -> None:
properties = properties or {}
# Call parent class constructor
super().__init__(properties)
self.locals_var_dict = locals().copy()
# Input/Output files
self.io_dict = {
"in": {"input_model_path": input_model_path, "input_dataset_path": input_dataset_path},
"out": {"output_results_path": output_results_path}
}
# Properties specific for BB
self.predictions = properties.get('predictions', [])
self.properties = properties
# Check the properties
self.check_properties(properties)
self.check_arguments()
[docs] def check_data_params(self, out_log, err_log):
""" Checks all the input/output paths and parameters """
self.io_dict["in"]["input_model_path"] = check_input_path(self.io_dict["in"]["input_model_path"], "input_model_path", False, out_log, self.__class__.__name__)
self.io_dict["out"]["output_results_path"] = check_output_path(self.io_dict["out"]["output_results_path"], "output_results_path", False, out_log, self.__class__.__name__)
if self.io_dict["in"]["input_dataset_path"]:
self.io_dict["in"]["input_dataset_path"] = check_input_path(self.io_dict["in"]["input_dataset_path"], "input_dataset_path", False, out_log, self.__class__.__name__)
[docs] @launchlogger
def launch(self) -> int:
"""Execute the :class:`PredictNeuralNetwork <neural_networks.neural_network_predict.PredictNeuralNetwork>` neural_networks.neural_network_predict.PredictNeuralNetwork object."""
# check input/output paths and parameters
self.check_data_params(self.out_log, self.err_log)
# Setup Biobb
if self.check_restart():
return 0
self.stage_files()
fu.log('Getting model from %s' % self.io_dict["in"]["input_model_path"], self.out_log, self.global_log)
with h5py.File(self.io_dict["in"]["input_model_path"], mode='r') as f:
variables = f.attrs['variables']
new_model = hdf5_format.load_model_from_hdf5(f)
# get dictionary with variables
vars_obj = json.loads(variables)
stringlist = []
new_model.summary(print_fn=lambda x: stringlist.append(x))
model_summary = "\n".join(stringlist)
fu.log('Model summary:\n\n%s\n' % model_summary, self.out_log, self.global_log)
if self.io_dict["in"]["input_dataset_path"]:
# load dataset from input_dataset_path file
fu.log('Getting dataset from %s' % self.io_dict["in"]["input_dataset_path"], self.out_log, self.global_log)
if 'features' not in vars_obj:
# recurrent
labels = None
skiprows = None
with open(self.io_dict["in"]["input_dataset_path"]) as csvfile:
reader = csv.reader(csvfile, quoting=csv.QUOTE_NONNUMERIC) # change contents to floats
for row in reader: # each row is a list
self.predictions.append(row)
else:
# classification or regression
if 'columns' in vars_obj['features']:
labels = getHeader(self.io_dict["in"]["input_dataset_path"])
skiprows = 1
else:
labels = None
skiprows = None
new_data_table = pd.read_csv(self.io_dict["in"]["input_dataset_path"], header=None, sep="\\s+|;|:|,|\t", engine="python", skiprows=skiprows, names=labels)
else:
if vars_obj['type'] != 'recurrent':
new_data_table = pd.DataFrame(data=get_list_of_predictors(self.predictions), columns=get_keys_of_predictors(self.predictions))
else:
new_data_table = pd.DataFrame(data=self.predictions, columns=get_num_cols(vars_obj['window_size']))
# prediction
if vars_obj['type'] != 'recurrent':
# classification or regression
# new_data_table = pd.DataFrame(data=get_list_of_predictors(self.predictions),columns=get_keys_of_predictors(self.predictions))
new_data = new_data_table
if vars_obj['scale']:
new_data = scale(new_data)
predictions = new_model.predict(new_data)
predictions = np.around(predictions, decimals=2)
clss = ''
# if predictions.shape[1] > 1:
if vars_obj['type'] == 'classification':
# classification
pr = tuple(map(tuple, predictions))
clss = ' (' + ', '.join(str(x) for x in vars_obj['vs']) + ')'
else:
# regression
pr = np.squeeze(np.asarray(predictions))
new_data_table[getTargetValue(vars_obj['target']) + clss] = pr
else:
# recurrent
# new_data_table = pd.DataFrame(data=self.predictions, columns=get_num_cols(vars_obj['window_size']))
predictions = []
for r in self.predictions:
row = np.asarray(r).reshape((1, vars_obj['window_size'], 1))
pred = new_model.predict(row)
pred = np.around(pred, decimals=2)
predictions.append(pred[0][0])
# pd.set_option('display.float_format', lambda x: '%.2f' % x)
new_data_table["predictions"] = predictions
fu.log('Predicting results\n\nPREDICTION RESULTS\n\n%s\n' % new_data_table, self.out_log, self.global_log)
fu.log('Saving results to %s' % self.io_dict["out"]["output_results_path"], self.out_log, self.global_log)
new_data_table.to_csv(self.io_dict["out"]["output_results_path"], index=False, header=True, float_format='%.3f')
# Copy files to host
self.copy_to_host()
self.tmp_files.extend([
self.stage_io_dict.get("unique_dir")
])
self.remove_tmp_files()
self.check_arguments(output_files_created=True, raise_exception=False)
return 0
[docs]def neural_network_predict(input_model_path: str, output_results_path: str, input_dataset_path: str = None, properties: dict = None, **kwargs) -> int:
"""Execute the :class:`PredictNeuralNetwork <neural_networks.neural_network_predict.PredictNeuralNetwork>` class and
execute the :meth:`launch() <neural_networks.neural_network_predict.PredictNeuralNetwork.launch>` method."""
return PredictNeuralNetwork(input_model_path=input_model_path,
output_results_path=output_results_path,
input_dataset_path=input_dataset_path,
properties=properties, **kwargs).launch()
[docs]def main():
"""Command line execution of this building block. Please check the command line documentation."""
parser = argparse.ArgumentParser(description="Makes predictions from an input dataset and a given classification model.", formatter_class=lambda prog: argparse.RawTextHelpFormatter(prog, width=99999))
parser.add_argument('--config', required=False, help='Configuration file')
# Specific args of each building block
required_args = parser.add_argument_group('required arguments')
required_args.add_argument('--input_model_path', required=True, help='Path to the input model. Accepted formats: h5.')
required_args.add_argument('--output_results_path', required=True, help='Path to the output results file. Accepted formats: csv.')
parser.add_argument('--input_dataset_path', required=False, help='Path to the dataset to predict. Accepted formats: csv.')
args = parser.parse_args()
args.config = args.config or "{}"
properties = settings.ConfReader(config=args.config).get_prop_dic()
# Specific call of each building block
neural_network_predict(input_model_path=args.input_model_path,
output_results_path=args.output_results_path,
input_dataset_path=args.input_dataset_path,
properties=properties)
if __name__ == '__main__':
main()