LSHTM_analysis/scripts/ml/ml_iterator_fs.py

156 lines
7.4 KiB
Python

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed Jun 29 20:29:36 2022
@author: tanu
"""
import sys, os
import pandas as pd
import numpy as np
import re
#import prettyprint as pp
###############################################################################
homedir = os.path.expanduser("~")
sys.path.append(homedir + '/git/LSHTM_analysis/scripts/ml/ml_functions')
sys.path
###############################################################################
#====================
# Import ML functions
#====================
from MultClfs import *
from GetMLData import *
from SplitTTS import *
from FS import *
# param dict for getmldata()
combined_model_paramD = {'data_combined_model' : False
, 'use_or' : False
, 'omit_all_genomic_features': False
, 'write_maskfile' : False
, 'write_outfile' : False }
###############################################################################
#ml_genes = ["pncA", "embB", "katG", "rpoB", "gid"]
outdir = homedir + '/git/Data/ml_combined/fs/'
ml_gene_drugD = {'pncA' : 'pyrazinamide'
# , 'embB' : 'ethambutol'
# , 'katG' : 'isoniazid'
# , 'rpoB' : 'rifampicin'
# , 'gid' : 'streptomycin'
}
gene_dataD={}
#split_types = ['70_30', '80_20', 'sl']
#split_data_types = ['actual', 'complete']
split_types = ['70_30']
split_data_types = ['actual', 'complete']
fs_models = [('Logistic Regression' , LogisticRegression(**rs) )]
# fs_models = [('AdaBoost Classifier' , AdaBoostClassifier(**rs) )
# , ('Decision Tree' , DecisionTreeClassifier(**rs) )
# , ('Extra Tree' , ExtraTreeClassifier(**rs) )
# , ('Extra Trees' , ExtraTreesClassifier(**rs) )
# , ('Gradient Boosting' , GradientBoostingClassifier(**rs) )
# , ('LDA' , LinearDiscriminantAnalysis() )
# , ('Logistic Regression' , LogisticRegression(**rs) )
# , ('Logistic RegressionCV' , LogisticRegressionCV(cv = 3, **rs))
# , ('Passive Aggresive' , PassiveAggressiveClassifier(**rs, **njobs) )
# , ('Random Forest' , RandomForestClassifier(**rs, n_estimators = 1000 ) )
# , ('Ridge Classifier' , RidgeClassifier(**rs) )
# , ('Ridge ClassifierCV' , RidgeClassifierCV(cv = 3) )
# , ('Stochastic GDescent' , SGDClassifier(**rs, **njobs) )
# ]
for gene, drug in ml_gene_drugD.items():
print ('\nGene:', gene
, '\nDrug:', drug)
gene_low = gene.lower()
gene_dataD[gene_low] = getmldata(gene, drug
, data_combined_model = False # this means it doesn't include 'gene_name' as a feauture as a single gene-target shouldn't have it.
, use_or = False
, omit_all_genomic_features = False
, write_maskfile = False
, write_outfile = False)
for split_type in split_types:
for data_type in split_data_types:
# unused per-split outfile
out_filename = outdir + gene.lower() + '_'+split_type+'_' + data_type + '.json'
tempD=split_tts(gene_dataD[gene_low]
, data_type = data_type
, split_type = split_type
, oversampling = True # TURN IT ON TO RUN THE OTHERS BIS
, dst_colname = 'dst'
, target_colname = 'dst_mode'
, include_gene_name = True
)
paramD = {
'baseline_paramD': { 'input_df' : tempD['X']
, 'target' : tempD['y']
, 'var_type' : 'mixed'
, 'resampling_type': 'none'}
,'smnc_paramD': { 'input_df' : tempD['X_smnc']
, 'target' : tempD['y_smnc']
, 'var_type' : 'mixed'
, 'resampling_type' : 'smnc'}
# , 'ros_paramD': { 'input_df' : tempD['X_ros']
# , 'target' : tempD['y_ros']
# , 'var_type' : 'mixed'
# , 'resampling_type' : 'ros'}
# , 'rus_paramD' : { 'input_df' : tempD['X_rus']
# , 'target' : tempD['y_rus']
# , 'var_type' : 'mixed'
# , 'resampling_type' : 'rus'}
# , 'rouC_paramD' : { 'input_df' : tempD['X_rouC']
# , 'target' : tempD['y_rouC']
# , 'var_type' : 'mixed'
# , 'resampling_type': 'rouC'}
}
#for m in fs_models:
# print(m)
out_fsD = {}
index = 1
for model_name, model_fn in fs_models:
print('\nRunning classifier with FS:', index
, '\nModel_name:' , model_name
, '\nModel func:' , model_fn)
#, '\nList of models:', models)
index = index+1
out_fsD[model_name] = {}
# current_model = {}
for k, v in paramD.items():
# out_filename = (gene.lower() + '_' + split_type + '_' + data_type + '_' + k + '.json')
fsD_params=paramD[k]
# print("XXXXXX THIS: ", len(fsD_params['input_df']) )
# print("XXXXXX THIS: ", out_filename )
# current_model[k] = fsgs_rfecv(
out_fsD[model_name][k] = fsgs_rfecv(
**fsD_params
, param_gridLd = [{'fs__min_features_to_select': [1]}]
, blind_test_df = tempD['X_bts']
, blind_test_target = tempD['y_bts']
, estimator = model_fn
, use_fs = False # uses estimator as the RFECV parameter for fs. Set to TRUE if you want to supply custom_fs as shown below
, custom_fs = RFECV(DecisionTreeClassifier(**rs), cv = skf_cv, scoring = 'matthews_corrcoef')
, cv_method = skf_cv
)
# write per-resampler outfile here
# with open(out_filename, 'w') as f:
# f.write(json.dumps(current_model
# , cls = NpEncoder )
# )
# write per-split outfile here
with open(out_filename, 'w') as f:
f.write(json.dumps(out_fsD
#, cls = NpEncoder
))
#%%############################################################################
# # Read output json
# testF = outdir + 'pnca_70_30_actual.json'
# testF = outdir + 'pnca_70_30_complete.json'
# with open(testF, 'r') as f:
# data = json.load(f)