tanu/LSHTM_analysis
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

added option to add confusion matrix and target numbers in the mult function

Browse source
This commit is contained in:
Tanushree Tunstall 2022-06-20 17:08:22 +01:00
parent 905327bf4e
commit 135efcee41
3 changed files with 144 additions and 140 deletions
Download patch file Download diff file Expand all files Collapse all files

117
scripts/ml/MultModelsCl_dissected.py
View file

@ -101,6 +101,9 @@ jacc_score_fn = {'jcc': make_scorer(jaccard_score)}
def MultModelsCl_dissected(input_df, target, skf_cv
, blind_test_input_df
, blind_test_target
, add_cm = True # adds confusion matrix based on cross_val_predict
, add_yn = True # adds target var class numbers
, feature_groups = ['']
, var_type = ['numerical', 'categorical','mixed']):
'''
@ -201,52 +204,88 @@ def MultModelsCl_dissected(input_df, target, skf_cv
, scoring = scoring_fn
, return_train_score = True)
#----------
# check 1
#----------
foo_df = pd.DataFrame.from_dict(skf_cv_modD, orient ='index')
#foo_df = pd.DataFrame.from_dict(skf_cv_modD)
#===================
# Confusion matrix: Not an easy problem to solve! STILL DOING it, USE with caution
#######################################################################
#======================================================
# Option 1: Add confusion matrix from cross_val_predict
# Understand and USE with caution
# cross_val_score, cross_val_predict, "Passing these predictions into an evaluation metric may not be a valid way to measure generalization performance. Results can differ from cross_validate and cross_val_score unless all tests sets have equal size and the metric decomposes over samples."
# https://stackoverflow.com/questions/65645125/producing-a-confusion-matrix-with-cross-validate
#===================
y_pred = cross_val_predict(model_pipeline, input_df, target, cv = 10, **njobs)
#_tn, _fp, _fn, _tp = confusion_matrix(y_pred, y).ravel() # internally
tn, fp, fn, tp = confusion_matrix(y_pred, target).ravel()
# create a dict of confusion matrix that can be appended to the one above
# cmD = {'TN' : np.array(tn)
# , 'FP': np.array(fp)
# , 'FN': np.array(fn)
# , 'TP': np.array(tp)}
#======================================================
if add_cm:
#-----------------------------------------------------------
# Initialise dict of Confusion Matrix (cm)
#-----------------------------------------------------------
cmD = {}
# Calculate cm
y_pred = cross_val_predict(model_pipeline, input_df, target, cv = skf_cv, **njobs)
#_tn, _fp, _fn, _tp = confusion_matrix(y_pred, y).ravel() # internally
tn, fp, fn, tp = confusion_matrix(y_pred, target).ravel()
# Build dict
cmD = {'TN' : tn
, 'FP': fp
, 'FN': fn
, 'TP': tp}
#---------------------------------
# Update cv dict with cmD and tbtD
#----------------------------------
skf_cv_modD.update(cmD)
else:
skf_cv_modD = skf_cv_modD
#######################################################################
#=============================================
# Option 2: Add targety numbers for data
#=============================================
if add_yn:
#-----------------------------------------------------------
# Initialise dict of target numbers: training and blind (tbt)
#-----------------------------------------------------------
tbtD = {}
cmD = {'TN' : tn
, 'FP': fp
, 'FN': fn
, 'TP': tp}
skf_cv_modD.update(cmD)
#----------
# check 2
#----------
#foo2_df = pd.DataFrame.from_dict(skf_cv_modD, orient ='index')
#foo_df = pd.DataFrame.from_dict(skf_cv_modD)
# training y
tyn = Counter(target)
tyn_neg = tyn[0]
tyn_pos = tyn[1]
# blind test y
btyn = Counter(blind_test_target)
btyn_neg = btyn[0]
btyn_pos = btyn[1]
# Build dict
tbtD = {'trainingY_neg' : tyn_neg
, 'trainingY_pos' : tyn_pos
, 'blindY_neg' : btyn_neg
, 'blindY_pos' : btyn_pos}
#---------------------------------
# Update cv dict with cmD and tbtD
#----------------------------------
skf_cv_modD.update(tbtD)
else:
skf_cv_modD = skf_cv_modD
#######################################################################
#==============================
# Extract mean values for CV
#==============================
mm_skf_scoresD[model_name] = {}
for key, value in skf_cv_modD.items():
print('\nkey:', key, '\nvalue:', value)
print('\nmean value:', np.mean(value))
mm_skf_scoresD[model_name][key] = round(np.mean(value),2)
#return(mm_skf_scoresD)
#%%
#=========================
# Blind test: BTS results
#=========================
# Build the final results with all scores for a feature selected model
# Build the final results with all scores for the model
#bts_predict = gscv_fs.predict(blind_test_input_df)
model_pipeline.fit(input_df, target)
bts_predict = model_pipeline.predict(blind_test_input_df)
@ -255,22 +294,6 @@ def MultModelsCl_dissected(input_df, target, skf_cv
print('\nMCC on Blind test:' , bts_mcc_score)
print('\nAccuracy on Blind test:', round(accuracy_score(blind_test_target, bts_predict),2))
# Diff b/w train and bts test scores
#train_test_diff_MCC = cvtrain_mcc - bts_mcc_score
# print('\nDiff b/w train and blind test score (MCC):', train_test_diff)
# # create a dict with all scores
# lr_btsD = { 'model_name': model_name
# , 'bts_mcc':None
# , 'bts_fscore':None
# , 'bts_precision':None
# , 'bts_recall':None
# , 'bts_accuracy':None
# , 'bts_roc_auc':None
# , 'bts_jaccard':None}
mm_skf_scoresD[model_name]['bts_mcc'] = bts_mcc_score
mm_skf_scoresD[model_name]['bts_fscore'] = round(f1_score(blind_test_target, bts_predict),2)
mm_skf_scoresD[model_name]['bts_precision'] = round(precision_score(blind_test_target, bts_predict),2)