mlquantify 0.0.1__py3-none-any.whl

mlquantify/utils/method_purposes/getHist.py

@@ -0,0 +1,13 @@
+import numpy as np
+
+def getHist(scores, nbins):
+
+    breaks = np.linspace(0, 1, int(nbins)+1)
+    breaks = np.delete(breaks, -1)
+    breaks = np.append(breaks,1.1)
+    
+    re = np.repeat(1/(len(breaks)-1), (len(breaks)-1))  
+    for i in range(1,len(breaks)):
+        re[i-1] = (re[i-1] + len(np.where((scores >= breaks[i-1]) & (scores < breaks[i]))[0]) ) / (len(scores)+ 1)
+
+    return re

mlquantify/utils/method_purposes/get_scores.py

@@ -0,0 +1,33 @@
+import pandas as pd
+import numpy as np
+from sklearn.model_selection import StratifiedKFold
+
+def get_scores(X, y, learner, folds:int=10, learner_fitted:bool=False) -> tuple:
+    if isinstance(X, np.ndarray):
+        X = pd.DataFrame(X)
+    if isinstance(y, np.ndarray):
+        y = pd.Series(y)
+        
+    if learner_fitted:
+        probabilities = learner.predict_proba(X)
+        y_label = y
+    else:
+    
+        skf = StratifiedKFold(n_splits=folds)    
+        probabilities = []
+        y_label = []
+        
+        for train_index, valid_index in skf.split(X,y):
+            
+            tr_data = pd.DataFrame(X.iloc[train_index])   #Train data and labels
+            tr_label = y.iloc[train_index]
+            
+            valid_data = pd.DataFrame(X.iloc[valid_index])  #Validation data and labels
+            valid_label = y.iloc[valid_index]
+            
+            learner.fit(tr_data, tr_label)
+            
+            probabilities.extend(learner.predict_proba(valid_data))     #evaluating scores
+            y_label.extend(valid_label)
+    
+    return np.asarray(y_label), np.asarray(probabilities)

mlquantify/utils/method_purposes/moss.py

@@ -0,0 +1,16 @@
+import numpy as np
+
+def MoSS(n:int, alpha:float, m:float):
+  
+  n_pos = int(n*alpha)
+  n_neg = int((1-alpha)*n)
+
+  x_pos = np.arange(1, n_pos, 1)
+  x_neg = np.arange(1, n_neg, 1)
+  
+  syn_plus = np.power(x_pos/(n_pos+1), m)
+  syn_neg = 1 - np.power(x_neg/(n_neg+1), m)
+
+  #moss = np.union1d(syn_plus, syn_neg)
+
+  return syn_plus, syn_neg

mlquantify/utils/method_purposes/ternary_search.py

@@ -0,0 +1,14 @@
+def ternary_search(left, right, f, eps=1e-4):
+    """This function applies Ternary search"""
+
+    while True:
+        if abs(left - right) < eps:
+            return(left + right) / 2
+    
+        leftThird  = left + (right - left) / 3
+        rightThird = right - (right - left) / 3
+    
+        if f(leftThird) > f(rightThird):
+            left = leftThird
+        else:
+            right = rightThird 

mlquantify/utils/method_purposes/tprfpr.py

@@ -0,0 +1,42 @@
+import numpy as np
+
+
+def compute_table(y, y_pred, classes):
+    TP = np.logical_and(y == y_pred, y == classes[1]).sum()
+    FP = np.logical_and(y != y_pred, y == classes[0]).sum()
+    FN = np.logical_and(y != y_pred, y == classes[1]).sum()
+    TN = np.logical_and(y == y_pred, y == classes[0]).sum()
+    return TP, FP, FN, TN
+
+
+def compute_tpr(TP, FN):
+    if TP + FN == 0:
+        return 0
+    return TP / (TP + FN)
+
+
+def compute_fpr(FP, TN):
+    if FP + TN == 0:
+        return 0
+    return FP / (FP + TN)
+
+
+def adjust_threshold(y, probabilities:np.ndarray, classes:np.ndarray) -> tuple:
+    unique_scores = np.linspace(0, 1, 101)
+    
+    tprs = []
+    fprs = []
+    
+    for threshold in unique_scores:
+        y_pred = np.where(probabilities >= threshold, classes[1], classes[0])
+        
+        TP, FP, FN, TN = compute_table(y, y_pred, classes)
+        
+        tpr = compute_tpr(TP, FN)
+        fpr = compute_fpr(FP, TN)
+        
+        tprs.append(tpr)
+        fprs.append(fpr)
+    
+    #best_tpr, best_fpr = self.adjust_threshold(np.asarray(tprs), np.asarray(fprs))
+    return (unique_scores, np.asarray(tprs), np.asarray(fprs))

mlquantify-0.0.1.dist-info/METADATA

@@ -0,0 +1,23 @@
+Metadata-Version: 2.1
+Name: mlquantify
+Version: 0.0.1
+Summary: Quantification Library
+Home-page: https://github.com/luizfernandolj/QuantifyML/tree/master
+Maintainer: Luiz Fernando Luth Junior
+Keywords: python,machine learning,quantification,quantify
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Science/Research
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: Unix
+Classifier: Operating System :: MacOS :: MacOS X
+Classifier: Operating System :: Microsoft :: Windows
+Description-Content-Type: text/markdown
+Requires-Dist: scikit-learn
+Requires-Dist: numpy
+Requires-Dist: scipy
+Requires-Dist: joblib
+Requires-Dist: tqdm
+Requires-Dist: pandas
+Requires-Dist: xlrd
+Requires-Dist: matplotlib
+

mlquantify-0.0.1.dist-info/RECORD

@@ -0,0 +1,73 @@
+mlquantify/__init__.py,sha256=BGxGCeZhlNsTYZkLoJJ-zcpVDRU4jBFVsz9ZmEZCtvs,166
+mlquantify/base.py,sha256=XpI7vynnn-f126Ykkw_YuuZI0CtuRMkPt0oBbDDZ7yY,8371
+mlquantify/model_selection.py,sha256=zN-qkrCe35GoJ8u1_Ab3EKXvrnAppk68Fj5kQVi61nU,8617
+mlquantify/classification/__init__.py,sha256=zxE6_ouh3kJVL56aJAhpFh36EBWKq9U4IcID4noROPQ,26
+mlquantify/classification/pwkclf.py,sha256=K1XBgI9FIh1Yu571bzdw8PM-TJXFGOQAzP4ewDV9b3Y,2323
+mlquantify/evaluation/__init__.py,sha256=_7orbRyYuJydD_W9o5l9Ko_-EopGGhewekoEhR82S3Q,48
+mlquantify/evaluation/measures/__init__.py,sha256=JvxIm3k31EFBzm-4Guj9Huh--B7u-o3gvln3VSFK2nk,761
+mlquantify/evaluation/measures/ae.py,sha256=eY8oKfj7TIpbNCsEfbkbeVVduSAaiKE5Vxy7xACwrL8,363
+mlquantify/evaluation/measures/bias.py,sha256=AmTZ9dP5XW6ZR-gdBfOF008IzVFeBNeL2SIuSLkABQc,497
+mlquantify/evaluation/measures/kld.py,sha256=f443cf31Sm904zEwk0q1093XwvEEQbroHUzNLKJj-BI,341
+mlquantify/evaluation/measures/mse.py,sha256=VS38G83ygVz_qxtsjqa-tTGbPVERQOraY_aga5JmVuo,408
+mlquantify/evaluation/measures/nae.py,sha256=D5W2qiD1EC53uy7AILRHvdZNvAcAQkLTuFNFNmHjkEs,499
+mlquantify/evaluation/measures/nkld.py,sha256=UsTDZP-CjiEgQjfyoKynZntpFV8msYPrEma3GtHK1-s,491
+mlquantify/evaluation/measures/nrae.py,sha256=iTxEG7k6DQRieSbwKrHfcynmbRR-qCviudDecoRkmdk,574
+mlquantify/evaluation/measures/rae.py,sha256=KYa9HY7k5w3acMp5yUSMVuwZIdzeUB9CzkSA-jpXn7Y,416
+mlquantify/evaluation/measures/se.py,sha256=aScxX56dmbSj1HxubD8Dlq4gTmbNrA-H3AdE145wgDI,396
+mlquantify/evaluation/protocol/_Protocol.py,sha256=XsLsu-sDQxHVyRqc-vGAs4407H5IJGSYAuH4sWAYQL8,7752
+mlquantify/evaluation/protocol/__init__.py,sha256=RO3fHIvhAK95t8EUaD67JfJV2WeLb17HMqbXo3LcUSI,42
+mlquantify/evaluation/protocol/app.py,sha256=WJ2OaYe49R8JqCoZDy6gMN50eIQ7jThPDT5W5lSRlK0,5713
+mlquantify/evaluation/protocol/npp.py,sha256=c287_9LmxYvv72hh2iwoCwSkwqTeFvknKOWWG4Jq0QY,1117
+mlquantify/methods/__init__.py,sha256=rvpN5T6vXdfX1bOHKYGC0aKZMc5bHSPb-pCVQ3mlb08,627
+mlquantify/methods/aggregative/__init__.py,sha256=rBtSfgTsTIjbH5rJhQ9V05n68o-9hm9QNZzxRCh2VME,209
+mlquantify/methods/aggregative/cc.py,sha256=HXFgcLqyM35CTszBn5UmUmT60Z1J1Ow0i4lZf5BctH8,1123
+mlquantify/methods/aggregative/emq.py,sha256=3N4PQhTlBu6vBwOaEpcNZf2QYNBF2XC8cCH3qsyKt9Y,3208
+mlquantify/methods/aggregative/fm.py,sha256=RgXVwsdoZsWV_DanvzcsRmll_e2Etvl27PNI7iYM9Nw,2912
+mlquantify/methods/aggregative/gac.py,sha256=gTIRqMN0S7teheSOgrk2NB80UxfC-KfbNuk0A1tflh8,3640
+mlquantify/methods/aggregative/gpac.py,sha256=VcDCcEYXiknEOTwVDGOXWeZCJS9T6vd4YFqEbnUB3YU,3531
+mlquantify/methods/aggregative/pcc.py,sha256=8WuiWFrYvskx0Xd2wTDgcn-GyswwB5hv-VH6HaRkilY,1251
+mlquantify/methods/aggregative/pwk.py,sha256=JCUzS8hrsiFMacLh00FR4Xhk7zhe4Vy4za-0QeFr-l8,1466
+mlquantify/methods/aggregative/ThreholdOptm/_ThreholdOptimization.py,sha256=amgklOfPe-htQIhuDb3l4usqj-cVkGFa7z0KXYmibUM,2433
+mlquantify/methods/aggregative/ThreholdOptm/__init__.py,sha256=ALB-ZEl9M_DyHr3U9PEw6E9XSMFfwci_aSwbpyzvp0E,155
+mlquantify/methods/aggregative/ThreholdOptm/acc.py,sha256=tgcMhYW9cD66j0T6mWDulaxP9-FA1mNozvlaKMQV2UQ,1082
+mlquantify/methods/aggregative/ThreholdOptm/max.py,sha256=A-_0RQ-HAiTE2FfeNvcInqauXpzn8xtSf0LDiuC5wDk,824
+mlquantify/methods/aggregative/ThreholdOptm/ms.py,sha256=BpVrDaiCZa8xe1_YX9qhGCsOAcCXZ_3EQF5ZCHWRCn8,739
+mlquantify/methods/aggregative/ThreholdOptm/ms2.py,sha256=Pxxc4p_41KLcNUaN0fhjih3mSVMJhoZrm-KUy45paT4,852
+mlquantify/methods/aggregative/ThreholdOptm/pacc.py,sha256=zfRQhklB8NuvQAn8-a1hJ9sm6iG6yRGIWh-xlLTnU3M,1432
+mlquantify/methods/aggregative/ThreholdOptm/t50.py,sha256=-zWNDXdGpW7CR-9O6MUQYOBqMuH15lLV-uK0u8nIAYA,728
+mlquantify/methods/aggregative/ThreholdOptm/x.py,sha256=Lwn5MtaL9tb4eRFMrFxerUjmumh2US90KvDfdNQxvF8,759
+mlquantify/methods/aggregative/mixtureModels/_MixtureModel.py,sha256=pcnCN0IArb_AJ71AyUnIt0uVPhN9IWnXZF9ad1d0g8c,3336
+mlquantify/methods/aggregative/mixtureModels/__init__.py,sha256=F7QvGhVkG7NEHjnau_fNm7T9MLCUUDwfrrNvncqPpPI,117
+mlquantify/methods/aggregative/mixtureModels/dys.py,sha256=JC3h_gdIYG-5pphl7C6Ez8Kbp0UPbccqX8ZVcsTAI3g,2356
+mlquantify/methods/aggregative/mixtureModels/dys_syn.py,sha256=GGt9TOXGUBBD9PdiozT6QpC-IGa6QvLf1QAGEI7El8s,3906
+mlquantify/methods/aggregative/mixtureModels/hdy.py,sha256=JykK4Fg5yPKmEuL1aAqsG9wnrmTbXNuxbK4ugI1XNBo,1986
+mlquantify/methods/aggregative/mixtureModels/smm.py,sha256=2r82umOXqOrffVO-SdnyUzsJKg6hz5_z82HzNsn7GRY,1050
+mlquantify/methods/aggregative/mixtureModels/sord.py,sha256=Kz5YAXd3RezeTI66wfrnXxY_AXDos44WS8N2fYE4oxA,3166
+mlquantify/methods/meta/__init__.py,sha256=uj--leWdOrpBs0u_djhCYlLCrQo3GIxg0AoJyAkHC-c,30
+mlquantify/methods/meta/ensemble.py,sha256=rX9fyGi6xMYVUo0jm_rttnSDTDt_XokvzZJUWDPxQTM,10396
+mlquantify/methods/non_aggregative/__init__.py,sha256=HD34lwNmAkw26qlEA5PDHay9V6ShgFiQvlPipOsrDAs,20
+mlquantify/methods/non_aggregative/hdx.py,sha256=vwodV_Z89_be8HBOSriX3lsFhzmPZVLdOgKewa6la-E,2420
+mlquantify/plots/__init__.py,sha256=IDnv_KVALIsVuohiudclnIMSM5F1dNRxUFpjygLblBI,118
+mlquantify/plots/distribution_plot.py,sha256=8l22Lq6LhlOd3yv6gHX6wg2p1MTozYASyNzPcLlPHzk,4038
+mlquantify/plots/protocol_plot.py,sha256=DoVg_oDO_8HcFHMmWBVY2TM9HR-sgw20EqPQAOVRKj4,5513
+mlquantify/utils/__init__.py,sha256=VKplOsrL4ONf0-9anhcWFOxXjvGSQTH0Kh922s_hGb0,63
+mlquantify/utils/general_purposes/__init__.py,sha256=-Pwx1t2M_rGCEHxIc2TyomSe92w6U5Bdp6wtpsdcMZI,380
+mlquantify/utils/general_purposes/convert_col_to_array.py,sha256=Pq6_U6BdsIAl4Vr_836_vM-g2iRQAJ1b4kGOsYoFukI,560
+mlquantify/utils/general_purposes/generate_artificial_indexes.py,sha256=L_Sb1hBYGllCxDbYKRNY_qqjCfSbbkXo6E1XuaWxQMs,1049
+mlquantify/utils/general_purposes/get_real_prev.py,sha256=nHSXaxAdp-rznyhauN7ta_1QCDrlfCZSJWxxQQ4_pvE,266
+mlquantify/utils/general_purposes/load_quantifier.py,sha256=6BfObUBVJ7_v-pzP-cQlMpblmg6ygpvDRTimOhMWepw,77
+mlquantify/utils/general_purposes/make_prevs.py,sha256=lamVz-MTRDP4bqDjS4xm2SMa3YpnMcL__1OF6ZpEX2I,661
+mlquantify/utils/general_purposes/normalize.py,sha256=kIRG22249IkBuDBGP-qQbLsH-edPFC8gPnJS_huxLhY,785
+mlquantify/utils/general_purposes/parallel.py,sha256=vkert2m-A5Rs_VKhLUk84nZ0scIVPFadvVsTF2bItaU,218
+mlquantify/utils/general_purposes/round_protocol_df.py,sha256=74zS7ws7mgzjObv6xoQi7rvH2xskschHF93czoFaVFY,474
+mlquantify/utils/method_purposes/__init__.py,sha256=MFB3LtYTsWuETmyFokn8EZoK2wwPZspCknIMdutLLCo,276
+mlquantify/utils/method_purposes/distances.py,sha256=l5neBoqLN_Npyf2L9VB19V_RHpB2w9AJuddTQ2pFzxU,428
+mlquantify/utils/method_purposes/getHist.py,sha256=dgO57Wo-Sbix6h1ixbT2XIKuRU0p-lZNLGI0ZGwadkM,402
+mlquantify/utils/method_purposes/get_scores.py,sha256=qdIVYUS8xd8Vt86k19yETDNfibTXaTur5pCrMA1b254,1151
+mlquantify/utils/method_purposes/moss.py,sha256=CVDDMHxPBnl_U2hz7Aqvne7jhB2mBUsVzTTsaiLQhOc,352
+mlquantify/utils/method_purposes/ternary_search.py,sha256=JpNrfJsA5kWuanVW_hyMucy7rQ9UzTSgazFpTRi9jMI,416
+mlquantify/utils/method_purposes/tprfpr.py,sha256=VKniG5aK8IwAA2fXEhkdHtwnx1zHH12qhwS4kKW5Dlo,1181
+mlquantify-0.0.1.dist-info/METADATA,sha256=hT32pA_BUXMWzHfQ12L9LmVzlc1plpWG8rVPSYSxPwE,774
+mlquantify-0.0.1.dist-info/WHEEL,sha256=yQN5g4mg4AybRjkgi-9yy4iQEFibGQmlz78Pik5Or-A,92
+mlquantify-0.0.1.dist-info/top_level.txt,sha256=tGEkYkbbFElwULvqENjam3u1uXtyC1J9dRmibsq8_n0,11
+mlquantify-0.0.1.dist-info/RECORD,,

mlquantify-0.0.1.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: bdist_wheel (0.41.2)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

mlquantify-0.0.1.dist-info/top_level.txt

@@ -0,0 +1 @@
+mlquantify