Add files using upload-large-folder tool

logistic_regression/OpenLID-v2/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       1.00      0.99      0.99      7223
+          da       0.98      0.98      0.98     20007
+          de       1.00      1.00      1.00     19996
+          el       1.00      1.00      1.00     20025
+          en       0.99      1.00      0.99     19981
+          es       1.00      1.00      1.00     20062
+          fi       0.99      0.99      0.99     19949
+          fr       1.00      1.00      1.00     20039
+          hr       0.98      0.98      0.98     19986
+          it       1.00      1.00      1.00     19953
+          ja       0.97      0.80      0.88     19983
+          ko       0.83      0.99      0.90     19990
+          lt       0.99      0.99      0.99     19939
+          mk       1.00      1.00      1.00     20014
+          nb       0.99      0.99      0.99     19962
+          nl       0.98      0.98      0.98     19987
+          pl       0.98      0.99      0.99     20033
+          pt       0.99      0.99      0.99     20037
+          ro       1.00      0.99      1.00     19961
+          ru       1.00      0.99      1.00     20029
+          sl       0.97      0.97      0.97     19959
+          sv       0.98      0.98      0.98     19922
+          uk       1.00      1.00      1.00     20037
+          zh       0.00      0.00      0.00         0
+
+    accuracy                           0.98    447074
+   macro avg       0.94      0.94      0.94    447074
+weighted avg       0.98      0.98      0.98    447074

logistic_regression/OpenLID-v2/bow_maxabs_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+1041.1416900157928,35.084471980497625,282.708448489507,26.858883835512284,0.09915644566638392,l2,50000,1,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1}",0.9802134882665875,0.9800990677001699,0.9800526645605366,0.9801217401757647,6.758499325136823e-05,5
+939.3910819689432,20.5468658244182,317.05882239341736,18.283980434928182,0.0029341230215132346,l2,50000,5,"{'clf__C': 0.0029341230215132346, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9614627159159564,0.9628094259777792,0.962487886482808,0.9622533427921812,0.0005742618768258299,7
+1015.3143272399902,17.818277752531323,288.7444857756297,29.35005618395361,6.155564318973006,l2,50000,3,"{'clf__C': 6.155564318973006, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9813014757359649,0.9813868293760803,0.98128974141665,0.9813260155095652,4.3267912150893565e-05,3
+1045.292368332545,26.252137003322023,304.16649746894836,20.372170331450015,0.3686905640846153,l2,50000,3,"{'clf__C': 0.3686905640846153, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.980860187060233,0.9810350589942068,0.9808662250888074,0.9809204903810823,8.104973692001813e-05,4
+1026.5849002997081,50.62719086675843,274.11564381917316,40.45962602434597,0.00029152036385288323,l2,50000,5,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9261539554418532,0.9223667914857019,0.9276794328098825,0.9254000599124792,0.0022334290992038095,8
+1096.3341462612152,28.168015469641972,286.9407886664073,22.380286434472303,0.0467378670778782,l2,50000,5,"{'clf__C': 0.0467378670778782, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9787622075249414,0.9786802276214641,0.978916716747495,0.9787863839646335,9.804812376194722e-05,6
+1029.094319343567,18.805606989775928,317.5117747783661,36.04870680504547,46.19347374377369,l2,50000,2,"{'clf__C': 46.19347374377369, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9813171969002967,0.9814151465206824,0.9813114953576457,0.9813479462595417,4.757473573366747e-05,2
+1000.5285371939341,77.87680351101064,364.13514431317645,22.861414563982052,0.00028267840503514715,l2,50000,5,"{'clf__C': 0.00028267840503514715, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9250467702719952,0.9214567568400774,0.9270901041175275,0.9245312104098667,0.0023285191096011877,9
+1054.4045932292938,35.90226742096984,327.0744074185689,28.62473167816423,456.60548734461145,l2,50000,2,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9813264357061897,0.981413553626723,0.9813146467257367,0.981351545352883,4.410982153278905e-05,1
+1146.8312856356304,55.61288203329003,357.35789163907367,44.32726586582983,0.00010144487859320231,l2,50000,5,"{'clf__C': 0.00010144487859320231, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.8890696284528324,0.8926661728325486,0.8935091752315354,0.8917483255056388,0.001925136650857023,10

logistic_regression/OpenLID-v2/bow_maxabs_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,25 @@
+{
+  "run_name": "bow_maxabs_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "bow",
+    "scaler": "MaxAbsScaler",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "OpenLID-v2",
+    "vectorizer_class": "CountVectorizer"
+  },
+  "params": {
+    "clf__C": "456.60548734461145",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "2"
+  },
+  "metrics": {
+    "accuracy": 0.9814102363367139,
+    "f1_macro": 0.9816951437188624,
+    "precision_macro": 0.983095730666879,
+    "recall_macro": 0.9816585186171294,
+    "inference_time_total_s": 257.3287,
+    "inference_time_ms_per_sample": 0.575584,
+    "throughput_samples_per_sec": 1737.37
+  }
+}

logistic_regression/OpenLID-v2/tfidf_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       1.00      0.99      1.00      7223
+          da       0.98      0.98      0.98     20007
+          de       1.00      1.00      1.00     19996
+          el       1.00      1.00      1.00     20025
+          en       0.99      1.00      0.99     19981
+          es       1.00      1.00      1.00     20062
+          fi       0.99      0.99      0.99     19949
+          fr       1.00      1.00      1.00     20039
+          hr       0.98      0.98      0.98     19986
+          it       1.00      1.00      1.00     19953
+          ja       0.84      0.98      0.91     19983
+          ko       0.98      0.82      0.89     19990
+          lt       0.99      0.99      0.99     19939
+          mk       1.00      1.00      1.00     20014
+          nb       0.99      0.99      0.99     19962
+          nl       0.99      0.99      0.99     19987
+          pl       0.99      0.99      0.99     20033
+          pt       1.00      1.00      1.00     20037
+          ro       1.00      1.00      1.00     19961
+          ru       1.00      1.00      1.00     20029
+          sl       0.97      0.98      0.97     19959
+          sv       0.97      0.99      0.98     19922
+          uk       1.00      1.00      1.00     20037
+          zh       0.00      0.00      0.00         0
+
+    accuracy                           0.98    447074
+   macro avg       0.94      0.94      0.94    447074
+weighted avg       0.98      0.98      0.98    447074

logistic_regression/OpenLID-v2/tfidf_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,param_vectorizer__sublinear_tf,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+1177.1960979302723,64.96314842995189,407.4506374200185,22.257658847090692,0.09915644566638392,l2,50000,1,True,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9767745965576218,0.9780597223864775,0.9776228208283391,0.9774857132574795,0.000533532856525455,5
+1115.4476176897685,22.72887433843238,503.8296180566152,3.286561121103191,71.77141927992008,l2,50000,1,True,"{'clf__C': 71.77141927992008, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9829905714200443,0.9829022509948367,0.9827327668540775,0.9828751964229863,0.00010697277208662466,2
+1405.8827237288158,93.36829860279127,407.3532235622406,35.56408136522917,0.0017707168643537816,l2,50000,3,True,"{'clf__C': 0.0017707168643537816, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3, 'vectorizer__sublinear_tf': True}",0.9119458869118672,0.9123948637816973,0.9126505763698679,0.9123304423544775,0.00029127237913234214,8
+1566.9878056049347,143.2470977187642,405.675767103831,37.325007308101526,0.00029152036385288323,l2,50000,5,True,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9042959539753,0.9043219443164305,0.8945926448030828,0.9010701810316045,0.0045803220824389085,10
+1225.2683494091034,71.00485253663666,404.9212706883748,34.14379754490904,6.440507553993708,l2,50000,5,True,"{'clf__C': 6.440507553993708, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9835441321616867,0.9832011618100668,0.9834958843964662,0.9834137261227399,0.00015159079123049942,1
+1583.2893005212147,169.70222837266766,306.29970502853394,33.645833048831065,0.00014610865886287189,l2,50000,2,False,"{'clf__C': 0.00014610865886287189, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9018519904046731,0.9015713091002491,0.9025068921663808,0.901976730557101,0.00039200254899588734,9
+1040.0020178953807,22.92054196428311,307.93366201718646,17.313707081885063,456.60548734461145,l2,50000,2,False,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9827716691434817,0.982855002014398,0.9826782338723861,0.9827683016767553,7.220456532983022e-05,3
+1260.0860554377239,98.08504339455561,311.4735604921977,44.80802938613155,0.0028483918709108036,l2,50000,1,True,"{'clf__C': 0.0028483918709108036, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9335510859068542,0.9336379793862851,0.9334658263678486,0.9335516305536626,7.028223065760043e-05,7
+1072.8258341948192,22.952734504253225,300.4805498123169,13.750253295733005,0.027160511446548484,l2,50000,2,True,"{'clf__C': 0.027160511446548484, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': True}",0.9712729908696486,0.9736375441788574,0.9729763227553202,0.9726289526012755,0.0009960848712043994,6
+949.5436635812124,42.612229722788996,303.22136847178143,25.983376233642552,0.28546978577971793,l2,50000,1,True,"{'clf__C': 0.28546978577971793, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.980137586107902,0.9806820754005156,0.9804126248521482,0.9804107621201886,0.00022229072520831727,4

logistic_regression/OpenLID-v2/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "OpenLID-v2",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "6.440507553993708",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.9835284539024859,
+    "f1_macro": 0.9837905549610888,
+    "precision_macro": 0.9849400616925207,
+    "recall_macro": 0.9837720372524135,
+    "inference_time_total_s": 240.6572,
+    "inference_time_ms_per_sample": 0.538294,
+    "throughput_samples_per_sec": 1857.72
+  }
+}

logistic_regression/flores_plus/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "flores_plus",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "6.440507553993708",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.984225621414914,
+    "f1_macro": 0.9859709060910451,
+    "precision_macro": 0.9872362994654668,
+    "recall_macro": 0.985541963917231,
+    "inference_time_total_s": 0.6405,
+    "inference_time_ms_per_sample": 0.30617,
+    "throughput_samples_per_sec": 3266.16
+  }
+}

logistic_regression/language-identification/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       1.00      1.00      1.00       433
+          el       1.00      0.99      1.00       418
+          en       1.00      1.00      1.00       441
+          es       1.00      1.00      1.00       436
+          fi       0.00      0.00      0.00         0
+          fr       1.00      1.00      1.00       417
+          hr       0.00      0.00      0.00         0
+          it       0.99      1.00      0.99       428
+          ja       1.00      0.87      0.93       449
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       0.99      0.99      0.99       399
+          pl       1.00      0.98      0.99       392
+          pt       1.00      1.00      1.00       423
+          ro       0.00      0.00      0.00         0
+          ru       1.00      1.00      1.00       425
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       0.00      0.00      0.00         0
+          zh       0.86      1.00      0.93       452
+
+    accuracy                           0.98      5113
+   macro avg       0.49      0.49      0.49      5113
+weighted avg       0.99      0.98      0.98      5113

logistic_regression/language-identification/bow_maxabs_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+19.384628613789875,3.8331015399707047,4.308941682179769,1.712907469276069,0.09915644566638392,l2,50000,1,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1}",0.9809062399697316,0.9777824340426097,0.9812220853968348,0.979970253136392,0.0015523860932543819,5
+21.579487403233845,3.835541701693018,5.787985722223918,0.7496542187761066,0.0029341230215132346,l2,50000,5,"{'clf__C': 0.0029341230215132346, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9663949590024239,0.9623920442901414,0.9653338066779228,0.9647069366568294,0.001693232818400207,7
+25.824681997299194,8.342065368857105,3.821570078531901,0.45336299425181253,6.155564318973006,l2,50000,3,"{'clf__C': 6.155564318973006, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9850077313197844,0.9820763019847915,0.9852936339773072,0.9841258890939609,0.0014539694107167755,3
+24.18393127123515,2.277633235290609,3.90667716662089,1.8438042331418356,0.3686905640846153,l2,50000,3,"{'clf__C': 0.3686905640846153, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9841817675979536,0.9804763945916357,0.9843422902946145,0.9830001508280679,0.0017857680027994977,4
+27.190136273701984,5.4925571086319955,6.402633984883626,2.5053355911824977,0.00029152036385288323,l2,50000,5,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9577989225040214,0.9545303188343118,0.9571315496631287,0.9564869303338206,0.0014101047354004037,8
+20.16955812772115,8.033087065958464,4.754613558451335,2.2203947955627217,0.0467378670778782,l2,50000,5,"{'clf__C': 0.0467378670778782, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9781771047408362,0.9759891429181439,0.978336182401398,0.977500810020126,0.001070881100133207,6
+21.257368246714275,3.1234232990216553,4.408258597056071,1.0393339616415767,46.19347374377369,l2,50000,2,"{'clf__C': 46.19347374377369, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9857186384335314,0.9823439632088586,0.9850189556796329,0.9843605191073409,0.0014542494645701043,1
+20.534382104873657,5.828078239173822,3.253084103266398,1.6859592587246428,0.00028267840503514715,l2,50000,5,"{'clf__C': 0.00028267840503514715, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9559121018642457,0.9532025736570295,0.9545571144729276,0.9545572633314009,0.0011061602635676635,9
+18.108100096384685,4.147995813634339,5.95265777905782,1.1191833050241058,456.60548734461145,l2,50000,2,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9857186384335314,0.9823439632088586,0.9850189556796329,0.9843605191073409,0.0014542494645701043,1
+22.98309310277303,2.101787814648165,2.8774582544962564,1.2359151172601692,0.00010144487859320231,l2,50000,5,"{'clf__C': 0.00010144487859320231, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.8442227223985483,0.8457379621801148,0.8496728941564529,0.8465445262450387,0.0022969548306086946,10

logistic_regression/language-identification/bow_maxabs_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,25 @@
+{
+  "run_name": "bow_maxabs_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "bow",
+    "scaler": "MaxAbsScaler",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "language-identification",
+    "vectorizer_class": "CountVectorizer"
+  },
+  "params": {
+    "clf__C": "46.19347374377369",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "2"
+  },
+  "metrics": {
+    "accuracy": 0.9837668687658908,
+    "f1_macro": 0.9844389961384462,
+    "precision_macro": 0.9863210921743177,
+    "recall_macro": 0.9841261942624783,
+    "inference_time_total_s": 3.4034,
+    "inference_time_ms_per_sample": 0.665643,
+    "throughput_samples_per_sec": 1502.31
+  }
+}

logistic_regression/language-identification/tfidf_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       1.00      1.00      1.00       433
+          el       1.00      1.00      1.00       418
+          en       0.99      1.00      1.00       441
+          es       1.00      1.00      1.00       436
+          fi       0.00      0.00      0.00         0
+          fr       1.00      1.00      1.00       417
+          hr       0.00      0.00      0.00         0
+          it       1.00      1.00      1.00       428
+          ja       1.00      0.96      0.98       449
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      1.00      1.00       399
+          pl       1.00      0.98      0.99       392
+          pt       0.99      1.00      0.99       423
+          ro       0.00      0.00      0.00         0
+          ru       1.00      1.00      1.00       425
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       0.00      0.00      0.00         0
+          zh       0.95      1.00      0.97       452
+
+    accuracy                           0.99      5113
+   macro avg       0.50      0.50      0.50      5113
+weighted avg       0.99      0.99      0.99      5113

logistic_regression/language-identification/tfidf_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,param_vectorizer__sublinear_tf,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+24.066272258758545,3.0391953849440956,3.621941884358724,2.263294500771926,0.09915644566638392,l2,50000,1,True,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.99336354741965,0.9924559086646587,0.9946730722573522,0.9934975094472204,0.0009100963201092047,5
+25.41511861483256,2.690370502531163,8.407360871632894,1.3489772727463232,71.77141927992008,l2,50000,1,True,"{'clf__C': 71.77141927992008, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9942439327297804,0.9937703879265899,0.9927321751759624,0.9935821652774441,0.0006313601981014462,3
+26.703498204549152,7.959151535802446,4.880916992823283,3.5688900413443685,0.0017707168643537816,l2,50000,3,True,"{'clf__C': 0.0017707168643537816, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3, 'vectorizer__sublinear_tf': True}",0.6538676278213051,0.4378628766170832,0.6546825298353354,0.5821376780912412,0.10201823291820244,8
+28.760569095611572,12.150343299108188,5.237223068873088,4.139852282823081,0.00029152036385288323,l2,50000,5,True,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.08352215580360571,0.07613279856629844,0.08020485967731297,0.07995327134907237,0.0030219334440405874,9
+22.985459327697754,2.718322957008903,5.419108311335246,3.4793378202761756,6.440507553993708,l2,50000,5,True,"{'clf__C': 6.440507553993708, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9950884907577199,0.9938057481504633,0.9943166404642939,0.9944036264574924,0.0005272773282801592,1
+31.711488644282024,3.727225812530605,5.710396846135457,1.9609189795634325,0.00014610865886287189,l2,50000,2,False,"{'clf__C': 0.00014610865886287189, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.012985256323549304,0.012881248872044759,0.05075443007662835,0.02554031175740747,0.01782912460646427,10
+21.026910384496052,8.837167900408234,3.331766049067179,1.3283306292448738,456.60548734461145,l2,50000,2,False,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9941113746512653,0.9941844849320045,0.9924423515739108,0.9935794037190603,0.0008045710923131994,4
+28.85173026720683,1.2566208938046495,4.797921975453694,2.233529341191331,0.0028483918709108036,l2,50000,1,True,"{'clf__C': 0.0028483918709108036, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.7777764179387172,0.6694205069966505,0.780275838418925,0.7424909211180976,0.05167865994517823,7
+31.233232895533245,5.221077561469457,4.673653205235799,1.1545385754722695,0.027160511446548484,l2,50000,2,True,"{'clf__C': 0.027160511446548484, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': True}",0.9898770663311244,0.9892655798114894,0.9917822383250359,0.9903082948225498,0.001071715455804161,6
+24.232994715372723,3.301897471422535,3.1286118030548096,1.2692860024427923,0.28546978577971793,l2,50000,1,True,"{'clf__C': 0.28546978577971793, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9946288669234727,0.993511461693859,0.9944209117261901,0.9941870801145072,0.000485219214210579,2

logistic_regression/language-identification/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "language-identification",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "6.440507553993708",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.99315470369646,
+    "f1_macro": 0.993344498787548,
+    "precision_macro": 0.9936685194721587,
+    "recall_macro": 0.9932338860235449,
+    "inference_time_total_s": 1.7974,
+    "inference_time_ms_per_sample": 0.351533,
+    "throughput_samples_per_sec": 2844.68
+  }
+}

logistic_regression/massive/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       1.00      1.00      1.00       957
+          de       1.00      1.00      1.00      1142
+          el       1.00      1.00      1.00      1141
+          en       0.99      1.00      1.00       963
+          es       1.00      0.99      0.99      1097
+          fi       1.00      1.00      1.00      1044
+          fr       1.00      1.00      1.00      1221
+          hr       0.00      0.00      0.00         0
+          it       1.00      1.00      1.00      1093
+          ja       1.00      0.92      0.96      1337
+          ko       1.00      0.93      0.96      1324
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      1.00      1.00      1129
+          pl       1.00      1.00      1.00      1002
+          pt       0.99      0.99      0.99      1070
+          ro       1.00      1.00      1.00      1010
+          ru       1.00      0.99      1.00      1023
+          sl       1.00      1.00      1.00       998
+          sv       1.00      0.99      1.00       957
+          uk       0.00      0.00      0.00         0
+          zh       0.80      1.00      0.89       874
+
+    accuracy                           0.99     19382
+   macro avg       0.74      0.74      0.74     19382
+weighted avg       0.99      0.99      0.99     19382

logistic_regression/massive/bow_maxabs_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+8.806038777033487,0.1752176687282589,1.8943706353505452,0.03727540730953514,0.09915644566638392,l2,50000,1,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1}",0.9869506520087205,0.9852729163102816,0.9865066022921292,0.9862433902037105,0.0007097697847759615,4
+8.512574275334677,0.5397481553457544,1.9069794019063313,0.05429025459154519,0.0029341230215132346,l2,50000,5,"{'clf__C': 0.0029341230215132346, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9475948789202777,0.9508172187995846,0.9444343815222648,0.947615493080709,0.00260582317574742,7
+8.828056812286377,0.16389161555023937,1.863707462946574,0.04273698349252348,6.155564318973006,l2,50000,3,"{'clf__C': 6.155564318973006, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9871552320784241,0.985144863081794,0.986231310151279,0.9861771351038323,0.0008216232216278756,6
+8.692294359207153,0.16293045207679602,1.9106624126434326,0.016190018826483594,0.3686905640846153,l2,50000,3,"{'clf__C': 0.3686905640846153, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9871550126544126,0.9851893415400785,0.9863800587405667,0.9862414709783526,0.0008084432193248496,5
+9.043086131413778,0.32208519903394495,1.9802601337432861,0.15603986443881065,0.00029152036385288323,l2,50000,5,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.8859559228005294,0.8900094519507787,0.8920350863679657,0.8893334870397579,0.002527416842101805,8
+7.779061317443848,0.12415004559575947,2.1071348190307617,0.032728674331174136,0.0467378670778782,l2,50000,5,"{'clf__C': 0.0467378670778782, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9874498749465922,0.9854895372668911,0.9863503679111729,0.9864299267082188,0.0008022793180492943,1
+8.55941637357076,0.06689128208385134,1.8991725444793701,0.01354718381536146,46.19347374377369,l2,50000,2,"{'clf__C': 46.19347374377369, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.987296221658316,0.9853964094505956,0.9864077689635226,0.9863668000241447,0.0007761359187866395,2
+9.052494287490845,0.3775782794355721,1.8512430985768635,0.014670465614623337,0.00028267840503514715,l2,50000,5,"{'clf__C': 0.00028267840503514715, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.8793390381001349,0.8873376386415608,0.8904146698241728,0.8856971155219563,0.004668039674288462,9
+8.631972710291544,0.20340224630571552,1.8465713659922283,0.036144672920940556,456.60548734461145,l2,50000,2,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.987296221658316,0.9853561509992572,0.9864077689635226,0.9863533805403653,0.0007929636880989923,3
+10.089497725168863,0.26447080603423045,1.4514545599619548,0.06531349435891155,0.00010144487859320231,l2,50000,5,"{'clf__C': 0.00010144487859320231, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.6869670009150837,0.6792802390547998,0.6901147121276733,0.6854539840325189,0.004550704469778963,10

logistic_regression/massive/bow_maxabs_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,25 @@
+{
+  "run_name": "bow_maxabs_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "bow",
+    "scaler": "MaxAbsScaler",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "massive",
+    "vectorizer_class": "CountVectorizer"
+  },
+  "params": {
+    "clf__C": "0.0467378670778782",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5"
+  },
+  "metrics": {
+    "accuracy": 0.9869466515323496,
+    "f1_macro": 0.987175281526507,
+    "precision_macro": 0.987050282253769,
+    "recall_macro": 0.9888379051984971,
+    "inference_time_total_s": 0.9894,
+    "inference_time_ms_per_sample": 0.051048,
+    "throughput_samples_per_sec": 19589.47
+  }
+}

logistic_regression/massive/tfidf_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.99      1.00      1.00       957
+          de       1.00      1.00      1.00      1142
+          el       1.00      1.00      1.00      1141
+          en       1.00      1.00      1.00       963
+          es       1.00      1.00      1.00      1097
+          fi       1.00      1.00      1.00      1044
+          fr       1.00      1.00      1.00      1221
+          hr       0.00      0.00      0.00         0
+          it       1.00      1.00      1.00      1093
+          ja       1.00      0.93      0.96      1337
+          ko       1.00      0.97      0.98      1324
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      1.00      1.00      1129
+          pl       1.00      1.00      1.00      1002
+          pt       0.99      1.00      0.99      1070
+          ro       1.00      1.00      1.00      1010
+          ru       1.00      1.00      1.00      1023
+          sl       1.00      1.00      1.00       998
+          sv       1.00      1.00      1.00       957
+          uk       0.00      0.00      0.00         0
+          zh       0.87      1.00      0.93       874
+
+    accuracy                           0.99     19382
+   macro avg       0.74      0.74      0.74     19382
+weighted avg       0.99      0.99      0.99     19382

logistic_regression/massive/tfidf_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,param_vectorizer__sublinear_tf,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+8.41204277674357,0.03800403238490002,2.130410114924113,0.08880732514408764,0.09915644566638392,l2,50000,1,True,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9729910424797432,0.9733399852969846,0.9738694150122245,0.9734001475963175,0.00036110866614740374,5
+7.746164083480835,0.24946507299764603,2.1380454699198403,0.058062556097476244,71.77141927992008,l2,50000,1,True,"{'clf__C': 71.77141927992008, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.989913260043953,0.9894861630512239,0.9884300658341301,0.989276496309769,0.0006233973440577038,3
+9.926965634028116,0.24857141234102267,1.9099247455596924,0.11720351930324341,0.0017707168643537816,l2,50000,3,True,"{'clf__C': 0.0017707168643537816, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3, 'vectorizer__sublinear_tf': True}",0.28948538807893903,0.2891538687960144,0.2880832375410509,0.2889074981386681,0.0005983479524635098,8
+10.602810462315878,0.472668710982073,1.9435614744822185,0.11176598364492388,0.00029152036385288323,l2,50000,5,True,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.06072823370136157,0.061920327623000766,0.0579689410422329,0.06020583412219841,0.0016548997890910644,9
+9.06920305887858,0.5940014475948462,2.001631021499634,0.13090904898254052,6.440507553993708,l2,50000,5,True,"{'clf__C': 6.440507553993708, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9906162773916051,0.9900120609761767,0.9899847503626515,0.9902043629101444,0.0002914808426300923,1
+11.624321222305298,0.2825827874130309,1.7265090942382812,0.11564576970282385,0.00014610865886287189,l2,50000,2,False,"{'clf__C': 0.00014610865886287189, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.02403926940150334,0.03542528702201226,0.03300912599136062,0.03082456080495874,0.004898271334791776,10
+7.584720770517985,0.19078849380875443,2.027594725290934,0.014661266984491505,456.60548734461145,l2,50000,2,False,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.989827648631286,0.9892548125214762,0.9886183646504715,0.9892336086010779,0.0004939157426416381,4
+9.725672086079916,0.11306594801186873,1.7257123788197835,0.08169984862783343,0.0028483918709108036,l2,50000,1,True,"{'clf__C': 0.0028483918709108036, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.5252265628707862,0.5276941666920951,0.5261065185506385,0.5263424160378399,0.0010211114444844223,7
+8.481626510620117,0.4627061651280554,1.7441190878550212,0.10620556747677802,0.027160511446548484,l2,50000,2,True,"{'clf__C': 0.027160511446548484, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': True}",0.9653768773786644,0.9647228220063488,0.9679398938158121,0.9660131977336084,0.0013883000281874598,6
+8.094027280807495,0.20726157673706772,1.7728882630666096,0.050696983855423185,0.28546978577971793,l2,50000,1,True,"{'clf__C': 0.28546978577971793, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9902870355876624,0.9897382107350143,0.9894092506684539,0.9898114989970436,0.0003620819134863722,2

logistic_regression/massive/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "massive",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "6.440507553993708",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.9916933237024043,
+    "f1_macro": 0.9917252601889304,
+    "precision_macro": 0.9911473317774038,
+    "recall_macro": 0.9929328394176677,
+    "inference_time_total_s": 0.9903,
+    "inference_time_ms_per_sample": 0.051093,
+    "throughput_samples_per_sec": 19572.3
+  }
+}

logistic_regression/mmarco/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       1.00      1.00      1.00     20033
+          el       0.00      0.00      0.00         0
+          en       0.99      1.00      0.99     20044
+          es       0.99      0.99      0.99     20016
+          fi       0.00      0.00      0.00         0
+          fr       1.00      1.00      1.00     19956
+          hr       0.00      0.00      0.00         0
+          it       1.00      1.00      1.00     19992
+          ja       1.00      0.87      0.93     19935
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      1.00      1.00     19942
+          pl       0.00      0.00      0.00         0
+          pt       0.99      0.99      0.99     20036
+          ro       0.00      0.00      0.00         0
+          ru       1.00      1.00      1.00     20086
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       0.00      0.00      0.00         0
+          zh       0.88      1.00      0.94     20005
+
+    accuracy                           0.98    200045
+   macro avg       0.41      0.41      0.41    200045
+weighted avg       0.99      0.98      0.98    200045

logistic_regression/mmarco/bow_maxabs_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+66.5223552385966,3.442939478354831,19.303172906239826,0.6179726425335146,0.09915644566638392,l2,50000,1,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1}",0.9816077153766996,0.9811656247847337,0.9814521436439888,0.9814084946018072,0.00018310279815575867,6
+64.78002214431763,2.3454436585317855,19.040038267771404,1.0511100641461475,0.0029341230215132346,l2,50000,5,"{'clf__C': 0.0029341230215132346, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9667275514753921,0.9667991298401912,0.9668098111214853,0.9667788308123563,3.652122891442162e-05,7
+70.61315234502156,3.9931379788581687,18.763609727223713,0.5701110154356371,6.155564318973006,l2,50000,3,"{'clf__C': 6.155564318973006, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9839623927384455,0.9843367748597099,0.9837529461180112,0.9840173712387221,0.0002414966894567856,3
+69.23730270067851,4.66706506108104,18.664097627003986,0.6371831089310802,0.3686905640846153,l2,50000,3,"{'clf__C': 0.3686905640846153, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9827264987172839,0.9838144848495279,0.9832643913175347,0.983268458294782,0.0004441777881410264,4
+65.5788471698761,1.4394848907851474,17.97521750132243,0.10428279780608445,0.00029152036385288323,l2,50000,5,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9428079552865585,0.9466900015549994,0.9428883510689122,0.94412876930349,0.0018113620759130098,8
+71.44474705060323,3.347553457202775,18.226688226064045,0.4661330272129226,0.0467378670778782,l2,50000,5,"{'clf__C': 0.0467378670778782, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.981402609170039,0.9812462487454214,0.9817275641254124,0.9814588073469576,0.00020047410484598003,5
+65.06078600883484,0.21780367153264815,17.72416130701701,0.16610170198210247,46.19347374377369,l2,50000,2,"{'clf__C': 46.19347374377369, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9839584446752623,0.9843887118316094,0.9837791994795599,0.9840421186621437,0.00025576985603689326,1
+64.34969687461853,1.327372786682874,17.40241042772929,0.18591026765558533,0.00028267840503514715,l2,50000,5,"{'clf__C': 0.00028267840503514715, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9426543210488647,0.9465419867199142,0.9426799998349568,0.9439587692012452,0.0018266407075252437,9
+65.24408038457234,0.8699221487258816,16.950232187906902,0.061915212666849136,456.60548734461145,l2,50000,2,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9839546992390027,0.9843887118316094,0.9837791994795599,0.9840408701833908,0.0002561840491622399,2
+64.38435085614522,0.7162908714508494,16.607149759928387,0.344745535331013,0.00010144487859320231,l2,50000,5,"{'clf__C': 0.00010144487859320231, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.939047143828233,0.9409774619794792,0.9378003040687191,0.9392749699588103,0.0013070352346198934,10

logistic_regression/mmarco/bow_maxabs_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,25 @@
+{
+  "run_name": "bow_maxabs_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "bow",
+    "scaler": "MaxAbsScaler",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "mmarco",
+    "vectorizer_class": "CountVectorizer"
+  },
+  "params": {
+    "clf__C": "46.19347374377369",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "2"
+  },
+  "metrics": {
+    "accuracy": 0.9837136644255042,
+    "f1_macro": 0.9836626999127922,
+    "precision_macro": 0.985201679742007,
+    "recall_macro": 0.9836709364057225,
+    "inference_time_total_s": 10.377,
+    "inference_time_ms_per_sample": 0.051874,
+    "throughput_samples_per_sec": 19277.64
+  }
+}

logistic_regression/mmarco/tfidf_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       1.00      1.00      1.00     20033
+          el       0.00      0.00      0.00         0
+          en       0.99      1.00      1.00     20044
+          es       0.99      0.99      0.99     20016
+          fi       0.00      0.00      0.00         0
+          fr       1.00      1.00      1.00     19956
+          hr       0.00      0.00      0.00         0
+          it       1.00      1.00      1.00     19992
+          ja       1.00      0.90      0.95     19935
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      1.00      1.00     19942
+          pl       0.00      0.00      0.00         0
+          pt       0.99      1.00      0.99     20036
+          ro       0.00      0.00      0.00         0
+          ru       1.00      1.00      1.00     20086
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       0.00      0.00      0.00         0
+          zh       0.91      0.99      0.95     20005
+
+    accuracy                           0.99    200045
+   macro avg       0.41      0.41      0.41    200045
+weighted avg       0.99      0.99      0.99    200045

logistic_regression/mmarco/tfidf_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,param_vectorizer__sublinear_tf,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+59.6538671652476,2.074384059881507,20.118187030156452,1.2479418357581145,0.09915644566638392,l2,50000,1,True,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9833258053335129,0.9838468317706035,0.9833805034256725,0.9835177135099297,0.00023379063632977408,5
+60.61395804087321,2.305601307861205,19.585657835006714,1.1408303303992218,71.77141927992008,l2,50000,1,True,"{'clf__C': 71.77141927992008, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9855790450750206,0.9865762503924931,0.9864237736487402,0.9861930230387513,0.0004385878812025037,2
+70.79162931442261,2.646390432395221,18.570846160252888,0.5584507400917724,0.0017707168643537816,l2,50000,3,True,"{'clf__C': 0.0017707168643537816, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3, 'vectorizer__sublinear_tf': True}",0.9619645279896675,0.9746191346656217,0.9676746803269509,0.9680861143274133,0.005174406631574839,8
+75.09138417243958,5.40426609528691,19.9089670976003,1.7869296876622531,0.00029152036385288323,l2,50000,5,True,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.928443540007254,0.9739812908058056,0.9710815108508912,0.957835447221317,0.02081690560108609,10
+71.04249183336894,5.58215604555665,18.799468994140625,0.7432630586734928,6.440507553993708,l2,50000,5,True,"{'clf__C': 6.440507553993708, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9861762802254667,0.9868189262341623,0.9873751997927988,0.986790135417476,0.0004898800639456188,1
+76.76116863886516,4.259923992078333,17.62615132331848,0.6644368377752696,0.00014610865886287189,l2,50000,2,False,"{'clf__C': 0.00014610865886287189, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9645103144613729,0.9415430800750798,0.9695522231577154,0.958535205898056,0.01219028208077935,9
+60.52088395754496,2.249377028691026,18.770265181859333,0.4240722708301269,456.60548734461145,l2,50000,2,False,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9854502990107626,0.9864001389352858,0.9862143327972349,0.986021590247761,0.0004110241179766901,3
+65.97150150934856,1.731910808777453,16.917045036951702,0.44875397436509445,0.0028483918709108036,l2,50000,1,True,"{'clf__C': 0.0028483918709108036, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9659258485440183,0.9734511205583004,0.9749578985871136,0.9714449558964774,0.003950780891841069,7
+61.16194780667623,1.9453262427812146,16.694405635197956,0.7564741669941186,0.027160511446548484,l2,50000,2,True,"{'clf__C': 0.027160511446548484, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': True}",0.9799072878065406,0.9807516459238188,0.9821819197889194,0.9809469511730929,0.000938827555734894,6
+59.47458577156067,0.5901416153251525,16.188791116078693,0.24599274884352676,0.28546978577971793,l2,50000,1,True,"{'clf__C': 0.28546978577971793, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9848508053302464,0.9855014124805159,0.9857010838976163,0.9853511005694596,0.0003630322882667097,4

logistic_regression/mmarco/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "mmarco",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "6.440507553993708",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.9872028793521458,
+    "f1_macro": 0.9871717798405111,
+    "precision_macro": 0.9879831577726538,
+    "recall_macro": 0.9871698822374911,
+    "inference_time_total_s": 10.3763,
+    "inference_time_ms_per_sample": 0.05187,
+    "throughput_samples_per_sec": 19278.98
+  }
+}

logistic_regression/multi_eurlex/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       1.00      1.00      1.00      6478
+          de       1.00      1.00      1.00      6514
+          el       1.00      1.00      1.00      6499
+          en       1.00      1.00      1.00      6537
+          es       1.00      1.00      1.00      6263
+          fi       1.00      1.00      1.00      5230
+          fr       1.00      1.00      1.00      6508
+          hr       1.00      1.00      1.00      1557
+          it       1.00      1.00      1.00      6499
+          ja       0.00      0.00      0.00         0
+          ko       0.00      0.00      0.00         0
+          lt       1.00      1.00      1.00      3330
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      1.00      1.00      6529
+          pl       1.00      1.00      1.00      3317
+          pt       1.00      1.00      1.00      6233
+          ro       1.00      1.00      1.00      2600
+          ru       0.00      0.00      0.00         0
+          sl       1.00      1.00      1.00      3313
+          sv       1.00      1.00      1.00      5257
+          uk       0.00      0.00      0.00         0
+          zh       0.00      0.00      0.00         0
+
+    accuracy                           1.00     82664
+   macro avg       0.67      0.67      0.67     82664
+weighted avg       1.00      1.00      1.00     82664

logistic_regression/multi_eurlex/bow_maxabs_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+58.11060563723246,0.379962077274111,21.211435317993164,0.9448053631930966,0.09915644566638392,l2,50000,1,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1}",0.9998117390231491,0.9998332454583383,0.9998188856607799,0.9998212900474224,8.943060437438874e-06,2
+59.3550066947937,0.13962543586908094,20.462625741958618,0.9450047828624563,0.0029341230215132346,l2,50000,5,"{'clf__C': 0.0029341230215132346, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9997595056959798,0.9997149870024714,0.9997521675168728,0.9997422200717746,1.94883258372628e-05,7
+59.568419535954796,0.664639820264726,20.143352508544922,0.8190291119801871,6.155564318973006,l2,50000,3,"{'clf__C': 6.155564318973006, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9998043903179885,0.9998332652846516,0.9998188856607799,0.9998188470878068,1.1788187331777226e-05,3
+59.52647336324056,0.8705879774657466,20.153676827748615,0.9539645421872754,0.3686905640846153,l2,50000,3,"{'clf__C': 0.3686905640846153, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9998043903179885,0.9998332652846516,0.9998188856607799,0.9998188470878068,1.1788187331777226e-05,3
+58.92317684491476,0.5066054578998324,20.894578059514362,0.9530331901121322,0.00029152036385288323,l2,50000,5,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9995934475075559,0.9993940553133949,0.9995556513181287,0.9995143847130264,8.647354622689244e-05,8
+68.48207505544026,3.2884036089716475,19.520317633946735,1.0031714327124717,0.0467378670778782,l2,50000,5,"{'clf__C': 0.0467378670778782, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9998264176705822,0.9998332437631596,0.9998335879365807,0.9998310831234409,3.301964225332328e-06,1
+58.04341173171997,0.6357297993833433,19.429731369018555,0.04803814652826426,46.19347374377369,l2,50000,2,"{'clf__C': 46.19347374377369, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9998043903179885,0.9998332652846516,0.9998188856607799,0.9998188470878068,1.1788187331777226e-05,3
+57.63342340787252,0.13569377155713697,19.285069227218628,0.1991914375848924,0.00028267840503514715,l2,50000,5,"{'clf__C': 0.00028267840503514715, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9995934475075559,0.9993940553133949,0.9995556513181287,0.9995143847130264,8.647354622689244e-05,8
+57.47682205835978,0.3791579637148409,19.386224110921223,0.13647914233662936,456.60548734461145,l2,50000,2,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9998043903179885,0.9998332652846516,0.9998188856607799,0.9998188470878068,1.1788187331777226e-05,3
+59.77876329421997,1.6395823468633917,19.146162192026775,0.6610139274757174,0.00010144487859320231,l2,50000,5,"{'clf__C': 0.00010144487859320231, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9991479032074926,0.9993074342715771,0.999465369882376,0.9993069024538151,0.00012960577286006444,10

logistic_regression/multi_eurlex/bow_maxabs_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,25 @@
+{
+  "run_name": "bow_maxabs_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "bow",
+    "scaler": "MaxAbsScaler",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "multi_eurlex",
+    "vectorizer_class": "CountVectorizer"
+  },
+  "params": {
+    "clf__C": "0.0467378670778782",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5"
+  },
+  "metrics": {
+    "accuracy": 0.9997822510403561,
+    "f1_macro": 0.9998217192528284,
+    "precision_macro": 0.9998226084060367,
+    "recall_macro": 0.9998208431704974,
+    "inference_time_total_s": 11.5672,
+    "inference_time_ms_per_sample": 0.13993,
+    "throughput_samples_per_sec": 7146.44
+  }
+}

logistic_regression/multi_eurlex/tfidf_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       1.00      1.00      1.00      6478
+          de       1.00      1.00      1.00      6514
+          el       1.00      1.00      1.00      6499
+          en       1.00      1.00      1.00      6537
+          es       1.00      1.00      1.00      6263
+          fi       1.00      1.00      1.00      5230
+          fr       1.00      1.00      1.00      6508
+          hr       1.00      1.00      1.00      1557
+          it       1.00      1.00      1.00      6499
+          ja       0.00      0.00      0.00         0
+          ko       0.00      0.00      0.00         0
+          lt       1.00      1.00      1.00      3330
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      1.00      1.00      6529
+          pl       1.00      1.00      1.00      3317
+          pt       1.00      1.00      1.00      6233
+          ro       1.00      1.00      1.00      2600
+          ru       0.00      0.00      0.00         0
+          sl       1.00      1.00      1.00      3313
+          sv       1.00      1.00      1.00      5257
+          uk       0.00      0.00      0.00         0
+          zh       0.00      0.00      0.00         0
+
+    accuracy                           1.00     82664
+   macro avg       0.67      0.67      0.67     82664
+weighted avg       1.00      1.00      1.00     82664

logistic_regression/multi_eurlex/tfidf_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,param_vectorizer__sublinear_tf,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+51.2232453028361,1.3142174686092774,21.26999632517497,1.6978969320646964,0.09915644566638392,l2,50000,1,True,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9996923440006453,0.999572573925271,0.999775889330342,0.9996802690854194,8.344116157741288e-05,5
+50.95029719670614,1.5331219888959293,19.845767895380657,0.22058695426570238,71.77141927992008,l2,50000,1,True,"{'clf__C': 71.77141927992008, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.999848279199895,0.9998155603678858,0.9998449055564081,0.9998362483747297,1.469332247368876e-05,1
+56.47531700134277,2.0276314353741824,20.515960216522217,1.5188026332487372,0.0017707168643537816,l2,50000,3,True,"{'clf__C': 0.0017707168643537816, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3, 'vectorizer__sublinear_tf': True}",0.9866035290502679,0.9877420796794085,0.9874921338758486,0.987279247535175,0.0004885794574158426,8
+73.0179849465688,1.3606154757052669,20.774610837300617,0.29146226969339684,0.00029152036385288323,l2,50000,5,True,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.6496726295709538,0.6463484848305601,0.6449876533201166,0.6470029225738768,0.001967818959768285,9
+54.08846044540405,3.811135046304561,21.75122380256653,1.5871673584552195,6.440507553993708,l2,50000,5,True,"{'clf__C': 6.440507553993708, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9998411050554616,0.9998085160514245,0.9998449055564081,0.9998315088877647,1.6332255285955568e-05,2
+67.77160255114238,3.4047254460362333,18.902366558710735,1.2611757935868972,0.00014610865886287189,l2,50000,2,False,"{'clf__C': 0.00014610865886287189, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.5994429376198614,0.5972865826053974,0.6216152761654814,0.6061149321302467,0.010995695086262238,10
+51.30253863334656,2.7206037266488687,20.6509002049764,0.46571918980480437,456.60548734461145,l2,50000,2,False,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9998556269244205,0.9998010153763707,0.999830207929822,0.999828950076871,2.2312805600302612e-05,3
+51.64417084058126,2.1085992095917288,20.048081477483112,0.6197062548115148,0.0028483918709108036,l2,50000,1,True,"{'clf__C': 0.0028483918709108036, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9887654249446605,0.991687090672797,0.9900873697759416,0.9901799617977997,0.0011945606217701348,7
+49.21649479866028,0.5427624612301145,19.9834201335907,0.4216165302661986,0.027160511446548484,l2,50000,2,True,"{'clf__C': 0.027160511446548484, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': True}",0.9995719670993201,0.9994570946027668,0.9995363490447238,0.9995218035822702,4.801111312038349e-05,6
+48.79159132639567,0.6494571625348741,18.51280887921651,0.6551590521276021,0.28546978577971793,l2,50000,1,True,"{'clf__C': 0.28546978577971793, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9997864891955355,0.9997170527525259,0.9997752511743214,0.9997595977074609,3.0431652085957768e-05,4

logistic_regression/multi_eurlex/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "multi_eurlex",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "71.77141927992008",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "1",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.9997943482047807,
+    "f1_macro": 0.9998310955808424,
+    "precision_macro": 0.9998300253353787,
+    "recall_macro": 0.9998322719298366,
+    "inference_time_total_s": 11.4756,
+    "inference_time_ms_per_sample": 0.138822,
+    "throughput_samples_per_sec": 7203.47
+  }
+}

logistic_regression/multilingual_cc_news/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       1.00      1.00      1.00     20015
+          da       0.99      0.99      0.99     20013
+          de       1.00      1.00      1.00     20008
+          el       1.00      1.00      1.00     20050
+          en       0.98      1.00      0.99     20005
+          es       0.00      0.00      0.00         0
+          fi       1.00      1.00      1.00     20084
+          fr       0.00      0.00      0.00         0
+          hr       1.00      1.00      1.00     19914
+          it       0.00      0.00      0.00         0
+          ja       0.71      0.98      0.82     20022
+          ko       0.99      0.83      0.90     20000
+          lt       1.00      0.99      1.00      4950
+          mk       1.00      1.00      1.00     18680
+          nb       0.99      0.99      0.99     19959
+          nl       1.00      1.00      1.00     20009
+          pl       1.00      1.00      1.00     20033
+          pt       1.00      1.00      1.00     19949
+          ro       1.00      1.00      1.00     20113
+          ru       1.00      1.00      1.00     20065
+          sl       1.00      1.00      1.00     19951
+          sv       1.00      1.00      1.00     20044
+          uk       1.00      0.99      1.00     20033
+          zh       0.97      0.76      0.85     19979
+
+    accuracy                           0.98    403876
+   macro avg       0.86      0.85      0.85    403876
+weighted avg       0.98      0.98      0.98    403876

logistic_regression/multilingual_cc_news/bow_maxabs_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+402.9399371147156,8.154657269017594,95.20252974828084,2.0797345240090186,0.09915644566638392,l2,50000,1,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1}",0.9752277520546296,0.974338503210626,0.9750254403671778,0.9748638985441445,0.00038058086927060075,5
+380.97084434827167,2.096794290577095,90.9351966381073,0.8475863264935782,0.0029341230215132346,l2,50000,5,"{'clf__C': 0.0029341230215132346, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9656756294108784,0.9646967887735727,0.9656415967419004,0.9653380049754504,0.00045362114795086784,7
+403.4427442550659,11.427001607226778,93.5841719309489,2.2056415233593736,6.155564318973006,l2,50000,3,"{'clf__C': 6.155564318973006, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9764374935614732,0.9756131467393677,0.9761156109011319,0.9760554170673243,0.0003392190973399652,3
+402.5956726074219,13.617858289354421,97.36782177289327,5.511217277327732,0.3686905640846153,l2,50000,3,"{'clf__C': 0.3686905640846153, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9759331912035696,0.9753399557527218,0.9757030313980376,0.9756587261181097,0.000244205232550765,4
+393.6945722103119,24.312571724344345,95.80627179145813,2.3690566482175197,0.00029152036385288323,l2,50000,5,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9522448367266754,0.951290233113672,0.9525833370485026,0.95203946896295,0.0005475164444116591,8
+387.80599387486774,22.64524855130065,93.43839979171753,3.9156075612376933,0.0467378670778782,l2,50000,5,"{'clf__C': 0.0467378670778782, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9744132497491936,0.9738388423746759,0.974262851645611,0.9741716479231601,0.00024320710956817393,6
+386.3445611000061,7.549293275136315,90.40917595227559,3.72332399859066,46.19347374377369,l2,50000,2,"{'clf__C': 46.19347374377369, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9764455705928385,0.9756172821037771,0.9761210137533212,0.9760612888166457,0.00034077436946226695,2
+361.68538037935895,13.157352840529368,91.73455119132996,2.695762613364933,0.00028267840503514715,l2,50000,5,"{'clf__C': 0.00028267840503514715, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9519740282093876,0.9510311614518805,0.952325945638046,0.9517770450997713,0.0005466371625930101,9
+394.5825621287028,4.753316303726586,88.7125723361969,3.3864155009464953,456.60548734461145,l2,50000,2,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9764473401367928,0.9756170092230521,0.9761480404733097,0.9760707966110514,0.00034335338187820946,1
+387.46650465329486,6.832365252854412,87.48951387405396,3.2864414083328612,0.00010144487859320231,l2,50000,5,"{'clf__C': 0.00010144487859320231, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.8945935376189939,0.905198488874407,0.9093624319141937,0.9030514861358648,0.006217570546184826,10

logistic_regression/multilingual_cc_news/bow_maxabs_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,25 @@
+{
+  "run_name": "bow_maxabs_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "bow",
+    "scaler": "MaxAbsScaler",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "multilingual_cc_news",
+    "vectorizer_class": "CountVectorizer"
+  },
+  "params": {
+    "clf__C": "456.60548734461145",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "2"
+  },
+  "metrics": {
+    "accuracy": 0.9757425546454852,
+    "f1_macro": 0.9769901432987629,
+    "precision_macro": 0.9815579981555338,
+    "recall_macro": 0.9763925161914471,
+    "inference_time_total_s": 54.8261,
+    "inference_time_ms_per_sample": 0.13575,
+    "throughput_samples_per_sec": 7366.5
+  }
+}

logistic_regression/multilingual_cc_news/tfidf_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       1.00      1.00      1.00     20015
+          da       0.99      0.99      0.99     20013
+          de       1.00      1.00      1.00     20008
+          el       1.00      1.00      1.00     20050
+          en       0.98      1.00      0.99     20005
+          es       0.00      0.00      0.00         0
+          fi       1.00      1.00      1.00     20084
+          fr       0.00      0.00      0.00         0
+          hr       1.00      1.00      1.00     19914
+          it       0.00      0.00      0.00         0
+          ja       0.76      0.96      0.85     20022
+          ko       0.98      0.87      0.92     20000
+          lt       1.00      0.99      1.00      4950
+          mk       1.00      1.00      1.00     18680
+          nb       0.99      0.99      0.99     19959
+          nl       1.00      1.00      1.00     20009
+          pl       1.00      1.00      1.00     20033
+          pt       1.00      1.00      1.00     19949
+          ro       1.00      1.00      1.00     20113
+          ru       1.00      1.00      1.00     20065
+          sl       1.00      1.00      1.00     19951
+          sv       1.00      1.00      1.00     20044
+          uk       1.00      0.99      1.00     20033
+          zh       0.95      0.81      0.87     19979
+
+    accuracy                           0.98    403876
+   macro avg       0.86      0.86      0.86    403876
+weighted avg       0.98      0.98      0.98    403876

logistic_regression/multilingual_cc_news/tfidf_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,param_vectorizer__sublinear_tf,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+387.14908870061237,10.42996720885376,103.22419436772664,4.436658127357272,0.09915644566638392,l2,50000,1,True,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9768182861408593,0.9767568931960271,0.9765458863566485,0.976707021897845,0.0001166641058660818,5
+391.8255871136983,15.368568481786543,100.51205062866211,3.220759423950863,71.77141927992008,l2,50000,1,True,"{'clf__C': 71.77141927992008, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9797900944731284,0.9794235244038101,0.9794082901409027,0.9795406363392803,0.00017650314652298794,2
+446.27149240175885,29.054544895738466,95.0022524992625,2.608151464896373,0.0017707168643537816,l2,50000,3,True,"{'clf__C': 0.0017707168643537816, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3, 'vectorizer__sublinear_tf': True}",0.9190900623481623,0.9173214284053496,0.9409196894876505,0.9257770600803874,0.01073177324677203,8
+452.4179784456889,29.159745115837524,93.67036859194438,2.7291077562187,0.00029152036385288323,l2,50000,5,True,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9120915218384804,0.9121332298416716,0.8050393995954123,0.8764213837585214,0.05047468792828181,10
+424.96267716089886,7.284722268470122,98.61706328392029,4.555820736147861,6.440507553993708,l2,50000,5,True,"{'clf__C': 6.440507553993708, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9800145518273617,0.9797307989053394,0.9797799799488286,0.9798417768938433,0.00012380919997832622,1
+463.7910005251567,26.796591656182194,89.84521587689717,4.76783496832902,0.00014610865886287189,l2,50000,2,False,"{'clf__C': 0.00014610865886287189, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9106661762467955,0.9106675015295603,0.9107332030472706,0.910688960274542,3.128904278526697e-05,9
+373.6284962495168,14.232401492412562,96.45695861180623,5.697114629543932,456.60548734461145,l2,50000,2,False,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9796566744404435,0.9792835403796589,0.9792844693383306,0.979408228052811,0.0001756785348048597,3
+393.46992659568787,23.09630129683921,87.79749663670857,1.0754166944404593,0.0028483918709108036,l2,50000,1,True,"{'clf__C': 0.0028483918709108036, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9586044191820471,0.9580164952150932,0.9589165200264395,0.9585124781411932,0.0003731407522679695,7
+380.7316460609436,6.8812991032022515,88.0167322953542,1.2114695598564937,0.027160511446548484,l2,50000,2,True,"{'clf__C': 0.027160511446548484, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': True}",0.9742560600989444,0.9742945797824765,0.9739362286516305,0.9741622895110171,0.00016062082806993106,6
+392.1845534642537,15.298243699958181,84.93731991449992,8.203447659422412,0.28546978577971793,l2,50000,1,True,"{'clf__C': 0.28546978577971793, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9784690777439445,0.9782639037336438,0.9780913210054424,0.9782747674943435,0.00015440974568412197,4

logistic_regression/multilingual_cc_news/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "multilingual_cc_news",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "6.440507553993708",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.9795209420713288,
+    "f1_macro": 0.9804527733547908,
+    "precision_macro": 0.9827989401352565,
+    "recall_macro": 0.9800439981432557,
+    "inference_time_total_s": 55.1468,
+    "inference_time_ms_per_sample": 0.136544,
+    "throughput_samples_per_sec": 7323.66
+  }
+}

logistic_regression/multilingual_toxicity_dataset/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       1.00      1.00      1.00       506
+          el       0.00      0.00      0.00         0
+          en       0.99      1.00      0.99       478
+          es       1.00      0.99      0.99       495
+          fi       0.00      0.00      0.00         0
+          fr       1.00      0.99      0.99       472
+          hr       0.00      0.00      0.00         0
+          it       1.00      1.00      1.00       478
+          ja       1.00      0.79      0.88       493
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       0.00      0.00      0.00         0
+          pl       0.00      0.00      0.00         0
+          pt       0.00      0.00      0.00         0
+          ro       0.00      0.00      0.00         0
+          ru       1.00      0.99      0.99       492
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       0.99      0.99      0.99       455
+          zh       0.79      1.00      0.89       455
+
+    accuracy                           0.97      4324
+   macro avg       0.37      0.36      0.36      4324
+weighted avg       0.98      0.97      0.97      4324

logistic_regression/multilingual_toxicity_dataset/bow_maxabs_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+4.040556033452352,0.2035284762584747,0.6972482204437256,0.02066138439134423,0.09915644566638392,l2,50000,1,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1}",0.9653392188400132,0.963977108186834,0.9597028752035628,0.96300640074347,0.002401221508939484,5
+4.047645092010498,0.08398335122140148,0.7023676236470541,0.014108892106838362,0.0029341230215132346,l2,50000,5,"{'clf__C': 0.0029341230215132346, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.8537109050099432,0.8494223545046622,0.8865347645938498,0.8632226747028184,0.01657685269031758,7
+4.27047594388326,0.045011945531176406,0.7021175225575765,0.021968218840048216,6.155564318973006,l2,50000,3,"{'clf__C': 6.155564318973006, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9682132389520426,0.9675306246864641,0.9632948604605684,0.9663462413663583,0.002175574197471764,1
+4.047166268030803,0.06535130427514227,0.7372770309448242,0.036618304507032066,0.3686905640846153,l2,50000,3,"{'clf__C': 0.3686905640846153, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9674354523313442,0.9664781486266119,0.9619639468590014,0.9652925159389859,0.0023858802269385725,4
+4.356929937998454,0.10613291928686613,0.7481387456258138,0.02214152640693379,0.00029152036385288323,l2,50000,5,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.8166368388753926,0.819233045626842,0.8179227976649744,0.8179308940557363,0.001059912429629702,8
+3.876397450764974,0.13087851195357808,0.726686159769694,0.03464620132678077,0.0467378670778782,l2,50000,5,"{'clf__C': 0.0467378670778782, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9644260968851864,0.9626887984958667,0.9568466707398555,0.9613205220403028,0.003242022055988025,6
+3.8898606300354004,0.03739246875719135,0.6777958075205485,0.0030575897224061487,46.19347374377369,l2,50000,2,"{'clf__C': 46.19347374377369, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9683916700693344,0.9683831894580818,0.9618557690445331,0.966210209523983,0.0030790563377974894,2
+4.0603954792022705,0.06339875502603774,0.6255292892456055,0.004404361708554165,0.00028267840503514715,l2,50000,5,"{'clf__C': 0.00028267840503514715, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.815543995896894,0.81805974836216,0.8154345079182215,0.8163460840590918,0.001212567774621574,9
+3.878629207611084,0.0632108031301941,0.6451913515726725,0.008300134344414923,456.60548734461145,l2,50000,2,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.9683916700693344,0.9683831894580818,0.9618557690445331,0.966210209523983,0.0030790563377974894,2
+4.218158880869548,0.241054313075401,0.5932675202687582,0.02689271267763404,0.00010144487859320231,l2,50000,5,"{'clf__C': 0.00010144487859320231, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.7313444062530139,0.7376569517327378,0.7344403744996182,0.7344805774951233,0.0025772426892619747,10

logistic_regression/multilingual_toxicity_dataset/bow_maxabs_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,25 @@
+{
+  "run_name": "bow_maxabs_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "bow",
+    "scaler": "MaxAbsScaler",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "multilingual_toxicity_dataset",
+    "vectorizer_class": "CountVectorizer"
+  },
+  "params": {
+    "clf__C": "6.155564318973006",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "3"
+  },
+  "metrics": {
+    "accuracy": 0.9699352451433858,
+    "f1_macro": 0.9697068633400848,
+    "precision_macro": 0.9742811759448592,
+    "recall_macro": 0.9705447736079997,
+    "inference_time_total_s": 0.3653,
+    "inference_time_ms_per_sample": 0.084493,
+    "throughput_samples_per_sec": 11835.32
+  }
+}

logistic_regression/multilingual_toxicity_dataset/tfidf_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       1.00      1.00      1.00       506
+          el       0.00      0.00      0.00         0
+          en       1.00      1.00      1.00       478
+          es       1.00      0.99      0.99       495
+          fi       0.00      0.00      0.00         0
+          fr       1.00      0.99      0.99       472
+          hr       0.00      0.00      0.00         0
+          it       0.99      1.00      1.00       478
+          ja       1.00      0.92      0.96       493
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       0.00      0.00      0.00         0
+          pl       0.00      0.00      0.00         0
+          pt       0.00      0.00      0.00         0
+          ro       0.00      0.00      0.00         0
+          ru       0.99      0.99      0.99       492
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       1.00      0.99      0.99       455
+          zh       0.90      1.00      0.95       455
+
+    accuracy                           0.99      4324
+   macro avg       0.37      0.37      0.37      4324
+weighted avg       0.99      0.99      0.99      4324

logistic_regression/multilingual_toxicity_dataset/tfidf_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,param_vectorizer__sublinear_tf,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+3.9588398138682046,0.20348158738211433,0.7723536491394043,0.0025173423011794578,0.09915644566638392,l2,50000,1,True,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9845364014625712,0.9854455679898189,0.985105666152761,0.9850292118683837,0.0003750821125803859,1
+3.9503840605417886,0.024054342345522246,0.7935018539428711,0.01792406073588315,71.77141927992008,l2,50000,1,True,"{'clf__C': 71.77141927992008, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9831628738697789,0.9842119286232374,0.9819523383582642,0.9831090469504268,0.0009232587374171034,2
+4.3404280344645185,0.08676557819811387,0.7395785649617513,0.021064106308070252,0.0017707168643537816,l2,50000,3,True,"{'clf__C': 0.0017707168643537816, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3, 'vectorizer__sublinear_tf': True}",0.5139475473905333,0.5726503036729831,0.5758080595813414,0.554135303548286,0.028446261199443323,8
+4.639631907145183,0.04971859206642914,0.7199817498524984,0.028834425139493622,0.00029152036385288323,l2,50000,5,True,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.034302515583123375,0.025982239501097307,0.023295542786146917,0.027860099290122533,0.004685661147143366,9
+4.061478853225708,0.06215993517270243,0.7521119117736816,0.02351433838498911,6.440507553993708,l2,50000,5,True,"{'clf__C': 6.440507553993708, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9836907160449173,0.9838789936753068,0.9802041985999305,0.9825913027733848,0.0016896867298839486,4
+4.818209648132324,0.29378475505492335,0.6650259494781494,0.049656153306557205,0.00014610865886287189,l2,50000,2,False,"{'clf__C': 0.00014610865886287189, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.02329192546583851,0.02329192546583851,0.023295542786146917,0.023293131239274647,1.7052211465316545e-06,10
+3.685816685358683,0.05627530678108159,0.7520771821339926,0.018130198468875022,456.60548734461145,l2,50000,2,False,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9821249321526742,0.9826325484874301,0.980029413093332,0.9815956312444788,0.0011267055393384378,5
+4.131646076838176,0.05632133127822176,0.6499122778574625,0.005002638936394404,0.0028483918709108036,l2,50000,1,True,"{'clf__C': 0.0028483918709108036, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.6541855665753074,0.6745892232208965,0.6813744904102502,0.6700497600688181,0.011554636561640236,7
+4.031825304031372,0.01084199153967617,0.6525913874308268,0.008953133516846668,0.027160511446548484,l2,50000,2,True,"{'clf__C': 0.027160511446548484, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': True}",0.8896291380999679,0.9752940758171058,0.890161144270522,0.9183614527291986,0.04025802972818929,6
+3.832550843556722,0.04273312072344384,0.6429457664489746,0.0124096852756637,0.28546978577971793,l2,50000,1,True,"{'clf__C': 0.28546978577971793, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9835010930237785,0.983022193578694,0.9812775009095039,0.9826002625039921,0.0009555486544442522,3

logistic_regression/multilingual_toxicity_dataset/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "multilingual_toxicity_dataset",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "0.09915644566638392",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "1",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.9854301572617946,
+    "f1_macro": 0.9853149526463483,
+    "precision_macro": 0.9859826911878914,
+    "recall_macro": 0.9856368834317932,
+    "inference_time_total_s": 0.3755,
+    "inference_time_ms_per_sample": 0.086836,
+    "throughput_samples_per_sec": 11515.93
+  }
+}

logistic_regression/stsb_multi_mt/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       0.99      0.99      0.99       783
+          el       0.00      0.00      0.00         0
+          en       0.98      0.99      0.99       738
+          es       0.99      0.99      0.99       818
+          fi       0.00      0.00      0.00         0
+          fr       1.00      0.99      1.00       784
+          hr       0.00      0.00      0.00         0
+          it       1.00      0.99      0.99       820
+          ja       0.00      0.00      0.00         0
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      0.99      0.99       766
+          pl       1.00      0.99      1.00       733
+          pt       0.99      0.99      0.99       801
+          ro       0.00      0.00      0.00         0
+          ru       1.00      1.00      1.00       721
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       0.00      0.00      0.00         0
+          zh       0.98      1.00      0.99       737
+
+    accuracy                           0.99      7701
+   macro avg       0.41      0.41      0.41      7701
+weighted avg       0.99      0.99      0.99      7701

logistic_regression/stsb_multi_mt/bow_maxabs_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+4.716332038243611,0.033669166650861865,1.0893371105194092,0.009346588189355357,0.09915644566638392,l2,50000,1,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1}",0.9882940669177043,0.990650339156544,0.9866336101764543,0.9885260054169009,0.0016480037470487697,5
+4.699404398600261,0.06490197621995386,1.092738151550293,0.012109625807892047,0.0029341230215132346,l2,50000,5,"{'clf__C': 0.0029341230215132346, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9778427787834453,0.9795172556595476,0.9773029116942148,0.9782209820457358,0.0009427293991055687,9
+4.706962903340657,0.09118766077772919,1.083038568496704,0.0012596301500125313,6.155564318973006,l2,50000,3,"{'clf__C': 6.155564318973006, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.988717093239116,0.9922552399519413,0.9880805642187355,0.9896842991365976,0.001836408597497985,3
+4.756442864735921,0.03578817266607296,1.0879961649576824,0.016369438873409463,0.3686905640846153,l2,50000,3,"{'clf__C': 0.3686905640846153, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3}",0.9886118944412576,0.9919450286841016,0.9874996325705132,0.9893521852319574,0.0018888108239088757,4
+5.062972068786621,0.04263837250513081,1.0298703511555989,0.009960285119840344,0.00029152036385288323,l2,50000,5,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9889564112322283,0.989344903242418,0.9864171821882393,0.9882394988876285,0.0012982963516727144,6
+4.678500413894653,0.04341698209804257,1.0894962151845295,0.016198095062798727,0.0467378670778782,l2,50000,5,"{'clf__C': 0.0467378670778782, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9869287782343079,0.9900604230513592,0.9859301073479683,0.9876397695445451,0.0017595469937541882,8
+4.742682456970215,0.15976416248502323,1.0770845413208008,0.04140823436453012,46.19347374377369,l2,50000,2,"{'clf__C': 46.19347374377369, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.988815118320473,0.9922625845079416,0.9882871806297044,0.9897882944860396,0.0017628126929608897,2
+5.043879667917888,0.06514892994177782,1.0084983507792156,0.012649626871591263,0.00028267840503514715,l2,50000,5,"{'clf__C': 0.00028267840503514715, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.9889582984604631,0.9890544746568818,0.9864308199381103,0.9881478643518183,0.0012147684564486633,7
+4.5661946932474775,0.16129829207674537,1.0662109057108562,0.018104495066729287,456.60548734461145,l2,50000,2,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2}",0.988815118320473,0.9922625845079416,0.9882906569719075,0.989789453266774,0.001761826425155435,1
+5.152834971745809,0.087561310196268,0.9657223224639893,0.024119761852694342,0.00010144487859320231,l2,50000,5,"{'clf__C': 0.00010144487859320231, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5}",0.8462281697436292,0.8400275040062753,0.8400185952794761,0.8420914230097937,0.0029251239285744733,10

logistic_regression/stsb_multi_mt/bow_maxabs_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,25 @@
+{
+  "run_name": "bow_maxabs_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "bow",
+    "scaler": "MaxAbsScaler",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "stsb_multi_mt",
+    "vectorizer_class": "CountVectorizer"
+  },
+  "params": {
+    "clf__C": "456.60548734461145",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "2"
+  },
+  "metrics": {
+    "accuracy": 0.9920789507856123,
+    "f1_macro": 0.9921168431288381,
+    "precision_macro": 0.9921252329617325,
+    "recall_macro": 0.9921589905758508,
+    "inference_time_total_s": 0.5729,
+    "inference_time_ms_per_sample": 0.074399,
+    "throughput_samples_per_sec": 13440.99
+  }
+}

logistic_regression/stsb_multi_mt/tfidf_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       1.00      0.99      1.00       783
+          el       0.00      0.00      0.00         0
+          en       0.98      1.00      0.99       738
+          es       1.00      0.99      1.00       818
+          fi       0.00      0.00      0.00         0
+          fr       1.00      0.99      1.00       784
+          hr       0.00      0.00      0.00         0
+          it       1.00      1.00      1.00       820
+          ja       0.00      0.00      0.00         0
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       1.00      0.99      0.99       766
+          pl       1.00      0.99      1.00       733
+          pt       1.00      1.00      1.00       801
+          ro       0.00      0.00      0.00         0
+          ru       1.00      1.00      1.00       721
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       0.00      0.00      0.00         0
+          zh       0.99      1.00      1.00       737
+
+    accuracy                           1.00      7701
+   macro avg       0.41      0.41      0.41      7701
+weighted avg       1.00      1.00      1.00      7701

logistic_regression/stsb_multi_mt/tfidf_lr_char_ngram_3_5/cv_results.csv

@@ -0,0 +1,11 @@
+mean_fit_time,std_fit_time,mean_score_time,std_score_time,param_clf__C,param_clf__penalty,param_vectorizer__max_features,param_vectorizer__min_df,param_vectorizer__sublinear_tf,params,split0_test_score,split1_test_score,split2_test_score,mean_test_score,std_test_score,rank_test_score
+4.862948735555013,0.02713757947858657,1.2074730396270752,0.01175361660279572,0.09915644566638392,l2,50000,1,True,"{'clf__C': 0.09915644566638392, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9909309928718406,0.9894025958880288,0.9735626007590467,0.984632063172972,0.007852122767549808,6
+4.342029333114624,0.08301768099923795,1.1870969136555989,0.0027877095213319826,71.77141927992008,l2,50000,1,True,"{'clf__C': 71.77141927992008, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9909543932315676,0.992202978478064,0.9902507982659593,0.9911360566585303,0.0008072600141446394,2
+5.297547499338786,0.10167505193669585,1.123515526453654,0.024829901913426686,0.0017707168643537816,l2,50000,3,True,"{'clf__C': 0.0017707168643537816, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 3, 'vectorizer__sublinear_tf': True}",0.6285875158014022,0.573677130341143,0.5828085546993258,0.5950244002806236,0.024023707322811297,8
+5.70340100924174,0.060631748756073245,1.1091025670369465,0.04083488103544944,0.00029152036385288323,l2,50000,5,True,"{'clf__C': 0.00029152036385288323, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.08293930192624308,0.09495161159647861,0.10702171359856576,0.09497087570709582,0.009831612832025554,10
+4.595508098602295,0.18645010041184368,1.192524512608846,0.008980676345992374,6.440507553993708,l2,50000,5,True,"{'clf__C': 6.440507553993708, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 5, 'vectorizer__sublinear_tf': True}",0.9924952584968227,0.9931932200866154,0.9901439770111983,0.9919441518648787,0.0013044180626047331,1
+5.940989335378011,0.2829331074385385,1.0091668764750164,0.021744272414171324,0.00014610865886287189,l2,50000,2,False,"{'clf__C': 0.00014610865886287189, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.08253890703815472,0.10650197930984706,0.10001139294506578,0.09635075976435585,0.010119531762467188,9
+4.284287134806315,0.06012390558164843,1.2297604878743489,0.06559458291513869,456.60548734461145,l2,50000,2,False,"{'clf__C': 456.60548734461145, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': False}",0.9903579756129239,0.9919081615067021,0.9900604843956573,0.9907755405050945,0.0008100403307036753,4
+4.994303385416667,0.06684144663787477,0.9738579591115316,0.010950182012580892,0.0028483918709108036,l2,50000,1,True,"{'clf__C': 0.0028483918709108036, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.8306698932575582,0.8029874551025236,0.8079964911188853,0.8138846131596557,0.012043859609142046,7
+4.858863830566406,0.14325703620079885,1.0002411206563313,0.03757748280223919,0.027160511446548484,l2,50000,2,True,"{'clf__C': 0.027160511446548484, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 2, 'vectorizer__sublinear_tf': True}",0.9892428099338192,0.9889992440808235,0.9889280148588308,0.9890566896244911,0.00013478124912785408,5
+4.592370271682739,0.03903599807242429,0.9755910237630209,0.02206236451445047,0.28546978577971793,l2,50000,1,True,"{'clf__C': 0.28546978577971793, 'clf__penalty': 'l2', 'vectorizer__max_features': 50000, 'vectorizer__min_df': 1, 'vectorizer__sublinear_tf': True}",0.9909267746319264,0.992690185367391,0.9897731482356809,0.9911300360783327,0.0011995173525604025,3

logistic_regression/stsb_multi_mt/tfidf_lr_char_ngram_3_5/run_metadata.json

@@ -0,0 +1,26 @@
+{
+  "run_name": "tfidf_lr_char_ngram_3_5",
+  "tags": {
+    "feature_type": "tfidf",
+    "scaler": "none",
+    "n_gram_range": "(3,5)",
+    "dataset_version": "stsb_multi_mt",
+    "vectorizer_class": "TfidfVectorizer"
+  },
+  "params": {
+    "clf__C": "6.440507553993708",
+    "clf__penalty": "l2",
+    "vectorizer__max_features": "50000",
+    "vectorizer__min_df": "5",
+    "vectorizer__sublinear_tf": "True"
+  },
+  "metrics": {
+    "accuracy": 0.9954551356966628,
+    "f1_macro": 0.9954399626268631,
+    "precision_macro": 0.9954229026821475,
+    "recall_macro": 0.9954906190072986,
+    "inference_time_total_s": 0.5699,
+    "inference_time_ms_per_sample": 0.073998,
+    "throughput_samples_per_sec": 13513.92
+  }
+}

logistic_regression/tweet_sentiment_multilingual/bow_maxabs_lr_char_ngram_3_5/classification_report.txt

@@ -0,0 +1,30 @@
+              precision    recall  f1-score   support
+
+          ca       0.00      0.00      0.00         0
+          da       0.00      0.00      0.00         0
+          de       0.99      0.99      0.99       271
+          el       0.00      0.00      0.00         0
+          en       0.99      1.00      0.99       292
+          es       1.00      0.99      0.99       288
+          fi       0.00      0.00      0.00         0
+          fr       0.99      1.00      1.00       299
+          hr       0.00      0.00      0.00         0
+          it       1.00      1.00      1.00       288
+          ja       0.00      0.00      0.00         0
+          ko       0.00      0.00      0.00         0
+          lt       0.00      0.00      0.00         0
+          mk       0.00      0.00      0.00         0
+          nb       0.00      0.00      0.00         0
+          nl       0.00      0.00      0.00         0
+          pl       0.00      0.00      0.00         0
+          pt       1.00      1.00      1.00       291
+          ro       0.00      0.00      0.00         0
+          ru       0.00      0.00      0.00         0
+          sl       0.00      0.00      0.00         0
+          sv       0.00      0.00      0.00         0
+          uk       0.00      0.00      0.00         0
+          zh       0.00      0.00      0.00         0
+
+    accuracy                           1.00      1729
+   macro avg       0.25      0.25      0.25      1729
+weighted avg       1.00      1.00      1.00      1729