first push to github

distribution_matching/commons.py

@@ -10,16 +10,16 @@ from sklearn.linear_model import LogisticRegression
 
 # set to True to get the full list of methods tested in the paper (reported in the appendix)
 # set to False to get the reduced list (shown in the body of the paper)
-FULL_METHOD_LIST = True
+FULL_METHOD_LIST = False
 
 if FULL_METHOD_LIST:
     ADJUSTMENT_METHODS = ['ACC', 'PACC']
     DISTR_MATCH_METHODS = ['HDy-OvA', 'DM-T', 'DM-HD', 'KDEy-HD',  'DM-CS', 'KDEy-CS']
-    MAX_LIKE_METHODS = ['DIR', 'EMQ', 'EMQ-BCTS', 'KDEy-ML', 'KDEx-ML']
+    MAX_LIKE_METHODS = ['DIR', 'EMQ', 'EMQ-BCTS', 'KDEy-ML']
 else:
     ADJUSTMENT_METHODS = ['PACC']
     DISTR_MATCH_METHODS = ['DM-T', 'DM-HD', 'KDEy-HD',  'DM-CS', 'KDEy-CS']
-    MAX_LIKE_METHODS = ['EMQ', 'KDEy-ML', 'KDEx-ML']
+    MAX_LIKE_METHODS = ['EMQ', 'KDEy-ML']
 
 # list of methods to consider
 METHODS  = ADJUSTMENT_METHODS + DISTR_MATCH_METHODS + MAX_LIKE_METHODS

distribution_matching/tables/gen_tables_compact.py

@@ -67,11 +67,11 @@ def make_table(tabs, eval, benchmark_groups, benchmark_names, compact=False):
     for i, (tab, group, name) in enumerate(zip(tabs, benchmark_groups, benchmark_names)):
         tablines = tab.latexTabular(benchmark_replace=nice_bench, endl='\\\\'+ cline, aslines=True)
         tablines[0] = tablines[0].replace('\multicolumn{1}{c|}{}', '\\textbf{'+name+'}')
-        if not compact:
+        if compact or len(tab.benchmarks)==1:
-            tabular += '\n'.join(tablines)
-        else:
             # if compact, keep the method names and the average; discard the rest
-            tabular += tablines[0] + '\n' + tablines[-1] + '\n'
+            tabular += tablines[0] + '\n' + tablines[1 if len(tab.benchmarks)==1 else -1] + '\n'
+        else:
+            tabular += '\n'.join(tablines)
 
         tabular += "\n" + "\\textit{Rank} & " + tab.getRankTable(prec_mean=0 if name.startswith('LeQua') else 1).latexAverage()
         if i < (len(tabs) - 1):
@@ -158,7 +158,7 @@ def gen_tables_tweet(eval):
 def gen_tables_lequa(Methods, task, eval):
     # generating table for LeQua-T1A or Lequa-T1B; only one table with two rows, one for MAE, another for MRAE
 
-    tab = new_table([f'Average'], Methods)
+    tab = new_table([task], Methods)
 
     print('Generating table for T1A@Lequa', eval, end='')
     dir_results = f'../results/lequa/{task}/{eval}'
@@ -168,7 +168,7 @@ def gen_tables_lequa(Methods, task, eval):
         if os.path.exists(result_path):
             df = pd.read_csv(result_path)
             print(f'{method}', end=' ')
-            tab.add('Average', method, df[eval].values)
+            tab.add(task, method, df[eval].values)
         else:
             print(f'MISSING-{method}', end=' ')
     print()
@@ -186,7 +186,7 @@ if __name__ == '__main__':
         tabs.append(gen_tables_uci_multiclass(eval))
         tabs.append(gen_tables_lequa(METHODS, 'T1B', eval))
 
-        names = ['Tweets', 'UCI-multi', 'LeQua-T1B']
+        names = ['Tweets', 'UCI-multi', 'LeQua']
         table = make_table(tabs, eval, benchmark_groups=tabs, benchmark_names=names)
         save_table(f'./latex/multiclass_{eval}.tex', table)
 
@@ -200,7 +200,7 @@ if __name__ == '__main__':
         
         # print uci-bin compacted plus lequa-T1A for the main body
         tabs.append(gen_tables_lequa(BIN_METHODS, 'T1A', eval))
-        table = make_table(tabs, eval, benchmark_groups=tabs, benchmark_names=['UCI-binary', 'LeQua-T1A'], compact=True)
+        table = make_table(tabs, eval, benchmark_groups=tabs, benchmark_names=['UCI-binary', 'LeQua'], compact=True)
         save_table(f'./latex/binary_{eval}.tex', table)
 
     print("[Tables Done] runing latex")

examples/uci_experiments.py

@@ -116,7 +116,7 @@ def run(experiment):
             model,
             protocol=APP(test, n_prevalences=21, repeats=100)
         )
-        test_true_prevalence = data.test.prevalence()
+        test_true_prevalence = data.mixture.prevalence()
 
         evaluate_experiment(true_prevalences, estim_prevalences)
         save_results(dataset_name, model_name, run, optim_loss,

laboratory/main_tweets_auto.py

@@ -48,7 +48,7 @@ if __name__ == '__main__':
         csv.write(f'Method\tDataset\tMAE\tMRAE\n')
         for data, quantifier, quant_name in gen_methods():
             quantifier.fit(data.training)
-            protocol = UPP(data.test, repeats=100)
+            protocol = UPP(data.mixture, repeats=100)
             report = qp.evaluation.evaluation_report(quantifier, protocol, error_metrics=['mae', 'mrae'], verbose=True)
             means = report.mean()
             csv.write(f'{quant_name}\t{data.name}\t{means["mae"]:.5f}\t{means["mrae"]:.5f}\n')

laboratory/method_dxs.py

@@ -133,7 +133,7 @@ if __name__ == '__main__':
         csv.write(f'Method\tDataset\tMAE\tMRAE\n')
         for data, quantifier, quant_name in gen_methods():
             quantifier.fit(data.training)
-            report = qp.evaluation.evaluation_report(quantifier, APP(data.test, repeats=repeats), error_metrics=['mae','mrae'], verbose=True)
+            report = qp.evaluation.evaluation_report(quantifier, APP(data.mixture, repeats=repeats), error_metrics=['mae', 'mrae'], verbose=True)
             means = report.mean()
             csv.write(f'{quant_name}\t{data.name}\t{means["mae"]:.5f}\t{means["mrae"]:.5f}\n')
 

quapy/method/neural.py

@@ -38,7 +38,7 @@ class QuaNetTrainer(BaseQuantifier):
     >>> # train QuaNet (QuaNet is an alias to QuaNetTrainer)
     >>> model = QuaNet(classifier, qp.environ['SAMPLE_SIZE'], device='cuda')
     >>> model.fit(dataset.training)
-    >>> estim_prevalence = model.quantify(dataset.test.instances)
+    >>> estim_prevalence = model.quantify(dataset.mixture.instances)
 
     :param classifier: an object implementing `fit` (i.e., that can be trained on labelled data),
         `predict_proba` (i.e., that can generate posterior probabilities of unlabelled examples) and