Update metric_for_tp_fp_samples.py

metric_for_tp_fp_samples.py

@@ -191,7 +191,7 @@ class metric_tp_fp_Datasets(evaluate.Metric):
 		return results
 
 	#Computes the metric for each prediction strategy##############################################
-	def _compute(self, predictions, references, prediction_strategies = [["argmax_max"],]):
+	def _compute(self, predictions, references, prediction_strategies = [["argmax_max"],], FPifWrong = False):
 		"""Returns the scores"""
 		# TODO: Compute the different scores of the metric
 		predictions = torch.from_numpy(np.array(predictions, dtype = 'float32'))
@@ -223,6 +223,9 @@ class metric_tp_fp_Datasets(evaluate.Metric):
 						if j[0] in i:
 							TP_data.loc[TP_data["class"] == j[0], "coincidence count"] += 1
 							TP_data = TP_data.sort_values(by=["class"], ignore_index  = True)
+						elif FPifWrong:
+							FP_data.loc[TP_data["class"] == j[0], "coincidence count"] += 1
+							FP_data = TP_data.sort_values(by=["class"], ignore_index  = True)
 				if j[1] == 2:
 					FP_data.loc[FP_data["class"] == j[0], "number of samples"] += 1
 					if len(i) >> 0: