Update humsetbias.py

humsetbias.py

@@ -47,8 +47,8 @@ _URLs = {
     "1.0.0": {
         "train": "data/humset_bias_train.jsonl",
         "dev": "data/humset_bias_val.jsonl",
-        "gender": "data/test_gender.jsonl",
-        "country": "data/test_country.jsonl"
+        #"gender": "data/test_gender.jsonl",
+        #"country": "data/test_country.jsonl"
     }
 }
 
@@ -194,18 +194,18 @@ class Humset(datasets.GeneratorBasedBuilder):
                     "filepath": downloaded_files["dev"]
                 },
             ),
-            datasets.SplitGenerator(
-                name="test_gender",
-                gen_kwargs={
-                    "filepath": downloaded_files["gender"]
-                },
-            ),
-            datasets.SplitGenerator(
-                name="test_country",
-                gen_kwargs={
-                    "filepath": downloaded_files["country"]
-                }
-            ),
+            #datasets.SplitGenerator(
+            #    name="test_gender",
+            #    gen_kwargs={
+            #        "filepath": downloaded_files["gender"]
+            #    },
+            #),
+            #datasets.SplitGenerator(
+            #    name="test_country",
+            #    gen_kwargs={
+            #        "filepath": downloaded_files["country"]
+            #    }
+            #),
         ]
 
         return splits
@@ -218,12 +218,12 @@ class Humset(datasets.GeneratorBasedBuilder):
             idx = 0
             for line in data:
                 row = json.loads(line)
-                if "train" in filepath:
-                    row = {k: v for k, v in row.items() if k in HUMSETBIAS_GENERAL}
-                elif "gender" in filepath:
-                    row = {k: v for k, v in row.items() if k in HUMSETBIAS_FEATURES_GENDER}
-                elif "country" in filepath:
-                    row = {k: v for k, v in row.items() if k in HUMSETBIAS_FEATURES_COUNTRY}
+                #if "train" in filepath:
+                #    row = {k: v for k, v in row.items() if k in HUMSETBIAS_GENERAL}
+                #elif "gender" in filepath:
+                #    row = {k: v for k, v in row.items() if k in HUMSETBIAS_FEATURES_GENDER}
+                #elif "country" in filepath:
+                #    row = {k: v for k, v in row.items() if k in HUMSETBIAS_FEATURES_COUNTRY}
                 #if self.config.name == "1.0.0":
                 yield idx, row
                 idx+=1