Update BabyLM.py

BabyLM.py

@@ -13,6 +13,21 @@ _HOMEPAGE = "https://babylm.github.io"
 
 filenames = [
     "aochildes.txt",
+    "aocatalan.txt",
+    "aocatalan_dev.txt",
+    "aocatalan_test.txt",
+    "aospanish.txt",
+    "aospanish_dev.txt",
+    "aospanish_test.txt",
+    "aoromanian.txt",
+    "aoromanian_dev.txt",
+    "aoromanian_test.txt",
+    "aoportuguese.txt",
+    "aoportuguese_dev.txt",
+    "aoportuguese_test.txt",
+    "aodutch.txt",
+    "aodutch_dev.txt",
+    "aodutch_test.txt",
     "aochinese.txt",
     "aochinese_dev.txt",
     "aochinese_test.txt",
@@ -115,7 +130,31 @@ class BabyLM(datasets.GeneratorBasedBuilder):
             description="GERMAN Cleaned version of the dataset, 10M words, unsupervised POS tags",
             version="1.0.0",
         ),
-        
+        datasets.BuilderConfig(
+            name="nld_lang_small",
+            description="DUTCH Cleaned version of the dataset, 10M words, unsupervised POS tags",
+            version="1.0.0",
+        ),
+        datasets.BuilderConfig(
+            name="es_lang_small",
+            description="SPANISH Cleaned version of the dataset, 10M words, unsupervised POS tags",
+            version="1.0.0",
+        ),
+        datasets.BuilderConfig(
+            name="cat_lang_small",
+            description="CATALAN Cleaned version of the dataset, 10M words, unsupervised POS tags",
+            version="1.0.0",
+        ),
+        datasets.BuilderConfig(
+            name="po_lang_small",
+            description="PORTUGUESE Cleaned version of the dataset, 10M words, unsupervised POS tags",
+            version="1.0.0",
+        ),
+        datasets.BuilderConfig(
+            name="ron_lang_small",
+            description="ROMANIAN Cleaned version of the dataset, 10M words, unsupervised POS tags",
+            version="1.0.0",
+        ),
         datasets.BuilderConfig(
             name="fr_lang_gold",
             description="FRENCH Cleaned version of the dataset, 100M words, gold POS tags",
@@ -126,7 +165,7 @@ class BabyLM(datasets.GeneratorBasedBuilder):
             description="JAPANESE Cleaned version of the dataset, 100M words, gold POS tags",
             version="1.0.0",
         ),
-                datasets.BuilderConfig(
+        datasets.BuilderConfig(
             name="de_lang_gold",
             description="GERMAN Cleaned version of the dataset, 100M words, gold POS tags",
             version="1.0.0",
@@ -174,6 +213,16 @@ class BabyLM(datasets.GeneratorBasedBuilder):
             train_data_dir = "ZH"
         elif "ja_lang_small" in self.config.name:
             train_data_dir = "JA"
+        elif "es_lang_small" in self.config.name:
+            train_data_dir = "SP"
+        elif "nld_lang_small" in self.config.name:
+            train_data_dir = "DUT"
+        elif "po_lang_small" in self.config.name:
+            train_data_dir = "PORT"
+        elif "ron_lang_small" in self.config.name:
+            train_data_dir = "ROM"
+        elif "cat_lang_small" in self.config.name:
+            train_data_dir = "CAT"
         elif "strict_small" in self.config.name: #default settings – English
             train_data_dir = "10M"
         else: 
@@ -208,6 +257,26 @@ class BabyLM(datasets.GeneratorBasedBuilder):
             urls_to_download["train"].append(f"{folder}/JA/aojapanese.txt")
             urls_to_download["dev"].append(f"{folder}/dev/aojapanese_dev.txt")
             urls_to_download["test"].append(f"{folder}/test/aojapanese_test.txt")
+        elif 'cat_lang_small' in self.config.name:
+            urls_to_download["train"].append(f"{folder}/CAT/aocatalan.txt")
+            urls_to_download["dev"].append(f"{folder}/dev/aocatalan_dev.txt")
+            urls_to_download["test"].append(f"{folder}/test/aocatalan_test.txt")
+        elif 'es_lang_small' in self.config.name:
+            urls_to_download["train"].append(f"{folder}/SP/aospanish.txt")
+            urls_to_download["dev"].append(f"{folder}/dev/aospanish_dev.txt")
+            urls_to_download["test"].append(f"{folder}/test/aospanish_test.txt")
+        elif 'nld_lang_small' in self.config.name:
+            urls_to_download["train"].append(f"{folder}/DUT/aodutch.txt")
+            urls_to_download["dev"].append(f"{folder}/dev/aodutch_dev.txt")
+            urls_to_download["test"].append(f"{folder}/test/aodutch_test.txt")
+        elif 'po_lang_small' in self.config.name:
+            urls_to_download["train"].append(f"{folder}/PORT/aoportuguese.txt")
+            urls_to_download["dev"].append(f"{folder}/dev/aoportuguese_dev.txt")
+            urls_to_download["test"].append(f"{folder}/test/aoportuguese_test.txt")
+        elif 'ron_lang_small' in self.config.name:
+            urls_to_download["train"].append(f"{folder}/ROM/aoromanian.txt")
+            urls_to_download["dev"].append(f"{folder}/dev/aoromanian_dev.txt")
+            urls_to_download["test"].append(f"{folder}/test/aoromanian_test.txt")
         else:
             urls_to_download["train"] = [f"{folder}/10M/{fn}" for fn in en_filenames]
             urls_to_download["dev"] = [f"{folder}/dev/{fn}" for fn in en_filenames]