resolving card bugs

README.md

@@ -9,39 +9,21 @@ task_categories:
 tags:
 - legal
 configs:
-- config_name: sentences_labeled
+- config_name: text_raw
   data_files:
   - split: train
-    path: sentences_labeled/train*.parquet
-  - split: test
-    path: sentences_labeled/test*.parquet
+    path: text_raw/train*.parquet
 - config_name: sentences_unlabeled
   data_files:
   - split: train
     path: sentences_unlabeled/train*.parquet
-- config_name: text_raw
+- config_name: sentences_labeled
   data_files:
   - split: train
-    path: text_raw/train-*
-dataset_info:
-  config_name: text_raw
-  features:
-  - name: source_id
-    dtype: int64
-  - name: source_name
-    dtype: string
-  - name: text_id
-    dtype: int64
-  - name: text
-    dtype: string
-  - name: extension
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 605308
-    num_examples: 867
-  download_size: 288502
-  dataset_size: 605308
+    path: labeled_sentences/train*.parquet
+  - split: test
+    path: sentences_labeled/test*.parquet
+
 ---
 
 # Paraguay Legislation

py_legislation.py

@@ -117,14 +117,14 @@ _metadata = {
     "license": "apache-2.0",
 
     "urls": {
-        "text_raw": "./text_raw",
-        "sentences_unlabeled": "./unlabeled",
-        "sentences_labeled": "./labeled",
+        "raw_text": "./raw_text",
+        "unlabeled_sentences": "./unlabeled",
+        "labeled_sentences": "./labeled",
     },
 
     # [@] Config Names:
 
-    "text_raw": {
+    "raw_text": {
         "description": textwrap.dedent("""
         Data extracted from the sources files (URls, PDFs and Word files) without any transformation or sentence splitter. It can be helpful because you can access the raw data extracted from the seeds (PDFs and Word files) and apply other preprocessing tasks from this point to prepare the data without returning to extract texts from source files.
         """),
@@ -140,7 +140,7 @@ _metadata = {
 
     },
 
-    "sentences_unlabeled": {
+    "unlabeled_sentences": {
         "description": textwrap.dedent("""
         Unlabeled corpora of Paraguay legislation. This data is prepared to be labeled by the experts. Each instance of the dataset represents a specific text passage, split by its original formatting extracted from raw text (from original documents)
 
@@ -165,7 +165,7 @@ _metadata = {
         }
     },
 
-    "sentences_labeled": {
+    "labeled_sentences": {
         "description": textwrap.dedent("""
         The labeled data is the ground truth data used to train the models. This data is annotated  by legal experts indicating the existence of administrative costs (and other types) in the legislation.
 
@@ -197,26 +197,29 @@ class PY_legislation(datasets.GeneratorBasedBuilder):
 
     BUILDER_CONFIGS = [
         datasets.BuilderConfig(
-            name="text_raw",
+            name="raw_text",
+            data_dir=_metadata["urls"]["raw_text"],
             version=VERSION,
-            description=_metadata["text_raw"]["description"],
+            description=_metadata["raw_text"]["description"],
         ),
 
         datasets.BuilderConfig(
-            name="sentences_unlabeled",
+            name="unlabeled_sentences",
             version=VERSION,
-            description=_metadata["sentences_unlabeled"]["description"],
+            data_dir=_metadata["urls"]["unlabeled_sentences"],
+            description=_metadata["unlabeled_sentences"]["description"],
         ),
 
         datasets.BuilderConfig(
-            name="sentences_labeled",
+            name="labeled_sentences",
             version=VERSION,
-            description=_metadata["sentences_labeled"]["description"],
+            data_dir=_metadata["urls"]["labeled_sentences"],
+            description=_metadata["labeled_sentences"]["description"],
         ),
     ]
 
     # It's not mandatory to have a default configuration. Just use one if it make sense.
-    DEFAULT_CONFIG_NAME = "text_raw"
+    DEFAULT_CONFIG_NAME = "raw_text"
 
     # [i] Info
     def _info(self):
@@ -227,8 +230,8 @@ class PY_legislation(datasets.GeneratorBasedBuilder):
         features = None
         description = ""
 
-        if self.config.name == "text_raw":
-            description = _metadata["text_raw"]["description"]
+        if self.config.name == "raw_text":
+            description = _metadata["raw_text"]["description"]
             features = datasets.Features(
                 {
                     "source_id": datasets.Value(dtype="int64"),
@@ -240,19 +243,19 @@ class PY_legislation(datasets.GeneratorBasedBuilder):
                 }
             )
 
-        if self.config.name == "sentences_unlabeled":
-            description = _metadata["sentences_unlabeled"]["description"]
+        if self.config.name == "unlabeled_sentences":
+            description = _metadata["unlabeled_sentences"]["description"]
             features = datasets.Features(
-                _metadata["sentences_unlabeled"]["features"])
+                _metadata["unlabeled_sentences"]["features"])
 
-        if self.config.name == "sentences_labeled":
-            description = _metadata["sentences_labeled"]["description"]
+        if self.config.name == "labeled_sentences":
+            description = _metadata["labeled_sentences"]["description"]
             features = datasets.Features(
-                _metadata["sentences_labeled"]["features"])
+                _metadata["labeled_sentences"]["features"])
 
         else:
             features = datasets.Features(
-                _metadata["text_raw"]["description"]
+                _metadata["raw_text"]["description"]
             )
 
         return datasets.DatasetInfo(
@@ -272,24 +275,24 @@ class PY_legislation(datasets.GeneratorBasedBuilder):
         # generators = [
         #     datasets.SplitGenerator(
         #         name=datasets.Split.TRAIN,
-        #         gen_kwargs={"filepath": downloaded_files["text_raw"]},
+        #         gen_kwargs={"filepath": downloaded_files["raw_text"]},
         #     ),
 
         #     datasets.SplitGenerator(
         #     name=datasets.Split.TRAIN,
         #     gen_kwargs={
-        #         "filepath": downloaded_files["sentences_unlabeled"]},
+        #         "filepath": downloaded_files["unlabeled_sentences"]},
         #     ),
 
         #     datasets.SplitGenerator(
         #         name=datasets.Split.TRAIN,
-        #         gen_kwargs={"filepath": downloaded_files["sentences_labeled_train"]},
+        #         gen_kwargs={"filepath": downloaded_files["labeled_sentences_train"]},
         #     ),
 
         #     datasets.SplitGenerator(
         #         name=datasets.Split.TEST,
         #         gen_kwargs={
-        #             "filepath": downloaded_files["sentences_labeled_test"]},
+        #             "filepath": downloaded_files["labeled_sentences_test"]},
         #     )
         # ]