Update etpc.py

etpc.py

@@ -65,19 +65,29 @@ class ETPC(datasets.GeneratorBasedBuilder):
                 "idx": datasets.Value("string"),
                 "sentence1": datasets.Value("string"),
                 "sentence2": datasets.Value("string"),
+                "sentence1_tokenized": datasets.Sequence(
+                    datasets.Value("string")
+                ),
+                "sentence2_tokenized": datasets.Sequence(
+                    datasets.Value("string")
+                ),
                 "etpc_label": datasets.Value("int8"),
                 "mrpc_label": datasets.Value("int8"),
                 "negation": datasets.Value("int8"),
-                "paraphrase_type": datasets.Value("string"),
-                "paraphrase_type_id": datasets.Value("int32"),
-                "sentence1_phrase_span": datasets.Sequence(
+                "paraphrase_types": datasets.Sequence(
+                    datasets.Value("string")
+                ),
+                "paraphrase_type_ids": datasets.Sequence(
+                    datasets.Value("string")
+                ),
+                "sentence1_segment_location": datasets.Sequence(
                     datasets.Value("int32")
                 ),
-                "sentence2_phrase_span": datasets.Sequence(
+                "sentence2_segment_location": datasets.Sequence(
                     datasets.Value("int32")
                 ),
-                "sentence1_phrase_text": datasets.Value("string"),
-                "sentence2_phrase_text": datasets.Value("string"),
+                "sentence1_segment_text": datasets.Value("string"),
+                "sentence2_segment_text": datasets.Value("string"),
             }
         )
 
@@ -116,7 +126,7 @@ class ETPC(datasets.GeneratorBasedBuilder):
         root_text_pairs = tree_text_pairs.getroot()
         root_paraphrase_types = tree_paraphrase_types.getroot()
 
-        idx = -1
+        idx = 0
 
         for row in root_text_pairs:
             current_pair_id = row.find(".//pair_id").text
@@ -126,49 +136,64 @@ class ETPC(datasets.GeneratorBasedBuilder):
             paraphrase_type_ids = root_paraphrase_types.xpath(
                 f".//pair_id[text()='{current_pair_id}']/parent::relation/type_id/text()"
             )
-            sentence1_phrase_text = root_paraphrase_types.xpath(
-                f".//pair_id[text()='{current_pair_id}']/parent::relation/s1_text/text()"
-            )
-            sentence2_phrase_text = root_paraphrase_types.xpath(
-                f".//pair_id[text()='{current_pair_id}']/parent::relation/s2_text/text()"
-            )
-            sentence1_phrase_spans = root_paraphrase_types.xpath(
+            sentence1_segment_location = root_paraphrase_types.xpath(
                 f".//pair_id[text()='{current_pair_id}']/parent::relation/s1_scope/text()"
             )
-            sentence2_phrase_spans = root_paraphrase_types.xpath(
+            sentence2_segment_location = root_paraphrase_types.xpath(
                 f".//pair_id[text()='{current_pair_id}']/parent::relation/s2_scope/text()"
             )
+            sentence1_segment_text = root_paraphrase_types.xpath(
+                f".//pair_id[text()='{current_pair_id}']/parent::relation/s1_text/text()"
+            )
+            sentence2_segment_text = root_paraphrase_types.xpath(
+                f".//pair_id[text()='{current_pair_id}']/parent::relation/s2_text/text()"
+            )
+            sentence1_tokenized = row.find(".//sent1_tokenized").text.split()
+            sentence2_tokenized = row.find(".//sent2_tokenized").text.split()
+
+            sentence1_segment_location_full = [0] * len(sentence1_tokenized)
+            sentence2_segment_location_full = [0] * len(sentence2_tokenized)
+
             for (
-                paraphrase_type,
+                sentence1_segment_location_example,
+                sentence2_segment_location_example,
                 paraphrase_type_id,
-                sentence1_phrase_span,
-                sentence2_phrase_span,
-                sentence1_phrase_text,
-                sentence2_phrase_text,
             ) in zip(
-                paraphrase_types,
+                sentence1_segment_location,
+                sentence2_segment_location,
                 paraphrase_type_ids,
-                sentence1_phrase_spans,
-                sentence2_phrase_spans,
-                sentence1_phrase_text,
-                sentence2_phrase_text,
             ):
-                idx += 1
-                yield idx, {
-                    "idx": row.find(".//pair_id").text
-                    + "_"
-                    + paraphrase_type_id
-                    + "_"
-                    + str(idx),
-                    "sentence1": row.find(".//sent1_raw").text,
-                    "sentence2": row.find(".//sent2_raw").text,
-                    "etpc_label": int(row.find(".//etpc_label").text),
-                    "mrpc_label": int(row.find(".//mrpc_label").text),
-                    "negation": int(row.find(".//negation").text),
-                    "paraphrase_type": paraphrase_type,
-                    "paraphrase_type_id": paraphrase_type_id,
-                    "sentence1_phrase_span": sentence1_phrase_span.split(","),
-                    "sentence2_phrase_span": sentence2_phrase_span.split(","),
-                    "sentence1_phrase_text": sentence1_phrase_text,
-                    "sentence2_phrase_text": sentence2_phrase_text,
-                }
+                print(paraphrase_type_id)
+                print(sentence1_segment_location_example.split(","))
+                print(sentence2_segment_location_example.split(","))
+
+                for (
+                    segment_location
+                ) in sentence1_segment_location_example.split(","):
+                    sentence1_segment_location_full[
+                        int(segment_location)
+                    ] = int(paraphrase_type_id)
+                for (
+                    segment_location
+                ) in sentence2_segment_location_example.split(","):
+                    sentence2_segment_location_full[
+                        int(segment_location)
+                    ] = int(paraphrase_type_id)
+            yield idx, {
+                "idx": row.find(".//pair_id").text + "_" + str(idx),
+                "sentence1": row.find(".//sent1_raw").text,
+                "sentence2": row.find(".//sent2_raw").text,
+                "sentence1_tokenized": sentence1_tokenized,
+                "sentence2_tokenized": sentence2_tokenized,
+                "etpc_label": int(row.find(".//etpc_label").text),
+                "mrpc_label": int(row.find(".//mrpc_label").text),
+                "negation": int(row.find(".//negation").text),
+                "paraphrase_types": paraphrase_types,
+                "paraphrase_type_ids": paraphrase_type_ids,
+                "sentence1_segment_location": sentence1_segment_location_full,
+                "sentence2_segment_location": sentence2_segment_location_full,
+                "sentence1_segment_text": sentence1_segment_text,
+                "sentence2_segment_text": sentence2_segment_text,
+            }
+
+            idx += 1