jsonlines

dataset_infos.json

@@ -1 +1,65 @@
-{"default": {"description": "The Norwegian Parliament Speeches is a collection of text passages from\n1998 to 2016 and pronounced at the Norwegian Parliament (Storting) by members\nof the two major parties: Fremskrittspartiet and Sosialistisk Venstreparti.\n", "citation": "\n@InProceedings{--,\n  author = {---},\n  title = {---},\n  booktitle = {---},\n  year = 2021,\n  address = \"---\"\n}\n", "homepage": "https://github.com/NBAiLab/notram/", "license": "", "features": {"text": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"num_classes": 2, "names": ["Fremskrittspartiet", "Sosialistisk Venstreparti"], "names_file": null, "id": null, "_type": "ClassLabel"}, "date": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "norwegian_parliament", "config_name": "default", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 6646256, "num_examples": 3600, "dataset_name": "norwegian_parliament"}, "validation": {"name": "validation", "num_bytes": 2384359, "num_examples": 1200, "dataset_name": "norwegian_parliament"}, "test": {"name": "test", "num_bytes": 2350756, "num_examples": 1200, "dataset_name": "norwegian_parliament"}}, "download_checksums": {"https://storage.googleapis.com/notram-public/finetune_datasets/parliament_speeches_1998_2016_frp_or_sv/train.csv": {"num_bytes": 6691149, "checksum": "5138fa8521424df0feca43d64a6a75fb1f4d9f2e759cc78027a1dc3bb8bde7ea"}, "https://storage.googleapis.com/notram-public/finetune_datasets/parliament_speeches_1998_2016_frp_or_sv/dev.csv": {"num_bytes": 2399289, "checksum": "b5397e989015c558bc5fb5cc706282b3306a41e09e950559ac0b3eaea8d72dd2"}, "https://storage.googleapis.com/notram-public/finetune_datasets/parliament_speeches_1998_2016_frp_or_sv/test.csv": {"num_bytes": 2365709, "checksum": "d1ebce8715e69884be78713e78d34635c2e6bb67db05c46e5833c2fb433f3825"}}, "download_size": 11456147, "post_processing_size": null, "dataset_size": 11381371, "size_in_bytes": 22837518}}
+{
+  "NbAiLab/norwegian_parliament": {
+    "description": "The Norwegian Parliament Speeches dataset contains transcriptions of speeches made by members of the Norwegian Parliament. The dataset covers speeches from two major parties: Fremskrittspartiet and Sosialistisk Venstreparti from 1998 to 2016.",
+    "citation": "@misc{nbailab_norwegian_parliament, title={Norwegian Parliament Speech Dataset}, author={NbAiLab}, year={2021}}",
+    "homepage": "https://github.com/NBAiLab/notram/",
+    "license": "",
+    "features": {
+      "label": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "text": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "full_label": {
+        "dtype": "string",
+        "_type": "Value"
+      },
+      "date": {
+        "dtype": "string",
+        "_type": "Value"
+      }
+    },
+    "splits": {
+      "train": {
+        "name": "train",
+        "num_bytes": 7513062,
+        "num_examples": 3600,
+        "dataset_name": "NbAiLab/norwegian_parliament"
+      },
+      "validation": {
+        "name": "validation",
+        "num_bytes": 2689316,
+        "num_examples": 1200,
+        "dataset_name": "NbAiLab/norwegian_parliament"
+      },
+      "test": {
+        "name": "test",
+        "num_bytes": 2651233,
+        "num_examples": 1200,
+        "dataset_name": "NbAiLab/norwegian_parliament"
+      }
+    },
+    "post_processed": null,
+    "supervised_keys": {
+      "input": "text",
+      "output": "label"
+    },
+    "builder_name": "norwegian_parliament",
+    "config_name": "default",
+    "version": {
+      "version_str": "1.0.0",
+      "description": "Initial version.",
+      "major": 1,
+      "minor": 0,
+      "patch": 0
+    },
+    "download_size": 12848611,
+    "post_processing_size": null,
+    "dataset_size": 12848611,
+    "size_in_bytes": 25697222
+  }
+}
+

utils/convert_csv_to_json.py

@@ -0,0 +1,47 @@
+import csv
+import json
+import argparse
+
+def convert_csv_to_jsonl(input_csv_path):
+    # Generate the output file name by replacing .csv with .jsonl
+    output_jsonl_path = input_csv_path.rsplit('.', 1)[0] + '.jsonl'
+
+    # Open the input CSV file and the output JSON lines file
+    with open(input_csv_path, mode='r', encoding='utf-8') as csv_file, \
+         open(output_jsonl_path, mode='w', encoding='utf-8') as jsonl_file:
+
+        # Create a CSV reader object to read the CSV file
+        csv_reader = csv.reader(csv_file, delimiter=',')
+        
+        # Iterate over each row in the CSV file
+        for row in csv_reader:
+            # Each row should have exactly four columns: label, text, full_label, date
+            if len(row) == 4:
+                label, text, full_label, date = row
+                # Create a dictionary to represent this record
+                record = {
+                    'label': label,
+                    'text': text,
+                    'full_label': full_label,
+                    'date': date
+                }
+                # Convert the dictionary to a JSON string and write to the JSON lines file
+                jsonl_file.write(json.dumps(record) + '\n')
+            else:
+                print(f"Skipping malformed row: {row}")
+
+def main():
+    # Create argument parser
+    parser = argparse.ArgumentParser(description="Convert CSV file to JSONL file.")
+    # Add argument for input CSV file
+    parser.add_argument("input_csv_path", type=str, help="Path to the input CSV file.")
+    
+    # Parse arguments
+    args = parser.parse_args()
+
+    # Convert CSV to JSONL using the provided input path
+    convert_csv_to_jsonl(args.input_csv_path)
+
+if __name__ == "__main__":
+    main()
+

utils/convert_tsv_to_json.py

@@ -0,0 +1,42 @@
+import csv
+import json
+import argparse
+
+def convert_tsv_to_jsonl(input_tsv_path):
+    # Generate the output file name by replacing .tsv with .jsonl
+    output_jsonl_path = input_tsv_path.rsplit('.', 1)[0] + '.jsonl'
+
+    # Open the input TSV file and the output JSON lines file
+    with open(input_tsv_path, mode='r', encoding='utf-8') as tsv_file, \
+         open(output_jsonl_path, mode='w', encoding='utf-8') as jsonl_file:
+
+        # Create a CSV reader object to read the TSV file
+        tsv_reader = csv.reader(tsv_file, delimiter='\t')
+        
+        # Iterate over each row in the TSV file
+        for row in tsv_reader:
+            # Each row should have exactly two columns: label and text
+            if len(row) == 2:
+                label, text = row
+                # Create a dictionary to represent this record
+                record = {'label': label, 'text': text}
+                # Convert the dictionary to a JSON string and write to the JSON lines file
+                jsonl_file.write(json.dumps(record) + '\n')
+            else:
+                print(f"Skipping malformed row: {row}")
+
+def main():
+    # Create argument parser
+    parser = argparse.ArgumentParser(description="Convert TSV file to JSONL file.")
+    # Add argument for input TSV file
+    parser.add_argument("input_tsv_path", type=str, help="Path to the input TSV file.")
+    
+    # Parse arguments
+    args = parser.parse_args()
+
+    # Convert TSV to JSONL using the provided input path
+    convert_tsv_to_jsonl(args.input_tsv_path)
+
+if __name__ == "__main__":
+    main()
+

utils/dataset_infos.json

@@ -0,0 +1 @@
+{"default": {"description": "The Norwegian Parliament Speeches is a collection of text passages from\n1998 to 2016 and pronounced at the Norwegian Parliament (Storting) by members\nof the two major parties: Fremskrittspartiet and Sosialistisk Venstreparti.\n", "citation": "\n@InProceedings{--,\n  author = {---},\n  title = {---},\n  booktitle = {---},\n  year = 2021,\n  address = \"---\"\n}\n", "homepage": "https://github.com/NBAiLab/notram/", "license": "", "features": {"text": {"dtype": "string", "id": null, "_type": "Value"}, "label": {"num_classes": 2, "names": ["Fremskrittspartiet", "Sosialistisk Venstreparti"], "names_file": null, "id": null, "_type": "ClassLabel"}, "date": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "builder_name": "norwegian_parliament", "config_name": "default", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 6646256, "num_examples": 3600, "dataset_name": "norwegian_parliament"}, "validation": {"name": "validation", "num_bytes": 2384359, "num_examples": 1200, "dataset_name": "norwegian_parliament"}, "test": {"name": "test", "num_bytes": 2350756, "num_examples": 1200, "dataset_name": "norwegian_parliament"}}, "download_checksums": {"https://storage.googleapis.com/notram-public/finetune_datasets/parliament_speeches_1998_2016_frp_or_sv/train.csv": {"num_bytes": 6691149, "checksum": "5138fa8521424df0feca43d64a6a75fb1f4d9f2e759cc78027a1dc3bb8bde7ea"}, "https://storage.googleapis.com/notram-public/finetune_datasets/parliament_speeches_1998_2016_frp_or_sv/dev.csv": {"num_bytes": 2399289, "checksum": "b5397e989015c558bc5fb5cc706282b3306a41e09e950559ac0b3eaea8d72dd2"}, "https://storage.googleapis.com/notram-public/finetune_datasets/parliament_speeches_1998_2016_frp_or_sv/test.csv": {"num_bytes": 2365709, "checksum": "d1ebce8715e69884be78713e78d34635c2e6bb67db05c46e5833c2fb433f3825"}}, "download_size": 11456147, "post_processing_size": null, "dataset_size": 11381371, "size_in_bytes": 22837518}}