Adding new translations for nob and nno

nob/nob_test.json.tar.gz → nno/nno_test.json.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e53b90d8579baebfb94b1cfea5695a75633d71c38bac43ad3d5c06586e2204e7
-size 3363690
+oid sha256:c3d4e9848cf1c5b52a9b7c8b3e4c54ac8a4cf83b76e7d90297aae8ae17ece42b
+size 9099898

nno/nno_train.json.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcebec789f9f7af7a149bedd0b49910115ee12dc053186fbc710dc1f0245abab
+size 162881864

nob/nob_validation.json.tar.gz → nno/nno_validation.json.gz

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:51ecfd130e61fb1860448427b83a18a64ca6f779f1785517366faeb8d5d790de
-size 3280742
+oid sha256:95bc60b000790dc4eab2b39cd95ec7d64074b06b148f94a32812aa3efc0ab05d
+size 8955006

nob/nob_train.json.tar.gz → nno/test/data-00000-of-00001.arrow

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c533b8c8596718c7f3954c0578562efa314e141b3880cb43145d6c7e3ef11053
-size 60171284
+oid sha256:2271215611a203d8cafac0af05c964e1b597348d396326ee8dd340b167d68f1b
+size 22634776

nno/test/dataset_info.json

@@ -0,0 +1,65 @@
+{
+  "builder_name": "xsum",
+  "citation": "\n@article{Narayan2018DontGM,\n  title={Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization},\n  author={Shashi Narayan and Shay B. Cohen and Mirella Lapata},\n  journal={ArXiv},\n  year={2018},\n  volume={abs/1808.08745}\n}\n",
+  "config_name": "default",
+  "dataset_size": 532255381,
+  "description": "\nExtreme Summarization (XSum) Dataset.\n\nThere are three features:\n  - document: Input news article.\n  - summary: One sentence summary of the article.\n  - id: BBC ID of the article.\n\n",
+  "download_checksums": {
+    "data/XSUM-EMNLP18-Summary-Data-Original.tar.gz": {
+      "num_bytes": 254582292,
+      "checksum": "10b48aa187fc9c904b30f76ca97e2da0de8d3a1238acc26acadef93e2001af90"
+    },
+    "https://raw.githubusercontent.com/EdinburghNLP/XSum/master/XSum-Dataset/XSum-TRAINING-DEV-TEST-SPLIT-90-5-5.json": {
+      "num_bytes": 2720574,
+      "checksum": "9c0c5d8f048a90bd68b19a34e4c30577ed270d3247b2119fa06a04ef46292068"
+    }
+  },
+  "download_size": 257302866,
+  "features": {
+    "document": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "summary": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "id": {
+      "dtype": "string",
+      "_type": "Value"
+    }
+  },
+  "homepage": "https://github.com/EdinburghNLP/XSum/tree/master/XSum-Dataset",
+  "license": "",
+  "size_in_bytes": 789558247,
+  "splits": {
+    "train": {
+      "name": "train",
+      "num_bytes": 479206363,
+      "num_examples": 204045,
+      "dataset_name": "xsum"
+    },
+    "validation": {
+      "name": "validation",
+      "num_bytes": 26292877,
+      "num_examples": 11332,
+      "dataset_name": "xsum"
+    },
+    "test": {
+      "name": "test",
+      "num_bytes": 26756141,
+      "num_examples": 11334,
+      "dataset_name": "xsum"
+    }
+  },
+  "supervised_keys": {
+    "input": "document",
+    "output": "summary"
+  },
+  "version": {
+    "version_str": "1.2.0",
+    "major": 1,
+    "minor": 2,
+    "patch": 0
+  }
+}

nno/test/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "ca9883b8f61f3596",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": "test"
+}

nno/train/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8cfe1eb2e1298cdb75e04c839ed09daa2b91451573cd8f6e3c8b3df8fd81a65c
+size 405633912

nno/train/dataset_info.json

@@ -0,0 +1,65 @@
+{
+  "builder_name": "xsum",
+  "citation": "\n@article{Narayan2018DontGM,\n  title={Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization},\n  author={Shashi Narayan and Shay B. Cohen and Mirella Lapata},\n  journal={ArXiv},\n  year={2018},\n  volume={abs/1808.08745}\n}\n",
+  "config_name": "default",
+  "dataset_size": 532255381,
+  "description": "\nExtreme Summarization (XSum) Dataset.\n\nThere are three features:\n  - document: Input news article.\n  - summary: One sentence summary of the article.\n  - id: BBC ID of the article.\n\n",
+  "download_checksums": {
+    "data/XSUM-EMNLP18-Summary-Data-Original.tar.gz": {
+      "num_bytes": 254582292,
+      "checksum": "10b48aa187fc9c904b30f76ca97e2da0de8d3a1238acc26acadef93e2001af90"
+    },
+    "https://raw.githubusercontent.com/EdinburghNLP/XSum/master/XSum-Dataset/XSum-TRAINING-DEV-TEST-SPLIT-90-5-5.json": {
+      "num_bytes": 2720574,
+      "checksum": "9c0c5d8f048a90bd68b19a34e4c30577ed270d3247b2119fa06a04ef46292068"
+    }
+  },
+  "download_size": 257302866,
+  "features": {
+    "document": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "summary": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "id": {
+      "dtype": "string",
+      "_type": "Value"
+    }
+  },
+  "homepage": "https://github.com/EdinburghNLP/XSum/tree/master/XSum-Dataset",
+  "license": "",
+  "size_in_bytes": 789558247,
+  "splits": {
+    "train": {
+      "name": "train",
+      "num_bytes": 479206363,
+      "num_examples": 204045,
+      "dataset_name": "xsum"
+    },
+    "validation": {
+      "name": "validation",
+      "num_bytes": 26292877,
+      "num_examples": 11332,
+      "dataset_name": "xsum"
+    },
+    "test": {
+      "name": "test",
+      "num_bytes": 26756141,
+      "num_examples": 11334,
+      "dataset_name": "xsum"
+    }
+  },
+  "supervised_keys": {
+    "input": "document",
+    "output": "summary"
+  },
+  "version": {
+    "version_str": "1.2.0",
+    "major": 1,
+    "minor": 2,
+    "patch": 0
+  }
+}

nno/train/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "5bac35fb99243876",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": "train"
+}

nno/validation/data-00000-of-00001.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b46d5ad6709969f1079b17dd317645af0ec84f00fcf7bb426be678b596cf4eee
+size 22308800

nno/validation/dataset_info.json

@@ -0,0 +1,65 @@
+{
+  "builder_name": "xsum",
+  "citation": "\n@article{Narayan2018DontGM,\n  title={Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization},\n  author={Shashi Narayan and Shay B. Cohen and Mirella Lapata},\n  journal={ArXiv},\n  year={2018},\n  volume={abs/1808.08745}\n}\n",
+  "config_name": "default",
+  "dataset_size": 532255381,
+  "description": "\nExtreme Summarization (XSum) Dataset.\n\nThere are three features:\n  - document: Input news article.\n  - summary: One sentence summary of the article.\n  - id: BBC ID of the article.\n\n",
+  "download_checksums": {
+    "data/XSUM-EMNLP18-Summary-Data-Original.tar.gz": {
+      "num_bytes": 254582292,
+      "checksum": "10b48aa187fc9c904b30f76ca97e2da0de8d3a1238acc26acadef93e2001af90"
+    },
+    "https://raw.githubusercontent.com/EdinburghNLP/XSum/master/XSum-Dataset/XSum-TRAINING-DEV-TEST-SPLIT-90-5-5.json": {
+      "num_bytes": 2720574,
+      "checksum": "9c0c5d8f048a90bd68b19a34e4c30577ed270d3247b2119fa06a04ef46292068"
+    }
+  },
+  "download_size": 257302866,
+  "features": {
+    "document": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "summary": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "id": {
+      "dtype": "string",
+      "_type": "Value"
+    }
+  },
+  "homepage": "https://github.com/EdinburghNLP/XSum/tree/master/XSum-Dataset",
+  "license": "",
+  "size_in_bytes": 789558247,
+  "splits": {
+    "train": {
+      "name": "train",
+      "num_bytes": 479206363,
+      "num_examples": 204045,
+      "dataset_name": "xsum"
+    },
+    "validation": {
+      "name": "validation",
+      "num_bytes": 26292877,
+      "num_examples": 11332,
+      "dataset_name": "xsum"
+    },
+    "test": {
+      "name": "test",
+      "num_bytes": 26756141,
+      "num_examples": 11334,
+      "dataset_name": "xsum"
+    }
+  },
+  "supervised_keys": {
+    "input": "document",
+    "output": "summary"
+  },
+  "version": {
+    "version_str": "1.2.0",
+    "major": 1,
+    "minor": 2,
+    "patch": 0
+  }
+}

nno/validation/state.json

@@ -0,0 +1,13 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00001.arrow"
+    }
+  ],
+  "_fingerprint": "a6c220ea2a1f3000",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": "validation"
+}

nob/dataset_dict.json

@@ -1 +0,0 @@
-{"splits": ["train", "validation", "test"]}

nob/nob_test.json.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0c323dc8b9a5bee9ed424842c3e8b02c889193ecad76aebe3655a0a4f13e8af
+size 9515838

nob/nob_train.json.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:923dbc055f3874dd98a89c7e9b22d48bb773af5caa8e9325f01fd410889c0989
+size 170294159

nob/nob_validation.json.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de17f72b1e7fdfb611e8aa636f83e1f36b6eea384bf8996a48bb4fa0e9c58b52
+size 9359670

nob/test/data-00000-of-00001.arrow

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b6b9eb616116b75a8c23eabde80b9ab21227c7f3f09738e9461548a13670c06
-size 48076152
+oid sha256:786585c00c182a27fe083d8956a81daf0c61a688cf1de8235b38bd9dc3bba153
+size 23875496

nob/test/state.json

@@ -4,7 +4,7 @@
       "filename": "data-00000-of-00001.arrow"
     }
   ],
-  "_fingerprint": "0cd0deb949ec246b",
+  "_fingerprint": "5cce8fd2208d7170",
   "_format_columns": null,
   "_format_kwargs": {},
   "_format_type": null,

nob/train/data-00000-of-00001.arrow

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e799e9bdddcdf2997088f16298df4614f8adb6068f634d52b576b11a66d3e04
-size 833308256
+oid sha256:67a00f2331a3afe0a984c8939c25a688fc92467ac86363cee5fadf402ac0f4d4
+size 427840440

nob/train/state.json

@@ -4,7 +4,7 @@
       "filename": "data-00000-of-00001.arrow"
     }
   ],
-  "_fingerprint": "09e18f5a398d83ac",
+  "_fingerprint": "b9a9c9cc7aef8975",
   "_format_columns": null,
   "_format_kwargs": {},
   "_format_type": null,

nob/validation/data-00000-of-00001.arrow

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0dccddfe9fc0209ae45e4ccf9279b0330b8a480dd5d1819a2ba8136e3695a320
-size 44194752
+oid sha256:1c182cfd9445cd9c6c35fbbb5cacd046a8f101d9d988983681da4e7a7a13a35a
+size 23504448

nob/validation/state.json

@@ -4,7 +4,7 @@
       "filename": "data-00000-of-00001.arrow"
     }
   ],
-  "_fingerprint": "3d58f12b55350560",
+  "_fingerprint": "c42300ebe4b92c7b",
   "_format_columns": null,
   "_format_kwargs": {},
   "_format_type": null,

translator.py

@@ -12,58 +12,80 @@ from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
 DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
 
-def to_lang_code(texts, lang_code, model, tokenizer, max_words=500):
+
+def split_into_chunks(text, tokenizer, max_tokens=128):
+    # Split tokenized text into sentences
+    sentences = nltk.sent_tokenize(text)
+
+    # Create chunks based on the maximum number of tokens
+    chunks = []
+    current_chunk = []
+    tokens_count = 0
+
+    for sentence in sentences:
+        sentence_tokens = tokenizer.encode(sentence, add_special_tokens=False)
+        sentence_token_count = len(sentence_tokens)
+
+        if tokens_count + sentence_token_count > max_tokens:
+            # If adding this sentence to the current chunk would exceed the maximum number of tokens, add the current chunk to the list of chunks
+            if current_chunk:
+                chunk_text = tokenizer.decode(current_chunk)
+                chunks.append(chunk_text)
+                current_chunk = []
+                tokens_count = 0
+
+        # Add the sentence to the current chunk
+        current_chunk.extend(sentence_tokens)
+        tokens_count += sentence_token_count
+
+    # Add any remaining tokens as the last chunk
+    if current_chunk:
+        chunk_text = tokenizer.decode(current_chunk)
+        chunks.append(chunk_text)
+
+    return chunks
+
+
+def to_lang_code(texts, lang_code, model, tokenizer, max_tokens=128):
     is_string = isinstance(texts, str)
     if is_string:
         texts = [texts]
     batch_size = len(texts)
     to_translate = []
-    merges = []
-    for index, text in enumerate(texts):
-        # Split in sentences if too long
-        merges.append(0)
-        if text.count(" ") > max_words:
-            sentences = nltk.sent_tokenize(text, "norwegian")
-            text_to_translate = ""
-            for sentence in sentences:
-                spaces = (text_to_translate + " " + sentence).count(" ")
-                if spaces >= max_words:
-                    to_translate.append(text_to_translate.strip())
-                    merges[-1] += 1
-                else:
-                    text_to_translate += sentence + " "
-        else:
-            to_translate.append(text)
+    lengths = []
+    for text in texts:
+        # Split in chunks of non-breaking sentences and keep lengths of chunks
+        chunks = split_into_chunks(text, tokenizer=tokenizer, max_tokens=max_tokens)
+        lengths.append(len(chunks))
+        to_translate += chunks
     translated_texts = []
     # Split in batches for translation
-    to_translate_batchs = [to_translate[i:i + batch_size] for i in range(0, len(to_translate), batch_size)]
-    for to_translate_batch in to_translate_batchs:
+    to_translate_batches = [to_translate[i:i + batch_size] for i in range(0, len(to_translate), batch_size)]
+    for to_translate_batch in to_translate_batches:
         inputs = tokenizer(to_translate_batch, return_tensors="pt", padding=True, truncation=True).to(DEVICE)
         translated_tokens = model.generate(
             **inputs,
             forced_bos_token_id=tokenizer.lang_code_to_id[lang_code],
-            max_length=int(len(inputs.tokens()) * 1.25)  # 25% more tokens for the translation just in case
+            max_new_tokens=512,
+            # max_length=int(len(inputs.tokens()) * 1.25)  # 25% more tokens for the translation just in case
         )
-        translated_texts += tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)
+        translated_text = tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)
+        translated_texts += translated_text
     # Merge outputs properly
     outputs = []
-    for merge in merges:
-        output = ""
-        if merge:
-            for i in range(len(outputs), len(outputs) + merge):
-                output += translated_texts[i] + " "
-            outputs.append(output.strip())
-        else:
-            outputs.append(translated_texts[len(outputs)].strip())
+    start = 0
+    for length in lengths:
+        outputs.append(" ".join(translated_texts[start:start + length]))
+        start += length
     return outputs[0] if is_string else outputs
 
 
 def main(
     dataset_name: str,
     dataset_columns: Union[list, tuple],
-    model_name: Optional[str]="facebook/nllb-200-3.3B",  # "facebook/nllb-200-distilled-600M"
+    model_name: Optional[str]="facebook/nllb-200-1.3B",  # "facebook/nllb-200-distilled-600M"
     model_revision: Optional[str]=None,
-    dataset_splits: Union[list, tuple]=("test", "validation", "train"),
+    dataset_splits: Union[list, tuple]=("train", "validation", "test"),
     dataset_config: Optional[str]=None,
     dataset_revision: Optional[str]=None,
     source_lang: Optional[str]="eng_Latn",
@@ -78,21 +100,24 @@ def main(
         model_name, revision=model_revision, use_auth_token=True, src_lang=source_lang,
     )
 
-    ds = load_dataset(dataset_name, name=dataset_config, revision=dataset_revision)
-    dss = {}
     for lang_code in target_langs:
-        translate = partial(to_lang_code, lang_code=lang_code, model=model, tokenizer=tokenizer)
-        dss[lang_code] = ds.map(
-            lambda batch: {col: translate(batch[col]) for col in dataset_columns},
-            batched=True,
-            batch_size=batch_size,
-            desc=f"Translating to {lang_code}",
-        )
-        lang_code_short = re.split(r"[-_ /]", lang_code)[0]
-        dss[lang_code].save_to_disk(output_dir / lang_code_short, max_shard_size="1GB")
         for split in dataset_splits:
-            json_filename = f"{lang_code_short}_{split}.json.tar.gz".lower()
-            dss[lang_code][split].to_pandas().to_json(
+            ds = load_dataset(dataset_name, name=dataset_config, revision=dataset_revision, split=split)
+            translate = partial(to_lang_code, lang_code=lang_code, model=model, tokenizer=tokenizer)
+            ds = ds.map(
+                lambda batch: {
+                    column: translate(batch[column])
+                    for column in dataset_columns
+                },
+                batched=True,
+                batch_size=batch_size,
+                desc=f"Translating to {lang_code} ({split})",
+            )
+            lang_code_short = re.split(r"[-_ /]", lang_code)[0]
+            ds.save_to_disk(output_dir / lang_code_short / split, max_shard_size="1GB")
+
+            json_filename = f"{lang_code_short}_{split}.json.gz".lower()
+            ds.to_pandas().to_json(
                 output_dir / lang_code_short / json_filename, orient='records', lines=True
             )
 
@@ -102,10 +127,10 @@ if __name__ == "__main__":
     parser = argparse.ArgumentParser(description="Translate datasets using Facebook's NLLB models")
     parser.add_argument('dataset_name')
     parser.add_argument('dataset_columns', help="Comma separated column names to translate")
-    parser.add_argument('--dataset_splits', default="test,validation,train", help="Comma separated splits to translate")
+    parser.add_argument('--dataset_splits', default="train,validation,test", help="Comma separated splits to translate")
     parser.add_argument('--dataset_config')
     parser.add_argument('--dataset_revision')
-    parser.add_argument('--model_name', default="facebook/nllb-200-3.3B")
+    parser.add_argument('--model_name', default="facebook/nllb-200-1.3B")
     parser.add_argument('--model_revision')
     parser.add_argument('--source_lang', default="eng_Latn")
     parser.add_argument('--target_langs', default="nob_Latn,nno_Latn", help="Comma separated target languages to translate to")