add data
Browse files- cs-en/test.jsonl +0 -0
- cs-en/train.jsonl +0 -0
- cs-en/validation.jsonl +0 -0
- data/test.json +0 -0
- data/validation.json +0 -0
- de-en/test.jsonl +0 -0
- de-en/train.jsonl +0 -0
- de-en/validation.jsonl +0 -0
- fr-en/test.jsonl +0 -0
- data/train.json → fr-en/train.jsonl +0 -0
- fr-en/validation.jsonl +0 -0
- hi-en/test.jsonl +0 -0
- hi-en/train.jsonl +0 -0
- hi-en/validation.jsonl +0 -0
- ru-en/test.jsonl +0 -0
- ru-en/train.jsonl +0 -0
- ru-en/validation.jsonl +0 -0
- wmt_14.py +77 -0
cs-en/test.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
cs-en/train.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
cs-en/validation.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
data/test.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
data/validation.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
de-en/test.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
de-en/train.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
de-en/validation.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
fr-en/test.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
data/train.json → fr-en/train.jsonl
RENAMED
|
The diff for this file is too large to render.
See raw diff
|
|
|
fr-en/validation.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
hi-en/test.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
hi-en/train.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
hi-en/validation.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ru-en/test.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ru-en/train.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ru-en/validation.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
wmt_14.py
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import datasets
|
| 2 |
+
import os
|
| 3 |
+
import json
|
| 4 |
+
|
| 5 |
+
|
| 6 |
+
_CITATION = ""
|
| 7 |
+
_DESCRIPTION = """
|
| 8 |
+
The 2014 Workshop on Statistical Machine Translation:
|
| 9 |
+
https://aclanthology.org/W14-3302.pdf
|
| 10 |
+
|
| 11 |
+
The scenario consists of 5 subsets, each of which is a parallel corpus between English and another language. The
|
| 12 |
+
non-English languages include Czech, German, French, Hindi, and Russian.
|
| 13 |
+
|
| 14 |
+
For each language pair, the validation and test set each includes around 3,000 examples, while the training set is
|
| 15 |
+
usually much larger. We therefore randomly downsample the training set to speedup data processing.
|
| 16 |
+
|
| 17 |
+
Task prompt structure:
|
| 18 |
+
|
| 19 |
+
Translate {source_language} to {target_language}:
|
| 20 |
+
{Hypothesis} = {Reference}
|
| 21 |
+
|
| 22 |
+
Example from WMT14 Fr-En:
|
| 23 |
+
|
| 24 |
+
Hypothesis: Assemblée générale
|
| 25 |
+
Reference: General Assembly
|
| 26 |
+
"""
|
| 27 |
+
|
| 28 |
+
class Summarization(datasets.GeneratorBasedBuilder):
|
| 29 |
+
VERSION = datasets.Version("1.0.0")
|
| 30 |
+
|
| 31 |
+
BUILDER_CONFIGS = [
|
| 32 |
+
datasets.BuilderConfig(name=name, version=datasets.Version("1.0.0"), description="")
|
| 33 |
+
for name in ["cs-en", "de-en", "fr-en", "hi-en", "ru-en", "en-en"]
|
| 34 |
+
]
|
| 35 |
+
|
| 36 |
+
def _info(self):
|
| 37 |
+
source_language, target_language = self.config.name.split('-')
|
| 38 |
+
features = datasets.Features(
|
| 39 |
+
{
|
| 40 |
+
source_language: datasets.Value("string"),
|
| 41 |
+
target_language: datasets.Value("string"),
|
| 42 |
+
|
| 43 |
+
}
|
| 44 |
+
)
|
| 45 |
+
return datasets.DatasetInfo(
|
| 46 |
+
description=_DESCRIPTION,
|
| 47 |
+
features=features,
|
| 48 |
+
homepage="",
|
| 49 |
+
license="",
|
| 50 |
+
citation=_CITATION,
|
| 51 |
+
)
|
| 52 |
+
|
| 53 |
+
def _split_generators(self, dl_manager):
|
| 54 |
+
train_json = dl_manager.download(os.path.join(self.config.name, "train.jsonl"))
|
| 55 |
+
test_json = dl_manager.download(os.path.join(self.config.name, "test.jsonl"))
|
| 56 |
+
val_json = dl_manager.download(os.path.join(self.config.name, "validation.jsonl"))
|
| 57 |
+
|
| 58 |
+
return [
|
| 59 |
+
datasets.SplitGenerator(
|
| 60 |
+
name=datasets.Split.TRAIN,
|
| 61 |
+
gen_kwargs={"path": train_json},
|
| 62 |
+
),
|
| 63 |
+
datasets.SplitGenerator(
|
| 64 |
+
name=datasets.Split.TEST,
|
| 65 |
+
gen_kwargs={"path": test_json},
|
| 66 |
+
),
|
| 67 |
+
datasets.SplitGenerator(
|
| 68 |
+
name=datasets.Split.VALIDATION,
|
| 69 |
+
gen_kwargs={"path": val_json},
|
| 70 |
+
)
|
| 71 |
+
]
|
| 72 |
+
|
| 73 |
+
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
| 74 |
+
def _generate_examples(self, path):
|
| 75 |
+
with open(path, encoding="utf-8") as f:
|
| 76 |
+
for key, row in enumerate(f):
|
| 77 |
+
yield key, json.loads(row)
|