Duplicate from AliMaatouk/Tele-Data

Co-authored-by: Ali Maatouk <AliMaatouk@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,62 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.lz4 filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+# Audio files - uncompressed
+*.pcm filter=lfs diff=lfs merge=lfs -text
+*.sam filter=lfs diff=lfs merge=lfs -text
+*.raw filter=lfs diff=lfs merge=lfs -text
+# Audio files - compressed
+*.aac filter=lfs diff=lfs merge=lfs -text
+*.flac filter=lfs diff=lfs merge=lfs -text
+*.mp3 filter=lfs diff=lfs merge=lfs -text
+*.ogg filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text
+# Image files - uncompressed
+*.bmp filter=lfs diff=lfs merge=lfs -text
+*.gif filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.tiff filter=lfs diff=lfs merge=lfs -text
+# Image files - compressed
+*.jpg filter=lfs diff=lfs merge=lfs -text
+*.jpeg filter=lfs diff=lfs merge=lfs -text
+*.webp filter=lfs diff=lfs merge=lfs -text
+# Video files - compressed
+*.mp4 filter=lfs diff=lfs merge=lfs -text
+*.webm filter=lfs diff=lfs merge=lfs -text
+arxiv/arxiv.jsonl filter=lfs diff=lfs merge=lfs -text
+standard/standard.jsonl filter=lfs diff=lfs merge=lfs -text
+web/web.jsonl filter=lfs diff=lfs merge=lfs -text
+wiki/wiki.jsonl filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,97 @@
+---
+license: mit
+language:
+- en
+tags:
+- telecom
+task_categories:
+- text-generation
+configs:
+  - config_name: default
+    data_files:
+      - split: data
+        path: Tele-Eval.jsonl
+---
+
+# Tele-Data
+
+## Dataset Summary
+
+Tele-Data is a comprehensive dataset of telecommunications material that revolves around four categories of sources: (1) scientific papers from arXiv, (2) 3GPP standards, (3) Wikipedia articles related to telecommunications, and (4) telecommunications-related websites extracted from Common Crawl dumps.
+
+LLM-based filtering was used to identify the relevant material from these sources, which then underwent extensive cleaning, format unification, and equation material standardization. The dataset consists of approximately 2.5 billion tokens, making it ideal for continually pretraining language models to adapt them to the telecommunications domain.
+
+
+## Dataset Structure
+
+### Data Fields
+
+The data fields are as follows:
+
+* `ID`: Provides a unique identifier for each data sample.
+* `Category`: Identifies the category of the sample.
+* `Content`: Includes the full text of the data sample.
+* `Metadata`: Includes a JSON object, cast as a string, with information relevant to each data sample, which varies depending on the category.
+
+### Data Instances
+
+An example of Tele-Data looks as follows:
+
+
+```json
+{
+    "ID": "standard_2413",
+    "Category": "standard",
+    "Content": "3rd Generation Partnership Project; \n Technical Specification Group Core Network and Terminals;\n Interworking between the Public Land Mobile Network (PLMN)\n supporting packet based services with\n Wireless Local Area Network (WLAN) Access and\n Packet Data Networks (PDN)\n (Release 12)\n Foreword\n This Technical Specification (TS) has been produced...",
+    "Metadata":
+      "Series": "29",
+      "Release": "12",
+      "File_name": "29161-c00"
+}
+```
+
+## Sample Code
+
+Below, we share a code snippet on how to get quickly started with using the dataset. First, make sure to `pip install datasets`, then copy the snippet below and adapt it to your usecase.
+
+#### Using the whole dataset
+
+```python
+import json
+from datasets import load_dataset
+
+Tele_Data = load_dataset("AliMaatouk/Tele-Data")
+data_sample = Tele_Data['train'][0]
+print(f"ID: {data_sample['id']}\nCategory: {data_sample['category']}  \nContent: {data_sample['content']}")
+for key, value in json.loads(data_sample['metadata']).items():
+	print(f"{key}: {value}")
+```
+
+#### Using a subset of the dataset
+
+```python
+import json
+from datasets import load_dataset
+
+Tele_Data = load_dataset("AliMaatouk/Tele-Data", name="standard")
+data_sample = Tele_Data['train'][0]
+print(f"ID: {data_sample['id']}\nCategory: {data_sample['category']}  \nContent: {data_sample['content']}")
+for key, value in json.loads(data_sample['metadata']).items():
+	print(f"{key}: {value}")
+```
+
+## Citation
+
+You can find the paper with all details about the dataset at https://arxiv.org/abs/2409.05314. Please cite it as follows:
+
+```
+@misc{maatouk2024telellmsseriesspecializedlarge,
+      title={Tele-LLMs: A Series of Specialized Large Language Models for Telecommunications}, 
+      author={Ali Maatouk and Kenny Chirino Ampudia and Rex Ying and Leandros Tassiulas},
+      year={2024},
+      eprint={2409.05314},
+      archivePrefix={arXiv},
+      primaryClass={cs.IT},
+      url={https://arxiv.org/abs/2409.05314}, 
+}
+```

Tele-Data.py

@@ -0,0 +1,51 @@
+import json
+import datasets
+
+class TeleData(datasets.GeneratorBasedBuilder):
+    """Tele-Data dataset with multiple subsets: arxiv, standard, web, and wiki"""
+
+    BUILDER_CONFIGS = [
+        datasets.BuilderConfig(name="arxiv", version=datasets.Version("1.0.0"), description="ArXiv data"),
+        datasets.BuilderConfig(name="standard", version=datasets.Version("1.0.0"), description="Standard data"),
+        datasets.BuilderConfig(name="web", version=datasets.Version("1.0.0"), description="Web data"),
+        datasets.BuilderConfig(name="wiki", version=datasets.Version("1.0.0"), description="Wiki data"),
+        datasets.BuilderConfig(name="full", version=datasets.Version("1.0.0"), description="Full dataset"),
+    ]
+
+    DEFAULT_CONFIG_NAME = "full"
+
+    def _info(self):
+        features = datasets.Features({
+            "id": datasets.Value("string"),
+            "category": datasets.Value("string"),
+            "content": datasets.Value("string"),
+            "metadata": datasets.Value("string"), 
+        })
+        return datasets.DatasetInfo(features=features)
+
+    def _split_generators(self, dl_manager):
+        if self.config.name == "full":
+            urls = [f"{name}/{name}.jsonl" for name in ["arxiv", "standard", "web", "wiki"]]
+        else:
+            urls = [f"{self.config.name}/{self.config.name}.jsonl"]
+
+        data_files = dl_manager.download_and_extract(urls)
+        
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={"filepaths": data_files if isinstance(data_files, list) else [data_files]},
+            )
+        ]
+
+    def _generate_examples(self, filepaths):
+        for filepath in filepaths:
+            with open(filepath, "r", encoding="utf-8") as f:
+                for id_, line in enumerate(f):
+                    data = json.loads(line)
+                    yield id_, {
+                        "id": data["id"],
+                        "category": data["category"],
+                        "content": data["content"],
+                        "metadata": json.dumps(data["metadata"]), 
+                    }

arxiv/arxiv.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f2d03a5cc1d8b0a13964d1a47e2456cd19813beddd8711e47d2e24102a824e7
+size 4309790362

standard/standard.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9560c7a62612c4724e986fc7443019d886a158a961ddc4c3e521a6650486fa9
+size 350404101

web/web.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a11dede469b0f5568298a8172ca451382c130e1e8db215a5158d3681598759e
+size 7266835533

wiki/wiki.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a5bb8acd97ad37a884a845a705206980039d337e49ab3ea5c278925b2e2977e
+size 129518949