423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-10-14 23:58:11 +08:00
Code Issues Packages Projects Releases Wiki Activity

add orca_dpo_pairs dataset

Browse Source 
Former-commit-id: af683aacbae462a2a37d76d37df583e217664bd5
This commit is contained in:
hiyouga 2024-03-20 20:09:06 +08:00
parent 04884a0911
commit 5ed234ca63
4 changed files with 11 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
data/belle_multiturn/belle_multiturn.py
View File

@ -2,6 +2,7 @@ import os
import json import json
import datasets import datasets
_HF_ENDPOINT = os.getenv("HF_ENDPOINT", "https://huggingface.co") _HF_ENDPOINT = os.getenv("HF_ENDPOINT", "https://huggingface.co")
_DESCRIPTION = "BELLE multiturn chat dataset." _DESCRIPTION = "BELLE multiturn chat dataset."
@ -15,9 +16,9 @@ _CITATION = """\
} }
""" """
_HOMEPAGE = f"{_HF_ENDPOINT}/datasets/BelleGroup/multiturn_chat_0.8M" _HOMEPAGE = "{}/datasets/BelleGroup/multiturn_chat_0.8M".format(_HF_ENDPOINT)
_LICENSE = "gpl-3.0" _LICENSE = "gpl-3.0"
_URL = f"{_HF_ENDPOINT}/datasets/BelleGroup/multiturn_chat_0.8M/resolve/main/multiturn_chat_0.8M.json" _URL = "{}/datasets/BelleGroup/multiturn_chat_0.8M/resolve/main/multiturn_chat_0.8M.json".format(_HF_ENDPOINT)
class BelleMultiturn(datasets.GeneratorBasedBuilder): class BelleMultiturn(datasets.GeneratorBasedBuilder):

9
data/hh_rlhf_en/hh_rlhf_en.py
View File

@ -3,15 +3,12 @@ import json
import datasets import datasets
from typing import List from typing import List
_HF_ENDPOINT = os.getenv("_HF_ENDPOINT", "https://huggingface.co") _HF_ENDPOINT = os.getenv("HF_ENDPOINT", "https://huggingface.co")
_DESCRIPTION = "Human preference data about helpfulness and harmlessness." _DESCRIPTION = "Human preference data about helpfulness and harmlessness."
_CITATION = "" _CITATION = ""
_HOMEPAGE = "{}/datasets/Anthropic/hh-rlhf".format(_HF_ENDPOINT)
_HOMEPAGE = f"{_HF_ENDPOINT}/datasets/Anthropic/hh-rlhf"
_LICENSE = "mit" _LICENSE = "mit"
_URL = f"{_HF_ENDPOINT}/datasets/Anthropic/hh-rlhf/resolve/main/" _URL = "{}/datasets/Anthropic/hh-rlhf/resolve/main/".format(_HF_ENDPOINT)
_URLS = { _URLS = {
"train": [ "train": [
_URL + "harmless-base/train.jsonl.gz", _URL + "harmless-base/train.jsonl.gz",

1
data/orca_rlhf.json.REMOVED.git-id Normal file
View File

@ -0,0 +1 @@
736bcedea2b24a1414765c6d69cbdafaea839f3c

8
data/ultra_chat/ultra_chat.py
View File

@ -3,7 +3,7 @@ import json
import datasets import datasets
from typing import List from typing import List
_HF_ENDPOINT = os.getenv("_HF_ENDPOINT", "https://huggingface.co") _HF_ENDPOINT = os.getenv("HF_ENDPOINT", "https://huggingface.co")
_DESCRIPTION = "UltraChat: Large-scale, Informative, and Diverse Multi-round Dialogue Data." _DESCRIPTION = "UltraChat: Large-scale, Informative, and Diverse Multi-round Dialogue Data."
@ -18,9 +18,9 @@ _CITATION = """\
} }
""" """
_HOMEPAGE = f"{_HF_ENDPOINT}/datasets/stingning/ultrachat" _HOMEPAGE = "{}/datasets/stingning/ultrachat".format(_HF_ENDPOINT)
_LICENSE = "cc-by-nc-4.0" _LICENSE = "cc-by-nc-4.0"
_BASE_DATA_URL = "{_HF_ENDPOINT}/datasets/stingning/ultrachat/resolve/main/train_{idx}.jsonl" _BASE_DATA_URL = "{}/datasets/stingning/ultrachat/resolve/main/train_{{idx}}.jsonl".format(_HF_ENDPOINT)
class UltraChat(datasets.GeneratorBasedBuilder): class UltraChat(datasets.GeneratorBasedBuilder):
@ -40,7 +40,7 @@ class UltraChat(datasets.GeneratorBasedBuilder):
) )
def _split_generators(self, dl_manager: datasets.DownloadManager): def _split_generators(self, dl_manager: datasets.DownloadManager):
file_paths = [dl_manager.download(_BASE_DATA_URL.format(_HF_ENDPOINT=_HF_ENDPOINT,idx=idx)) for idx in range(10)] # multiple shards file_paths = [dl_manager.download(_BASE_DATA_URL.format(idx=idx)) for idx in range(10)] # multiple shards
return [ return [
datasets.SplitGenerator( datasets.SplitGenerator(
name=datasets.Split.TRAIN, name=datasets.Split.TRAIN,