improve aligner

README.md

@@ -174,6 +174,7 @@ Please refer to [constants.py](src/llmtuner/extras/constants.py) for a full list
 - [CodeAlpaca 20k (en)](https://huggingface.co/datasets/sahil2801/CodeAlpaca-20k)
 - [Alpaca CoT (multilingual)](https://huggingface.co/datasets/QingyiSi/Alpaca-CoT)
 - [OpenOrca (en)](https://huggingface.co/datasets/Open-Orca/OpenOrca)
+- [SlimOrca (en)](https://huggingface.co/datasets/Open-Orca/SlimOrca)
 - [MathInstruct (en)](https://huggingface.co/datasets/TIGER-Lab/MathInstruct)
 - [Firefly 1.1M (zh)](https://huggingface.co/datasets/YeungNLP/firefly-train-1.1M)
 - [Wiki QA (en)](https://huggingface.co/datasets/wiki_qa)

README_zh.md

@@ -174,6 +174,7 @@ https://github.com/hiyouga/LLaMA-Factory/assets/16256802/6ba60acc-e2e2-4bec-b846
 - [CodeAlpaca 20k (en)](https://huggingface.co/datasets/sahil2801/CodeAlpaca-20k)
 - [Alpaca CoT (multilingual)](https://huggingface.co/datasets/QingyiSi/Alpaca-CoT)
 - [OpenOrca (en)](https://huggingface.co/datasets/Open-Orca/OpenOrca)
+- [SlimOrca (en)](https://huggingface.co/datasets/Open-Orca/SlimOrca)
 - [MathInstruct (en)](https://huggingface.co/datasets/TIGER-Lab/MathInstruct)
 - [Firefly 1.1M (zh)](https://huggingface.co/datasets/YeungNLP/firefly-train-1.1M)
 - [Wiki QA (en)](https://huggingface.co/datasets/wiki_qa)

data/README.md

@@ -11,7 +11,7 @@ If you are using a custom dataset, please provide your dataset definition in the
   "folder": "the name of the folder of the dataset repository on the Hugging Face hub. (optional, default: None)",
   "ranking": "whether the dataset is a preference dataset or not. (default: false)",
   "formatting": "the format of the dataset. (optional, default: alpaca, can be chosen from {alpaca, sharegpt})",
-  "columns": {
+  "columns (optional)": {
     "prompt": "the column name in the dataset containing the prompts. (default: instruction)",
     "query": "the column name in the dataset containing the queries. (default: input)",
     "response": "the column name in the dataset containing the responses. (default: output)",
@@ -20,14 +20,14 @@ If you are using a custom dataset, please provide your dataset definition in the
     "system": "the column name in the dataset containing the system prompts. (default: None)",
     "tools": "the column name in the dataset containing the tool description. (default: None)"
   },
-  "tags": {
+  "tags (optional, used for the sharegpt format)": {
     "role_tag": "the key in the message represents the identity. (default: from)",
     "content_tag": "the key in the message represents the content. (default: value)",
     "user_tag": "the value of the role_tag represents the user. (default: human)",
     "assistant_tag": "the value of the role_tag represents the assistant. (default: gpt)",
     "observation_tag": "the value of the role_tag represents the tool results. (default: observation)",
     "function_tag": "the value of the role_tag represents the function call. (default: function_call)",
-    "system_tag": "the value of the role_tag represents the system prompt. (default: None) incompatible with system column"
+    "system_tag": "the value of the role_tag represents the system prompt. (default: system, can override system column)"
   }
 }
 ```

data/README_zh.md

@@ -11,7 +11,7 @@
   "folder": "Hugging Face 仓库的文件夹名称（可选，默认：None）",
   "ranking": "是否为偏好数据集（可选，默认：False）",
   "formatting": "数据集格式（可选，默认：alpaca，可以为 alpaca 或 sharegpt）",
-  "columns": {
+  "columns（可选）": {
     "prompt": "数据集代表提示词的表头名称（默认：instruction）",
     "query": "数据集代表请求的表头名称（默认：input）",
     "response": "数据集代表回答的表头名称（默认：output）",
@@ -20,13 +20,14 @@
     "system": "数据集代表系统提示的表头名称（默认：None）",
     "tools": "数据集代表工具描述的表头名称（默认：None）"
   },
-  "tags": {
+  "tags（可选，用于 sharegpt 格式）": {
     "role_tag": "消息中代表发送者身份的键名（默认：from）",
     "content_tag": "消息中代表文本内容的键名（默认：value）",
     "user_tag": "消息中代表用户的 role_tag（默认：human）",
     "assistant_tag": "消息中代表助手的 role_tag（默认：gpt）",
     "observation_tag": "消息中代表工具返回结果的 role_tag（默认：observation）",
-    "function_tag": "消息中代表工具调用的 role_tag（默认：function_call）"
+    "function_tag": "消息中代表工具调用的 role_tag（默认：function_call）",
+    "system_tag": "消息中代表系统提示的 role_tag（默认：system，会覆盖 system 列）"
   }
 }
 ```

data/dataset_info.json

@@ -15,9 +15,6 @@
     "file_name": "alpaca_gpt4_data_zh.json",
     "file_sha1": "3eaa3bda364ccdd59925d7448a698256c31ef845"
   },
-  "alpaca-gpt4_de": {
-    "hf_hub_url": "mayflowergmbh/alpaca-gpt4_de"
-  },
   "self_cognition": {
     "file_name": "self_cognition.json",
     "file_sha1": "6287a730ada924fc5d9eadc6d8f865e01b7a6f67"
@@ -42,9 +39,6 @@
       "history": "history"
     }
   },
-  "oasst_de": {
-    "hf_hub_url": "mayflowergmbh/oasst_de"
-  },
   "lima": {
     "file_name": "lima.json",
     "file_sha1": "9db59f6b7007dc4b17529fc63379b9cd61640f37",
@@ -126,44 +120,8 @@
       "system": "system_prompt"
     }
   },
- "slimorca": {
+  "slimorca": {
-  "hf_hub_url": "Open-Orca/SlimOrca",
+    "hf_hub_url": "Open-Orca/SlimOrca"
-  "formatting": "sharegpt",
-  "columns": {
-    "messages": "conversations"
-  },
-  "tags": {
-    "role_tag": "from",
-    "content_tag": "value",
-    "user_tag": "human",
-    "assistant_tag": "gpt",
-    "system_tag": "system"
-  }
- },
-  "intel_orca_dpo_pairs_de" : {
-    "hf_hub_url": "mayflowergmbh/intel_orca_dpo_pairs_de",
-    "ranking": true
-  },
-  "airoboros-3.0_de": {
-    "hf_hub_url": "mayflowergmbh/airoboros-3.0_de"
-  },
-  "booksum_de": {
-    "hf_hub_url": "mayflowergmbh/booksum_de"
-  },
-  "dolphin_de": {
-    "hf_hub_url": "mayflowergmbh/dolphin_de"
-  },
-  "wiki_qa_de": {
-    "hf_hub_url": "mayflowergmbh/wiki_qa_de"
-  },
-  "evol-instruct_de": {
-    "hf_hub_url": "mayflowergmbh/evol-instruct_de"
-  },
-  "openschnabeltier_de": {
-    "hf_hub_url": "mayflowergmbh/openschnabeltier_de"
-  },
-  "dolly-15k_de": {
-    "hf_hub_url": "mayflowergmbh/dolly-15k_de"
   },
   "mathinstruct": {
     "hf_hub_url": "TIGER-Lab/MathInstruct",
@@ -180,6 +138,13 @@
       "response": "target"
     }
   },
+  "wikiqa": {
+    "hf_hub_url": "wiki_qa",
+    "columns": {
+      "prompt": "question",
+      "response": "answer"
+    }
+  },
   "webqa": {
     "hf_hub_url": "suolyer/webqa",
     "ms_hub_url": "AI-ModelScope/webqa",
@@ -193,7 +158,8 @@
     "ms_hub_url": "AI-ModelScope/webnovel_cn"
   },
   "nectar_sft": {
-    "hf_hub_url": "mlinmg/SFT-Nectar"
+    "hf_hub_url": "mlinmg/SFT-Nectar",
+    "ms_hub_url": "AI-ModelScope/SFT-Nectar"
   },
   "deepctrl": {
     "ms_hub_url": "deepctrl/deepctrl-sft-data"
@@ -229,9 +195,6 @@
     },
     "formatting": "sharegpt"
   },
-  "ultrachat_chat_de": {
-    "hf_hub_url": "mayflowergmbh/ultra-chat_de"
-  },
   "agent_instruct": {
     "hf_hub_url": "THUDM/AgentInstruct",
     "ms_hub_url": "ZhipuAI/AgentInstruct",
@@ -253,8 +216,36 @@
   },
   "evol_instruct": {
     "hf_hub_url": "WizardLM/WizardLM_evol_instruct_V2_196k",
+    "ms_hub_url": "AI-ModelScope/WizardLM_evol_instruct_V2_196k",
     "formatting": "sharegpt"
   },
+  "oasst_de": {
+    "hf_hub_url": "mayflowergmbh/oasst_de"
+  },
+  "dolly_15k_de": {
+    "hf_hub_url": "mayflowergmbh/dolly-15k_de"
+  },
+  "alpaca-gpt4_de": {
+    "hf_hub_url": "mayflowergmbh/alpaca-gpt4_de"
+  },
+  "openschnabeltier_de": {
+    "hf_hub_url": "mayflowergmbh/openschnabeltier_de"
+  },
+  "evol_instruct_de": {
+    "hf_hub_url": "mayflowergmbh/evol-instruct_de"
+  },
+  "dolphin_de": {
+    "hf_hub_url": "mayflowergmbh/dolphin_de"
+  },
+  "booksum_de": {
+    "hf_hub_url": "mayflowergmbh/booksum_de"
+  },
+  "airoboros_de": {
+    "hf_hub_url": "mayflowergmbh/airoboros-3.0_de"
+  },
+  "ultrachat_de": {
+    "hf_hub_url": "mayflowergmbh/ultra-chat_de"
+  },
   "hh_rlhf_en": {
     "script_url": "hh_rlhf_en",
     "columns": {
@@ -298,6 +289,11 @@
   },
   "nectar_rm": {
     "hf_hub_url": "mlinmg/RLAIF-Nectar",
+    "ms_hub_url": "AI-ModelScope/RLAIF-Nectar",
+    "ranking": true
+  },
+  "orca_dpo_de" : {
+    "hf_hub_url": "mayflowergmbh/intel_orca_dpo_pairs_de",
     "ranking": true
   },
   "wiki_demo": {
@@ -329,6 +325,7 @@
   },
   "wikipedia_en": {
     "hf_hub_url": "olm/olm-wikipedia-20221220",
+    "ms_hub_url": "AI-ModelScope/olm-wikipedia-20221220",
     "columns": {
       "prompt": "text"
     }
@@ -342,6 +339,7 @@
   },
   "pile": {
     "hf_hub_url": "EleutherAI/pile",
+    "ms_hub_url": "AI-ModelScope/pile",
     "columns": {
       "prompt": "text"
     },
@@ -349,6 +347,7 @@
   },
   "skypile": {
     "hf_hub_url": "Skywork/SkyPile-150B",
+    "ms_hub_url": "AI-ModelScope/SkyPile-150B",
     "columns": {
       "prompt": "text"
     }

src/llmtuner/data/aligner.py

@@ -49,40 +49,32 @@ def convert_sharegpt(examples: Dict[str, List[Any]], dataset_attr: "DatasetAttr"
         dataset_attr.function_tag: Role.FUNCTION,
         dataset_attr.system_tag: Role.SYSTEM,
     }
+    odd_tags = (dataset_attr.user_tag, dataset_attr.observation_tag)
+    even_tags = (dataset_attr.assistant_tag, dataset_attr.function_tag)
+    accept_tags = (odd_tags, even_tags)
     for i, messages in enumerate(examples[dataset_attr.messages]):
-        if len(messages) <= 1:
+        if dataset_attr.system_tag and messages[0][dataset_attr.role_tag] == dataset_attr.system_tag:
+            system = messages[0][dataset_attr.content_tag]
+            messages = messages[1:]
+        else:
+            system = examples[dataset_attr.system][i] if dataset_attr.system else ""
+
+        messages = messages[: len(messages) // 2 * 2]  # should be multiples of 2
+        if len(messages) == 0:
             continue
 
-        prompt = []
+        aligned_messages = []
-        response = []
-        n_sys = 0
         for turn_idx, message in enumerate(messages):
-            if dataset_attr.system_tag and message[dataset_attr.role_tag] == dataset_attr.system_tag:
+            if message[dataset_attr.role_tag] not in accept_tags[turn_idx % 2]:
-                outputs["system"].append(message[dataset_attr.content_tag])
-                n_sys = 1
-                continue
-
-            if (turn_idx - n_sys) % 2 == 0:
-                accept_tags = [dataset_attr.user_tag, dataset_attr.observation_tag]
-            else:
-                accept_tags = [dataset_attr.assistant_tag, dataset_attr.function_tag]
-
-            if message[dataset_attr.role_tag] not in accept_tags:
                 raise ValueError("Invalid role tag in {}.".format(messages))
 
-            prompt.append(
+            aligned_messages.append(
                 {"role": tag_mapping[message[dataset_attr.role_tag]], "content": message[dataset_attr.content_tag]}
             )
 
-        if len(prompt) % 2 == 1:
+        outputs["prompt"].append(aligned_messages[:-1])
-            # Last message was neither from assistant nor function
+        outputs["response"].append(aligned_messages[-1:])
-            prompt.pop(-1)
+        outputs["system"].append(system)
-        last_message = prompt.pop(-1)
-        response.append(last_message)
-        outputs["prompt"].append(prompt)
-        outputs["response"].append(response)
-        if n_sys == 0:
-            outputs["system"].append(examples[dataset_attr.system][i] if dataset_attr.system else "")
         outputs["tools"].append(examples[dataset_attr.tools][i] if dataset_attr.tools else "")
 
     return outputs
@@ -93,8 +85,8 @@ def align_dataset(
 ) -> Union["Dataset", "IterableDataset"]:
     r"""
     Aligned dataset:
-        prompt: [{"role": "user", "content": "..."}]
+        prompt: [{"role": "user", "content": "..."}] * (2T - 1)
-        response: [{"role": "assistant", "content": "..."}]
+        response: [{"role": "assistant", "content": "..."}] * N (N > 1 for ranking dataset)
         system: "..."
         tools: "..."
     """

src/llmtuner/data/loader.py

@@ -30,6 +30,7 @@ def load_single_dataset(
     model_args: "ModelArguments",
     data_args: "DataArguments",
 ):
+    logger.info("Loading dataset {}...".format(dataset_attr))
     data_path, data_name, data_dir, data_files = None, None, None, None
     if dataset_attr.load_from in ["hf_hub", "ms_hub"]:
         data_path = dataset_attr.dataset_name
@@ -60,7 +61,7 @@ def load_single_dataset(
         if data_path is None:
             raise ValueError("File extension must be txt, csv, json or jsonl.")
 
-        checksum(data_files, dataset_attr.dataset_sha1)
+        checksum(data_files, dataset_attr.file_sha1)
     else:
         raise NotImplementedError
 
@@ -157,7 +158,7 @@ def get_dataset(
 
     with training_args.main_process_first(desc="load dataset"):
         all_datasets = []
-        for dataset_attr in get_dataset_list(data_args):  # TODO: add split
+        for dataset_attr in get_dataset_list(data_args):
             all_datasets.append(load_single_dataset(dataset_attr, model_args, data_args))
         dataset = merge_dataset(all_datasets, data_args, training_args)
 
@@ -185,6 +186,6 @@ def get_dataset(
             try:
                 print_function(next(iter(dataset)))
             except StopIteration:
-                raise RuntimeError("Empty dataset!")
+                raise RuntimeError("Cannot find valid samples, check `data/README.md` for the data format.")
 
         return dataset

src/llmtuner/data/parser.py

@@ -1,7 +1,7 @@
 import json
 import os
 from dataclasses import dataclass
-from typing import TYPE_CHECKING, List, Literal, Optional
+from typing import TYPE_CHECKING, Any, Dict, List, Literal, Optional
 
 from ..extras.constants import DATA_CONFIG
 from ..extras.misc import use_modelscope
@@ -13,38 +13,44 @@ if TYPE_CHECKING:
 
 @dataclass
 class DatasetAttr:
+    r"""
+    Dataset attributes.
+    """
+
+    """ basic configs """
     load_from: Literal["hf_hub", "ms_hub", "script", "file"]
     dataset_name: Optional[str] = None
-    dataset_sha1: Optional[str] = None
+    """ extra configs """
+    file_sha1: Optional[str] = None
     subset: Optional[str] = None
     folder: Optional[str] = None
     ranking: Optional[bool] = False
     formatting: Optional[Literal["alpaca", "sharegpt"]] = "alpaca"
-
+    """ columns """
     system: Optional[str] = None
-
+    """ columns for the alpaca format """
     prompt: Optional[str] = "instruction"
     query: Optional[str] = "input"
     response: Optional[str] = "output"
     history: Optional[str] = None
-
+    """ columns for the sharegpt format """
     messages: Optional[str] = "conversations"
     tools: Optional[str] = None
-
+    """ tags for the sharegpt format """
     role_tag: Optional[str] = "from"
     content_tag: Optional[str] = "value"
     user_tag: Optional[str] = "human"
     assistant_tag: Optional[str] = "gpt"
     observation_tag: Optional[str] = "observation"
     function_tag: Optional[str] = "function_call"
-    system_tag: Optional[str] = None
+    system_tag: Optional[str] = "system"
-
-    assert system_tag is None or system is None, f"Can not provide both system message (system_tag={system_tag}) and system column(system={system})"
-
 
     def __repr__(self) -> str:
         return self.dataset_name
 
+    def set_attr(self, key: str, obj: Dict[str, Any], default: Optional[Any] = None) -> None:
+        setattr(self, key, obj.get(key, default))
+
 
 def get_dataset_list(data_args: "DataArguments") -> List["DatasetAttr"]:
     dataset_names = [ds.strip() for ds in data_args.dataset.split(",")] if data_args.dataset is not None else []
@@ -77,30 +83,36 @@ def get_dataset_list(data_args: "DataArguments") -> List["DatasetAttr"]:
         elif "script_url" in dataset_info[name]:
             dataset_attr = DatasetAttr("script", dataset_name=dataset_info[name]["script_url"])
         else:
-            dataset_attr = DatasetAttr(
+            dataset_attr = DatasetAttr("file", dataset_name=dataset_info[name]["file_name"])
-                "file",
-                dataset_name=dataset_info[name]["file_name"],
-                dataset_sha1=dataset_info[name].get("file_sha1", None),
-            )
 
-        dataset_attr.subset = dataset_info[name].get("subset", None)
+        dataset_attr.set_attr("file_sha1", dataset_info[name])
-        dataset_attr.folder = dataset_info[name].get("folder", None)
+        dataset_attr.set_attr("subset", dataset_info[name])
-        dataset_attr.ranking = dataset_info[name].get("ranking", False)
+        dataset_attr.set_attr("folder", dataset_info[name])
-        dataset_attr.formatting = dataset_info[name].get("formatting", "alpaca")
+        dataset_attr.set_attr("ranking", dataset_info[name], default=False)
+        dataset_attr.set_attr("formatting", dataset_info[name], default="alpaca")
 
         if "columns" in dataset_info[name]:
+            column_names = ["system"]
             if dataset_attr.formatting == "alpaca":
-                column_names = ["prompt", "query", "response", "history"]
+                column_names.extend(["prompt", "query", "response", "history"])
             else:
-                column_names = ["messages", "tools"]
+                column_names.extend(["messages", "tools"])
 
-            column_names += ["system"]
             for column_name in column_names:
-                setattr(dataset_attr, column_name, dataset_info[name]["columns"].get(column_name, None))
+                dataset_attr.set_attr(column_name, dataset_info[name]["columns"])
 
         if dataset_attr.formatting == "sharegpt" and "tags" in dataset_info[name]:
-            for tag in ["role_tag", "content_tag", "user_tag", "assistant_tag", "observation_tag", "function_tag", "system_tag"]:
+            tag_names = (
-                setattr(dataset_attr, tag, dataset_info[name]["tags"].get(tag, None))
+                "role_tag",
+                "content_tag",
+                "user_tag",
+                "assistant_tag",
+                "observation_tag",
+                "function_tag",
+                "system_tag",
+            )
+            for tag in tag_names:
+                dataset_attr.set_attr(tag, dataset_info[name]["tags"])
 
         dataset_list.append(dataset_attr)
 

src/llmtuner/data/template.py

@@ -247,7 +247,7 @@ def _add_or_replace_eos_token(tokenizer: "PreTrainedTokenizer", eos_token: str)
         logger.info("Replace eos token: {}".format(tokenizer.eos_token))
 
     if is_oov:
-        logger.warning("New token is added, you must enable `resize_vocab` to activate it.")
+        logger.warning("New tokens have been added, make sure `resize_vocab` is True.")
 
 
 def get_template_and_fix_tokenizer(

src/llmtuner/data/utils.py

@@ -19,9 +19,9 @@ logger = get_logger(__name__)
 class Role(str, Enum):
     USER = "user"
     ASSISTANT = "assistant"
+    SYSTEM = "system"
     OBSERVATION = "observation"
     FUNCTION = "function"
-    SYSTEM = "system"
 
 
 def checksum(data_files: List[str], file_sha1: Optional[str] = None) -> None:

src/llmtuner/hparams/parser.py

@@ -67,7 +67,7 @@ def _verify_model_args(model_args: "ModelArguments", finetuning_args: "Finetunin
             raise ValueError("Quantized model only accepts a single adapter. Merge them first.")
 
     if model_args.adapter_name_or_path is not None and finetuning_args.finetuning_type != "lora":
-        raise ValueError("Only LoRA method has adapters.")
+        raise ValueError("Adapter is only valid for the LoRA method.")
 
 
 def _parse_train_args(args: Optional[Dict[str, Any]] = None) -> _TRAIN_CLS:
@@ -125,6 +125,14 @@ def get_train_args(args: Optional[Dict[str, Any]] = None) -> _TRAIN_CLS:
 
     _verify_model_args(model_args, finetuning_args)
 
+    if (
+        training_args.do_train
+        and finetuning_args.finetuning_type == "lora"
+        and model_args.resize_vocab
+        and finetuning_args.additional_target is None
+    ):
+        logger.warning("Add token embeddings to `additional_target` to make the added tokens trainable.")
+
     if training_args.do_train and model_args.quantization_bit is not None and (not model_args.upcast_layernorm):
         logger.warning("We recommend enable `upcast_layernorm` in quantized training.")