更新依赖，重构代码以使用LLaMA-Factory替代Llmtuner，更新文档以反映新环境配置

.gitignore

@@ -155,3 +155,4 @@ test.*
 WeClone-audio/src/output*.wav
 WeClone-audio/uv.lock
 Spark-TTS-0.5B/
+uv.lock

README.md

@@ -93,6 +93,7 @@ export USE_MODELSCOPE_HUB=1 # Windows 使用 `set USE_MODELSCOPE_HUB=1`
 git lfs install
 git clone https://www.modelscope.cn/ZhipuAI/chatglm3-6b.git
 ```
+魔搭社区的`modeling_chatglm.py`文件需要更换为Hugging Face的
 
 ### 配置参数并微调模型
 

WeClone-audio/README.md

@@ -38,6 +38,19 @@ python ./WeClone-audio/get_sample_audio.py --db-path "导出数据库路径" --M
 
 ## 2. 语音合成推理
 ### Spark-TTS模型
+
+**环境安装**
+可不创建新环境，直接安装依赖组到WeClone共主环境
+
+```bash
+uv venv .venv-sparktts --python=3.9
+source .venv-sparktts/bin/activate
+uv pip install -e '.[sparktts]'
+
+cd WeClone-audio/src
+git clone https://github.com/SparkAudio/Spark-TTS.git
+```
+
 **模型下载**
 
 通过python下载:

WeClone-audio/src/infer.py

@@ -15,3 +15,4 @@ with torch.no_grad():
         prompt_text="对，这就是我万人敬仰的太乙真人，虽然有点婴儿肥，但也掩不住我逼人的帅气。",
     )
     sf.write(os.path.join(os.path.dirname(__file__), "output.wav"), wav, samplerate=16000)
+    print("生成成功！")

pyproject.toml

@@ -8,7 +8,8 @@ authors = [
 readme = "README.md"
 requires-python = ">=3.9,<3.10"
 dependencies = [
-    "llmtuner>=0.7.1",
+    "transformers==4.46.2",
+    "llamafactory>=0.9.2",
     "pandas",
     "openai==0.28.0",
 ]
@@ -16,16 +17,45 @@ dependencies = [
 [project.optional-dependencies]
 xcodec = ["xcodec2==0.1.3"]
 wx = ["pywxdump"]
+sparktts = [
+    "einops>=0.8.1",
+    "einx>=0.3.0",
+    "numpy==1.26.4",
+    "omegaconf>=2.3.0",
+    "packaging>=24.2",
+    "safetensors>=0.5.2",
+    "soundfile>=0.12.1",
+    "soxr>=0.5.0.post1",
+    "torch>=2.5.1",
+    "torchaudio>=2.5.1",
+    "tqdm>=4.66.5",
+    "transformers==4.46.2"
+]
 
 [dependency-groups]
 xcodec = ["xcodec2==0.1.3"]
 wx = ["pywxdump"]
+sparktts = [
+    "einops>=0.8.1",
+    "einx>=0.3.0",
+    "numpy==1.26.4",
+    "omegaconf>=2.3.0",
+    "packaging>=24.2",
+    "safetensors>=0.5.2",
+    "soundfile>=0.12.1",
+    "soxr>=0.5.0.post1",
+    "torch>=2.5.1",
+    "torchaudio>=2.5.1",
+    "tqdm>=4.66.5",
+    "transformers==4.46.2"
+]
 
 [tool.uv]
 conflicts = [
     [
       { group = "xcodec" },
       { group = "wx" },
+      { group = "sparktts" },
     ],
 ]
 

settings.json

@@ -46,7 +46,8 @@
         "model_name_or_path": "./chatglm3-6b",
         "adapter_name_or_path": "./model_output",
         "template": "chatglm3-weclone",
-        "finetuning_type": "lora"
+        "finetuning_type": "lora",
+        "trust_remote_code": true
     },
     "_comment": "adapter_name_or_path同时做为train_sft_args的output_dir "
 }

src/api_service.py

@@ -1,7 +1,7 @@
 import os
 import uvicorn
-from llmtuner.chat import ChatModel
-from llmtuner.api.app import create_app
+from llamafactory.chat import ChatModel
+from llamafactory.api.app import create_app
 from template import template_register
 from utils.config import load_config
 

src/cli_demo.py

@@ -1,5 +1,5 @@
-from llmtuner.chat import ChatModel
-from llmtuner.extras.misc import torch_gc
+from llamafactory.chat import ChatModel
+from llamafactory.extras.misc import torch_gc
 
 
 try:

src/evaluate.py

@@ -1,4 +1,4 @@
-from llmtuner.eval.evaluator import Evaluator
+from llamafactory.eval.evaluator import Evaluator
 
 
 def main():

src/export_model.py

@@ -1,4 +1,4 @@
-from llmtuner.train.tuner import export_model
+from llamafactory.train.tuner import export_model
 
 
 def main():

src/template.py

@@ -1,21 +1,24 @@
-from llmtuner.data.formatter import FunctionFormatter, StringFormatter
-from llmtuner.data.template import _register_template
+from llamafactory.data.formatter import FunctionFormatter, StringFormatter, ToolFormatter, EmptyFormatter
+from llamafactory.data.template import register_template
 
 default_prompt = "请你扮演一名人类，不要说自己是人工智能"
 
 
 def template_register():
-    _register_template(
+    register_template(
         name="chatglm3-weclone",
         default_system=(
             default_prompt
         ),
         format_user=StringFormatter(slots=[{"token": "<|user|>"}, "\n", "{{content}}", {"token": "<|assistant|>"}]),
         format_assistant=StringFormatter(slots=["\n", "{{content}}"]),
-        format_system=StringFormatter(slots=[{"token": "[gMASK]"}, {"token": "sop"}, {"token": "<|system|>"}, "\n", "{{content}}"]),
-        format_function=FunctionFormatter(slots=["{{name}}\n{{arguments}}"]),
-        format_observation=StringFormatter(slots=[{"token": "<|observation|>"}, "\n", "{{content}}"]),
+        format_system=StringFormatter(slots=[{"token": "<|system|>"}, "\n", "{{content}}"]),
+        format_function=FunctionFormatter(slots=["{{content}}"], tool_format="glm4"),
+        format_observation=StringFormatter(
+            slots=[{"token": "<|observation|>"}, "\n", "{{content}}", {"token": "<|assistant|>"}]
+        ),
+        format_tools=ToolFormatter(tool_format="glm4"),
+        format_prefix=EmptyFormatter(slots=[{"token": "[gMASK]"}, {"token": "sop"}]),
         stop_words=["<|user|>", "<|observation|>"],
         efficient_eos=True,
-        force_system=True
     )

src/train_pt.py

@@ -1,4 +1,4 @@
-from llmtuner.train.tuner import run_exp
+from llamafactory.train.tuner import run_exp
 from utils.config import load_config
 
 config = load_config('train_pt')

src/train_sft.py

@@ -1,4 +1,4 @@
-from llmtuner.train.tuner import run_exp
+from llamafactory.train.tuner import run_exp
 from template import template_register
 from utils.config import load_config
 

src/web_demo.py

@@ -1,4 +1,4 @@
-from llmtuner.webui.interface import create_web_demo
+from llamafactory.webui.interface import create_web_demo
 from template import template_register
 from utils.config import load_config