Fix readme & update docs (#3018)

Author: Jintao-Huang

README.md

@@ -79,7 +79,7 @@ You can contact us and communicate with us by adding our group:
 
 ## 🎉 News
 
-- 🎁 2024.01.23: SWIFT support the `sample` command, this is a very important feature for complex CoT and RFT. Meanwhile, we support an [Reinforced Fine-tuning script](docs/source_en/Instruction/Reinforced_Fine_tuning.md).
+- 🎁 2025.01.23: SWIFT support the `sample` command, this is a very important feature for complex CoT and RFT. Meanwhile, we support an [Reinforced Fine-tuning script](docs/source_en/Instruction/Reinforced_Fine_tuning.md).
 - 🎁 2024.12.04: **SWIFT3.0** major version update. Please check the [Release Notes and Changes](https://swift.readthedocs.io/en/latest/Instruction/ReleaseNote3.0.html).
 - 🎉 2024.08.12: The SWIFT paper has been published on arXiv, and you can read it [here](https://arxiv.org/abs/2408.05517).
 - 🔥 2024.08.05: Support for using [evalscope](https://github.com/modelscope/evalscope/) as a backend for evaluating large models and multimodal models.

README_CN.md

@@ -74,7 +74,7 @@
 - **模型量化**：支持AWQ、GPTQ和BNB的量化导出，导出的模型支持使用vLLM/LmDeploy推理加速，并支持继续训练。
 
 ## 🎉 新闻
-- 🎁 2024.01.23: SWIFT支持了`sample`命令, 这是一个对CoT和RFT非常重要的命令。同时, 我们支持了一个[强化微调脚本](docs/source/Instruction/强化微调.md)。
+- 🎁 2025.01.23: SWIFT支持了`sample`命令, 这是一个对CoT和RFT非常重要的命令。同时, 我们支持了一个[强化微调脚本](docs/source/Instruction/强化微调.md)。
 - 🎁 2024.12.04: **SWIFT3.0**大版本更新。请查看[发布说明和更改](https://swift.readthedocs.io/zh-cn/latest/Instruction/ReleaseNote3.0.html)。
 - 🎉 2024.08.12: SWIFT论文已经发布到arXiv上，可以点击[这里](https://arxiv.org/abs/2408.05517)阅读。
 - 🔥 2024.08.05: 支持使用[evalscope](https://github.com/modelscope/evalscope/)作为后端进行大模型和多模态模型的评测。

docs/source/Customization/自定义数据集.md

@@ -86,7 +86,7 @@ query-response格式：
 
 ### 多模态
 
-对于多模态数据集，和上述任务的格式相同。区别在于增加了`images`, `videos`, `audios`几个key，分别代表多模态资源的url或者path（推荐使用绝对路径），`<image>` `<video>` `<audio>`标签代表了插入图片/视频/音频的位置，ms-swift支持多图片/视频/音频的情况。下面给出的四条示例分别展示了纯文本，以及包含图像、视频和音频数据的数据格式。
+对于多模态数据集，和上述任务的格式相同。区别在于增加了`images`, `videos`, `audios`几个key，分别代表多模态资源的url或者path（推荐使用绝对路径），`<image>` `<video>` `<audio>`标签代表了插入图片/视频/音频的位置，ms-swift支持多图片/视频/音频的情况。这些特殊tokens将在预处理的时候进行替换，参考[这里](https://github.com/modelscope/ms-swift/blob/main/swift/llm/template/template/qwen.py#L198)。下面给出的四条示例分别展示了纯文本，以及包含图像、视频和音频数据的数据格式。
 
 预训练：
 ```

docs/source/GetStarted/快速开始.md

@@ -56,7 +56,14 @@ swift sft \
     --model_name swift-robot
 ```
 
-训练完成后，使用以下命令对训练后的权重进行推理，这里的`--adapters`替换成训练生成的last checkpoint文件夹。由于adapters文件夹中包含了训练的参数文件，因此不需要额外指定`--model`, `--system`。
+小贴士：
+- 如果要使用自定义数据集进行训练，你可以参考[这里](../Customization/自定义数据集.md)组织数据集格式，并指定`--dataset <dataset_path>`。
+- `--model_author`和`--model_name`参数只有当数据集中包含`swift/self-cognition`时才生效。
+- 如果要使用其他模型进行训练，你只需要修改`--model <model_id/model_path>`即可。
+- 默认使用ModelScope进行模型和数据集的下载。如果要使用HuggingFace，指定`--use_hf true`即可。
+
+训练完成后，使用以下命令对训练后的权重进行推理：
+- 这里的`--adapters`需要替换成训练生成的last checkpoint文件夹。由于adapters文件夹中包含了训练的参数文件`args.json`，因此不需要额外指定`--model`，`--system`，swift会自动读取这些参数。如果要关闭此行为，可以设置`--load_args false`。
 
 ```shell
 # 使用交互式命令行进行推理
@@ -79,6 +86,17 @@ swift infer \
     --max_new_tokens 2048
 ```
 
+最后，使用以下命令将模型推送到ModelScope：
+```shell
+CUDA_VISIBLE_DEVICES=0 \
+swift export \
+    --adapters output/vx-xxx/checkpoint-xxx \
+    --push_to_hub true \
+    --hub_model_id '<your-model-id>' \
+    --hub_token '<your-sdk-token>' \
+    --use_hf false
+```
+
 ## 了解更多
 
 - 更多Shell脚本：[https://github.com/modelscope/ms-swift/tree/main/examples](https://github.com/modelscope/ms-swift/tree/main/examples)

docs/source_en/Customization/Custom-dataset.md

@@ -87,7 +87,7 @@ The following outlines the standard dataset format for ms-swift, where the "syst
 
 ### Multimodal
 
-For multimodal datasets, the format is the same as the aforementioned tasks. The difference lies in the addition of several keys: `images`, `videos`, and `audios`, which respectively represent the URLs or paths (absolute paths are recommended) of multimodal resources. The tags `<image>`, `<video>`, and `<audio>` indicate the positions where images, videos, and audio should be inserted. MS-Swift supports the inclusion of multiple images, videos, and audio. The four examples provided below respectively demonstrate data formats for plain text and those containing image, video, and audio data.
+For multimodal datasets, the format is the same as the aforementioned tasks. The difference lies in the addition of several keys: `images`, `videos`, and `audios`, which represent the URLs or paths (preferably absolute paths) of multimodal resources. The tags `<image>`, `<video>`, and `<audio>` indicate where to insert images, videos, or audio. MS-Swift supports multiple images, videos, and audio files. These special tokens will be replaced during preprocessing, as referenced [here](https://github.com/modelscope/ms-swift/blob/main/swift/llm/template/template/qwen.py#L198). The four examples below respectively demonstrate the data format for plain text, as well as formats containing image, video, and audio data.
 
 
 Pre-training:

docs/source_en/GetStarted/Quick-start.md

@@ -56,7 +56,16 @@ swift sft \
     --model_name swift-robot
 ```
 
-After training is complete, use the following command to perform inference with the trained weights. The `--adapters` option should be replaced with the last checkpoint folder generated from the training. Since the adapters folder contains the parameter files from the training, there is no need to specify `--model` or `--system` separately.
+Tips:
+
+- If you want to train with a custom dataset, you can refer to [this guide](https://idealab.alibaba-inc.com/Customization/Custom_Dataset.md) to organize your dataset format and specify `--dataset <dataset_path>`.
+- The `--model_author` and `--model_name` parameters are only effective when the dataset includes `swift/self-cognition`.
+- To train with a different model, simply modify `--model <model_id/model_path>`.
+- By default, ModelScope is used for downloading models and datasets. If you want to use HuggingFace, simply specify `--use_hf true`.
+
+After training is complete, use the following command to infer with the trained weights:
+
+- Here, `--adapters` should be replaced with the last checkpoint folder generated during training. Since the adapters folder contains the training parameter file `args.json`, there is no need to specify `--model`, `--system` separately; Swift will automatically read these parameters. To disable this behavior, you can set `--load_args false`.
 
 ```shell
 # Using an interactive command line for inference.
@@ -79,6 +88,18 @@ swift infer \
     --max_new_tokens 2048
 ```
 
+Finally, use the following command to push the model to ModelScope:
+
+```shell
+CUDA_VISIBLE_DEVICES=0 \
+swift export \
+    --adapters output/vx-xxx/checkpoint-xxx \
+    --push_to_hub true \
+    --hub_model_id '<your-model-id>' \
+    --hub_token '<your-sdk-token>' \
+    --use_hf false
+```
+
 ## Learn More
 - More Shell scripts: [https://github.com/modelscope/ms-swift/tree/main/examples](https://github.com/modelscope/ms-swift/tree/main/examples)
 - Using Python: [https://github.com/modelscope/ms-swift/blob/main/examples/notebook/qwen2_5-self-cognition/self-cognition-sft.ipynb](https://github.com/modelscope/ms-swift/blob/main/examples/notebook/qwen2_5-self-cognition/self-cognition-sft.ipynb)

examples/train/pretrain/train.sh

@@ -13,8 +13,8 @@ swift pt \
     --learning_rate 1e-5 \
     --gradient_accumulation_steps $(expr 256 / $nproc_per_node) \
     --warmup_ratio 0.03 \
-    --eval_steps 100 \
-    --save_steps 100 \
+    --eval_steps 500 \
+    --save_steps 500 \
     --save_total_limit 2 \
     --logging_steps 5 \
     --deepspeed zero3 \