update qwen2 docs (#1097)

modelscope · Jun 7, 2024 · 66d9cf4 · 66d9cf4
1 parent 9858e4f
commit 66d9cf4
Show file tree

Hide file tree

Showing 13 changed files with 315 additions and 218 deletions.
diff --git a/docs/source/LLM/Qwen1.5全流程最佳实践.md b/docs/source/LLM/Qwen1.5全流程最佳实践.md
@@ -193,8 +193,7 @@ sft_args = SftArguments(
              f'{DatasetName.self_cognition}#500'],
     logging_steps=5,
     max_length=2048,
-    learning_rate=5e-5,
-    warmup_ratio=0.4,
+    learning_rate=1e-4,
     output_dir='output',
     lora_target_modules=['ALL'],
     model_name=['小黄', 'Xiao Huang'],
@@ -216,8 +215,7 @@ swift sft \
     --dataset alpaca-zh#500 alpaca-en#500 self-cognition#500 \
     --logging_steps 5 \
     --max_length 2048 \
-    --learning_rate 5e-5 \
-    --warmup_ratio 0.4 \
+    --learning_rate 1e-4 \
     --output_dir output \
     --lora_target_modules ALL \
     --model_name 小黄 'Xiao Huang' \
@@ -235,8 +233,7 @@ swift sft \
     --dataset alpaca-zh#500 alpaca-en#500 self-cognition#500 \
     --logging_steps 5 \
     --max_length 2048 \
-    --learning_rate 5e-5 \
-    --warmup_ratio 0.4 \
+    --learning_rate 1e-4 \
     --output_dir output \
     --lora_target_modules ALL \
     --model_name 小黄 'Xiao Huang' \
@@ -486,8 +483,7 @@ swift sft \
     --dataset alpaca-zh#500 alpaca-en#500 self-cognition#500 \
     --logging_steps 5 \
     --max_length 4096 \
-    --learning_rate 5e-5 \
-    --warmup_ratio 0.4 \
+    --learning_rate 1e-4 \
     --output_dir output \
     --lora_target_modules ALL \
     --model_name 小黄 'Xiao Huang' \

diff --git a/docs/source/LLM/VLLM推理加速与部署.md b/docs/source/LLM/VLLM推理加速与部署.md
@@ -530,8 +530,7 @@ swift sft \
     --dataset self-cognition#500 sharegpt-gpt4:default#1000 \
     --logging_steps 5 \
     --max_length 4096 \
-    --learning_rate 5e-5 \
-    --warmup_ratio 0.4 \
+    --learning_rate 1e-4 \
     --output_dir output \
     --lora_target_modules ALL \
     --model_name 小黄 'Xiao Huang' \

diff --git a/docs/source/LLM/命令行参数.md b/docs/source/LLM/命令行参数.md
@@ -104,7 +104,7 @@
 - `--ignore_args_error`: 是否忽略命令行传参错误抛出的Error, 默认为`False`. 如果需要拷贝代码到notebook中运行, 需要设置成True.
 - `--check_model_is_latest`: 检查模型是否是最新, 默认为`True`. 如果你需要断网进行训练, 请将该参数设置为`False`.
 - `--logging_dir`: 默认为`None`. 即设置为`f'{self.output_dir}/runs'`, 表示tensorboard文件存储路径.
-- `--report_to`: 默认为`['tensorboard']`.
+- `--report_to`: 默认为`['tensorboard']`. 可以设置`--report_to all`来报告所有已安装的集成.
 - `--acc_strategy`: 默认为`'token'`, 可选择的值包括: 'token', 'sentence'.
 - `--save_on_each_node`: 该参数在多机训练时生效, 默认为`True`.
 - `--save_strategy`: 保存checkpoint的策略, 默认为`'steps'`, 可选择的值包括: 'steps', 'epoch', 'no'.

diff --git a/docs/source/LLM/自我认知微调最佳实践.md b/docs/source/LLM/自我认知微调最佳实践.md
diff --git a/docs/source_en/LLM/Command-line-parameters.md b/docs/source_en/LLM/Command-line-parameters.md
@@ -104,7 +104,7 @@
 - `--ignore_args_error`: Whether to ignore Error thrown by command line parameter errors, default is `False`. Set to True if need to copy code to notebook to run.
 - `--check_model_is_latest`: Check if model is latest, default is `True`. Set this to `False` if you need to train offline.
 - `--logging_dir`: Default is `None`. I.e. set to `f'{self.output_dir}/runs'`, representing path to store tensorboard files.
-- `--report_to`: Default is `['tensorboard']`.
+- `--report_to`: Default is `['tensorboard']`. You can set `--report_to all` to report to all installed integrations.
 - `--acc_strategy`: Default is `'token'`, options include: 'token', 'sentence'.
 - `--save_on_each_node`: Takes effect during multi-machine training, default is `True`.
 - `--save_strategy`: Strategy for saving checkpoint, default is `'steps'`, options include: 'steps', 'epoch', no'.

diff --git a/docs/source_en/LLM/Qwen1.5-best-practice.md b/docs/source_en/LLM/Qwen1.5-best-practice.md
@@ -191,8 +191,7 @@ sft_args = SftArguments(
              f'{DatasetName.self_cognition}#500'],
     logging_steps=5,
     max_length=2048,
-    learning_rate=5e-5,
-    warmup_ratio=0.4,
+    learning_rate=1e-4,
     output_dir='output',
     lora_target_modules=['ALL'],
     model_name=['小黄', 'Xiao Huang'],
@@ -215,8 +214,7 @@ swift sft \
     --dataset alpaca-zh#500 alpaca-en#500 self-cognition#500 \
     --logging_steps 5 \
     --max_length 2048 \
-    --learning_rate 5e-5 \
-    --warmup_ratio 0.4 \
+    --learning_rate 1e-4 \
     --output_dir output \
     --lora_target_modules ALL \
     --model_name 小黄 'Xiao Huang' \
@@ -234,8 +232,7 @@ swift sft \
     --dataset alpaca-zh#500 alpaca-en#500 self-cognition#500 \
     --logging_steps 5 \
     --max_length 2048 \
-    --learning_rate 5e-5 \
-    --warmup_ratio 0.4 \
+    --learning_rate 1e-4 \
     --output_dir output \
     --lora_target_modules ALL \
     --model_name 小黄 'Xiao Huang' \
@@ -488,8 +485,7 @@ swift sft \
     --dataset alpaca-zh#500 alpaca-en#500 self-cognition#500 \
     --logging_steps 5 \
     --max_length 4096 \
-    --learning_rate 5e-5 \
-    --warmup_ratio 0.4 \
+    --learning_rate 1e-4 \
     --output_dir output \
     --lora_target_modules ALL \
     --model_name 小黄 'Xiao Huang' \