p04798526
/
LLaMA-Factory-Mirror
1
0
Fork 1
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update readme

This commit is contained in:
hiyouga 2024-02-25 16:26:08 +08:00
parent 4aab19c7ef
commit 261f631a1c
2 changed files with 8 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
README.md
View File

@ -516,7 +516,7 @@ python src/export_model.py \
> [!TIP]
> Use `--export_quantization_bit 4` and `--export_quantization_dataset data/c4_demo.json` to quantize the model after merging the LoRA weights.
### API Demo
### Inference with OpenAI-style API
```bash
python src/api_demo.py \
@ -529,7 +529,7 @@ python src/api_demo.py \
> [!TIP]
> Visit `http://localhost:8000/docs` for API documentation.
### CLI Demo
### Inference with command line
```bash
python src/cli_demo.py \
@ -539,7 +539,7 @@ python src/cli_demo.py \
--finetuning_type lora
```
### Web Demo
### Inference with web browser
```bash
python src/web_demo.py \
@ -576,7 +576,7 @@ CUDA_VISIBLE_DEVICES=0 python src/train_bash.py \
--template default \
--finetuning_type lora \
--output_dir path_to_predict_result \
--per_device_eval_batch_size 8 \
--per_device_eval_batch_size 1 \
--max_samples 100 \
--predict_with_generate \
--fp16

8
README_zh.md
View File

@ -516,7 +516,7 @@ python src/export_model.py \
> [!TIP]
> 合并 LoRA 权重之后可再次使用 `--export_quantization_bit 4` 和 `--export_quantization_dataset data/c4_demo.json` 量化模型。
### API 服务
### 使用 OpenAI 风格 API 推理
```bash
python src/api_demo.py \
@ -529,7 +529,7 @@ python src/api_demo.py \
> [!TIP]
> 关于 API 文档请见 `http://localhost:8000/docs`。
### 命令行测试
### 使用命令行推理
```bash
python src/cli_demo.py \
@ -539,7 +539,7 @@ python src/cli_demo.py \
--finetuning_type lora
```
### 浏览器测试
### 使用浏览器推理
```bash
python src/web_demo.py \
@ -576,7 +576,7 @@ CUDA_VISIBLE_DEVICES=0 python src/train_bash.py \
--template default \
--finetuning_type lora \
--output_dir path_to_predict_result \
--per_device_eval_batch_size 8 \
--per_device_eval_batch_size 1 \
--max_samples 100 \
--predict_with_generate \
--fp16