From 892e561c28cbbab1ee38a8022ddd9b397c873563 Mon Sep 17 00:00:00 2001 From: hiyouga <467089858@qq.com> Date: Thu, 13 Jun 2024 03:26:10 +0800 Subject: [PATCH] update examples --- examples/README.md | 10 ++++++---- examples/README_zh.md | 10 ++++++---- 2 files changed, 12 insertions(+), 8 deletions(-) diff --git a/examples/README.md b/examples/README.md index 3372afb9..180d5f7b 100644 --- a/examples/README.md +++ b/examples/README.md @@ -11,6 +11,8 @@ Make sure to execute these commands in the `LLaMA-Factory` directory. - [Inferring LoRA Fine-Tuned Models](#inferring-lora-fine-tuned-models) - [Extras](#extras) +Use `CUDA_VISIBLE_DEVICES` (GPU) or `ASCEND_RT_VISIBLE_DEVICES` (NPU) to choose computing devices. + ## Examples ### LoRA Fine-Tuning @@ -87,7 +89,7 @@ FORCE_TORCHRUN=1 NNODES=2 RANK=1 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llama #### Supervised Fine-Tuning with DeepSpeed ZeRO-3 (Weight Sharding) ```bash -FORCE_TORCHRUN=1 llamafactory-cli train examples/train_lora/llama3_lora_sft_ds.yaml +FORCE_TORCHRUN=1 llamafactory-cli train examples/train_lora/llama3_lora_sft_ds3.yaml ``` ### QLoRA Fine-Tuning @@ -121,14 +123,14 @@ CUDA_VISIBLE_DEVICES=0 llamafactory-cli train examples/train_qlora/llama3_lora_s #### Supervised Fine-Tuning on Single Node ```bash -FORCE_TORCHRUN=1 llamafactory-cli train examples/train_full/llama3_full_sft.yaml +FORCE_TORCHRUN=1 llamafactory-cli train examples/train_full/llama3_full_sft_ds3.yaml ``` #### Supervised Fine-Tuning on Multiple Nodes ```bash -FORCE_TORCHRUN=1 NNODES=2 RANK=0 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llamafactory-cli train examples/train_full/llama3_full_sft.yaml -FORCE_TORCHRUN=1 NNODES=2 RANK=1 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llamafactory-cli train examples/train_full/llama3_full_sft.yaml +FORCE_TORCHRUN=1 NNODES=2 RANK=0 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llamafactory-cli train examples/train_full/llama3_full_sft_ds3.yaml +FORCE_TORCHRUN=1 NNODES=2 RANK=1 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llamafactory-cli train examples/train_full/llama3_full_sft_ds3.yaml ``` #### Batch Predicting and Computing BLEU and ROUGE Scores diff --git a/examples/README_zh.md b/examples/README_zh.md index 64c31fbd..b6168a95 100644 --- a/examples/README_zh.md +++ b/examples/README_zh.md @@ -11,6 +11,8 @@ - [推理 LoRA 模型](#推理-lora-模型) - [杂项](#杂项) +使用 `CUDA_VISIBLE_DEVICES`（GPU）或 `ASCEND_RT_VISIBLE_DEVICES`（NPU）选择计算设备。 + ## 示例 ### LoRA 微调 @@ -87,7 +89,7 @@ FORCE_TORCHRUN=1 NNODES=2 RANK=1 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llama #### 使用 DeepSpeed ZeRO-3 平均分配显存 ```bash -FORCE_TORCHRUN=1 llamafactory-cli train examples/train_lora/llama3_lora_sft_ds.yaml +FORCE_TORCHRUN=1 llamafactory-cli train examples/train_lora/llama3_lora_sft_ds3.yaml ``` ### QLoRA 微调 @@ -121,14 +123,14 @@ llamafactory-cli train examples/train_qlora/llama3_lora_sft_aqlm.yaml #### 在单机上进行指令监督微调 ```bash -FORCE_TORCHRUN=1 llamafactory-cli train examples/train_full/llama3_full_sft.yaml +FORCE_TORCHRUN=1 llamafactory-cli train examples/train_full/llama3_full_sft_ds3.yaml ``` #### 在多机上进行指令监督微调 ```bash -FORCE_TORCHRUN=1 NNODES=2 RANK=0 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llamafactory-cli train examples/train_full/llama3_full_sft.yaml -FORCE_TORCHRUN=1 NNODES=2 RANK=1 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llamafactory-cli train examples/train_full/llama3_full_sft.yaml +FORCE_TORCHRUN=1 NNODES=2 RANK=0 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llamafactory-cli train examples/train_full/llama3_full_sft_ds3.yaml +FORCE_TORCHRUN=1 NNODES=2 RANK=1 MASTER_ADDR=192.168.0.1 MASTER_PORT=29500 llamafactory-cli train examples/train_full/llama3_full_sft_ds3.yaml ``` #### 批量预测并计算 BLEU 和 ROUGE 分数