File size: 6,704 Bytes
dee9fba
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
# Training customization

TRL is designed with modularity in mind so that users to be able to efficiently customize the training loop for their needs. Below are some examples on how you can apply and test different techniques.  Note: Although these examples use the DPOTrainer, the customization applies to most (if not all) trainers.

## Train on multiple GPUs / nodes

The trainers in TRL use 🤗 Accelerate to enable distributed training across multiple GPUs or nodes. To do so, first create an 🤗 Accelerate config file by running

```bash
accelerate config
```

and answering the questions according to your multi-gpu / multi-node setup. You can then launch distributed training by running:

```bash
accelerate launch your_script.py
```

We also provide config files in the [examples folder](https://github.com/huggingface/trl/tree/main/examples/accelerate_configs) that can be used as templates. To use these templates, simply pass the path to the config file when launching a job, e.g.:

```shell
accelerate launch --config_file=examples/accelerate_configs/multi_gpu.yaml --num_processes {NUM_GPUS} path_to_script.py --all_arguments_of_the_script
```

Refer to the [examples page](https://github.com/huggingface/trl/tree/main/examples) for more details.

### Distributed training with DeepSpeed

All of the trainers in TRL can be run on multiple GPUs together with DeepSpeed ZeRO-{1,2,3} for efficient sharding of the optimizer states, gradients, and model weights. To do so, run:

```shell
accelerate launch --config_file=examples/accelerate_configs/deepspeed_zero{1,2,3}.yaml --num_processes {NUM_GPUS} path_to_your_script.py --all_arguments_of_the_script
```

Note that for ZeRO-3, a small tweak is needed to initialize your reward model on the correct device via the `zero3_init_context_manager()` context manager. In particular, this is needed to avoid DeepSpeed hanging after a fixed number of training steps. Here is a snippet of what is involved from the [`sentiment_tuning`](https://github.com/huggingface/trl/blob/main/examples/scripts/ppo.py) example:

```python
ds_plugin = ppo_trainer.accelerator.state.deepspeed_plugin
if ds_plugin is not None and ds_plugin.is_zero3_init_enabled():
    with ds_plugin.zero3_init_context_manager(enable=False):
        sentiment_pipe = pipeline("sentiment-analysis", model="lvwerra/distilbert-imdb", device=device)
else:
    sentiment_pipe = pipeline("sentiment-analysis", model="lvwerra/distilbert-imdb", device=device)
```

Consult the 🤗 Accelerate [documentation](https://huggingface.co/docs/accelerate/usage_guides/deepspeed) for more information about the DeepSpeed plugin.


## Use different optimizers and schedulers

By default, the `DPOTrainer` creates a `torch.optim.AdamW` optimizer. You can create and define a different optimizer and pass it to `DPOTrainer` as follows:

```python
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoTokenizer
from torch import optim
from trl import DPOConfig, DPOTrainer

model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
dataset = load_dataset("trl-lib/ultrafeedback_binarized", split="train")
training_args = DPOConfig(output_dir="Qwen2.5-0.5B-DPO")

optimizer = optim.SGD(model.parameters(), lr=training_args.learning_rate)

trainer = DPOTrainer(
    model=model,
    args=training_args,
    train_dataset=dataset,
    tokenizer=tokenizer,
    optimizers=(optimizer, None),
)
trainer.train()
```

### Add a learning rate scheduler

You can also play with your training by adding learning rate schedulers.

```python
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoTokenizer
from torch import optim
from trl import DPOConfig, DPOTrainer

model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
dataset = load_dataset("trl-lib/ultrafeedback_binarized", split="train")
training_args = DPOConfig(output_dir="Qwen2.5-0.5B-DPO")

optimizer = optim.AdamW(model.parameters(), lr=training_args.learning_rate)
lr_scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=30, gamma=0.1)

trainer = DPOTrainer(
    model=model,
    args=training_args,
    train_dataset=dataset,
    tokenizer=tokenizer,
    optimizers=(optimizer, lr_scheduler),
)
trainer.train()
```

## Memory efficient fine-tuning by sharing layers

Another tool you can use for more memory efficient fine-tuning is to share layers between the reference model and the model you want to train.

```python
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoTokenizer
from trl import create_reference_model, DPOConfig, DPOTrainer

model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
ref_model = create_reference_model(model, num_shared_layers=6)
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
dataset = load_dataset("trl-lib/ultrafeedback_binarized", split="train[:1%]")
training_args = DPOConfig(output_dir="Qwen2.5-0.5B-DPO")

trainer = DPOTrainer(
    model=model,
    ref_model=ref_model,
    args=training_args,
    train_dataset=dataset,
    tokenizer=tokenizer,
)
trainer.train()
```

## Pass 8-bit reference models 
 
Since `trl` supports all keyword arguments when loading a model from `transformers` using `from_pretrained`, you can also leverage `load_in_8bit` from `transformers` for more memory efficient fine-tuning.

Read more about 8-bit model loading in `transformers` [here](https://huggingface.co/docs/transformers/en/peft#load-in-8bit-or-4bit).

```python
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoTokenizer
from trl import DPOConfig, DPOTrainer

model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
quantization_config = BitsAndBytesConfig(load_in_8bit=True)
ref_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct", quantization_config= quantization_config)
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
dataset = load_dataset("trl-lib/ultrafeedback_binarized", split="train")
training_args = DPOConfig(output_dir="Qwen2.5-0.5B-DPO")

trainer = DPOTrainer(
    model=model,
    ref_model=ref_model,
    args=training_args,
    train_dataset=dataset,
    tokenizer=tokenizer,
)
trainer.train()
```

## Use the CUDA cache optimizer

When training large models, you should better handle the CUDA cache by iteratively clearing it. To do so, simply pass `optimize_cuda_cache=True` to `DPOConfig`:

```python
training_args = DPOConfig(..., optimize_cuda_cache=True)
```