Instructions to use rmihaylov/gpt2-medium-bg with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rmihaylov/gpt2-medium-bg with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="rmihaylov/gpt2-medium-bg", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("rmihaylov/gpt2-medium-bg", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("rmihaylov/gpt2-medium-bg", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use rmihaylov/gpt2-medium-bg with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rmihaylov/gpt2-medium-bg"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rmihaylov/gpt2-medium-bg",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/rmihaylov/gpt2-medium-bg

SGLang

How to use rmihaylov/gpt2-medium-bg with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "rmihaylov/gpt2-medium-bg" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rmihaylov/gpt2-medium-bg",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "rmihaylov/gpt2-medium-bg" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rmihaylov/gpt2-medium-bg",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use rmihaylov/gpt2-medium-bg with Docker Model Runner:
```
docker model run hf.co/rmihaylov/gpt2-medium-bg
```

gpt2-medium-bg / modeling_gpt2.py

rmihaylov

add model

a0a9adf over 4 years ago

Raw

History Blame Contribute Delete

1.02 kB

	from torch import nn

	from transformers import GPT2LMHeadModel as GPT2LMHeadModelBase
	from transformers.models.gpt2.modeling_gpt2 import GPT2Block as GPT2BlockBase


	class GPT2Block(GPT2BlockBase):
	def forward(self, x, layer_past=None,
	attention_mask=None, head_mask=None, use_cache=False,
	encoder_hidden_states=None, encoder_attention_mask=None, output_attentions=None):

	x = self.ln_1(x)
	output_attn = self.attn(
	x, layer_past=layer_past,
	attention_mask=attention_mask,
	head_mask=head_mask,
	use_cache=use_cache)

	a = output_attn[0]
	x = x + a

	m = self.mlp(self.ln_2(x))
	x = x + m

	outputs = (x,) + output_attn[1:]
	return outputs


	class GPT2LMHeadModel(GPT2LMHeadModelBase):
	def __init__(self, config):
	super().__init__(config)
	self.transformer.h = nn.ModuleList([GPT2Block(config, layer_idx) for layer_idx in range(config.n_layer)])