Text Generation
Transformers
ONNX
Safetensors
opt
trl
sft
optimum
danbooru
text-generation-inference
Instructions to use p1atdev/dart-v1-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use p1atdev/dart-v1-sft with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="p1atdev/dart-v1-sft")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("p1atdev/dart-v1-sft") model = AutoModelForCausalLM.from_pretrained("p1atdev/dart-v1-sft", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use p1atdev/dart-v1-sft with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "p1atdev/dart-v1-sft" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "p1atdev/dart-v1-sft", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/p1atdev/dart-v1-sft
- SGLang
How to use p1atdev/dart-v1-sft 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 "p1atdev/dart-v1-sft" \ --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": "p1atdev/dart-v1-sft", "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 "p1atdev/dart-v1-sft" \ --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": "p1atdev/dart-v1-sft", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use p1atdev/dart-v1-sft with Docker Model Runner:
docker model run hf.co/p1atdev/dart-v1-sft
Update tokenization_dart.py
Browse files- tokenization_dart.py +1 -23
tokenization_dart.py
CHANGED
|
@@ -1,7 +1,5 @@
|
|
| 1 |
import logging
|
| 2 |
-
import
|
| 3 |
-
from typing import Dict, List
|
| 4 |
-
from pydantic.dataclasses import dataclass
|
| 5 |
|
| 6 |
from transformers import PreTrainedTokenizerFast
|
| 7 |
from tokenizers.decoders import Decoder
|
|
@@ -57,26 +55,6 @@ PROMPT_TEMPLATE = (
|
|
| 57 |
# fmt: on
|
| 58 |
|
| 59 |
|
| 60 |
-
@dataclass
|
| 61 |
-
class Category:
|
| 62 |
-
name: str
|
| 63 |
-
bos_token_id: int
|
| 64 |
-
eos_token_id: int
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
@dataclass
|
| 68 |
-
class TagCategoryConfig:
|
| 69 |
-
categories: Dict[str, Category]
|
| 70 |
-
category_to_token_ids: Dict[str, List[int]]
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
def load_tag_category_config(config_json: str):
|
| 74 |
-
with open(config_json, "rb") as file:
|
| 75 |
-
config: TagCategoryConfig = TagCategoryConfig(**json.loads(file.read()))
|
| 76 |
-
|
| 77 |
-
return config
|
| 78 |
-
|
| 79 |
-
|
| 80 |
class DartDecoder:
|
| 81 |
def __init__(self, special_tokens: List[str]):
|
| 82 |
self.special_tokens = list(special_tokens)
|
|
|
|
| 1 |
import logging
|
| 2 |
+
from typing import List
|
|
|
|
|
|
|
| 3 |
|
| 4 |
from transformers import PreTrainedTokenizerFast
|
| 5 |
from tokenizers.decoders import Decoder
|
|
|
|
| 55 |
# fmt: on
|
| 56 |
|
| 57 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
class DartDecoder:
|
| 59 |
def __init__(self, special_tokens: List[str]):
|
| 60 |
self.special_tokens = list(special_tokens)
|