Quick Start¶
Welcome to MindNLP! This page will help you get started quickly.
Getting Started¶
For a comprehensive guide on how to use MindNLP, including loading pretrained models and fine-tuning them for your specific tasks, please visit our detailed tutorial:
Quick Start Tutorial - Learn how to fine-tune BERT for sentiment classification
Quick Examples¶
Using HuggingFace Transformers with MindSpore¶
import mindspore
import mindnlp
from transformers import pipeline
# Create a text generation pipeline with Qwen
pipe = pipeline(
task="text-generation",
model="Qwen/Qwen3-8B",
ms_dtype=mindspore.bfloat16,
device_map="auto"
)
chat = [
{"role": "user", "content": "Hello, how are you?"}
]
response = pipe(chat, max_new_tokens=512)
print(response[0]["generated_text"][-1]["content"])
Using MindNLP Native Interface¶
from mindnlp.transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
model = AutoModel.from_pretrained("bert-base-uncased")
inputs = tokenizer("Hello world!", return_tensors="ms")
outputs = model(**inputs)
More Tutorials¶
- Use Trainer - Training with MindNLP's Trainer API
- PEFT/LoRA - Parameter-efficient fine-tuning
- Data Preprocessing - Dataset handling and processing
- Use Mirror - Using model mirrors for faster downloads