metadata
license: apache-2.0
language:
- en
- zh
datasets:
- timdettmers/openassistant-guanaco
pipeline_tag: text-generation
Tags:
- Chat
widget:
- text: How can I find a girlfriend?
example_title: How can I find a girlfriend
- text: 你吃什么?
example_title: 你吃什么?
This is a beginner's test model.
from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch
modelName='MingZhuang/mz_model_merged'
model = AutoModelForCausalLM.from_pretrained(modelName)
tokenizer = AutoTokenizer.from_pretrained(modelName)
generate = transformers.pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
)
Result = generate(
"</s>Human: How can I find a girlfriend?<s> Assistant:",
max_new_tokens=120
)
Result