Open stingoChen opened 11 months ago
from transformers import AutoTokenizer, AutoConfig from optimum.onnxruntime import ORTModelForCausalLM from optimum.onnxruntime import ORTModel import torch
from optimum.onnxruntime import ORTModelForCustomTasks
model_id = '/data1/wxd/HuangYuxuan/Projects/ChatGLM2-6B/model'
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True) inputs = tokenizer("hellp", return_tensors="pt")
config = AutoConfig.from_pretrained('THUDM/chatglm2-6b', trust_remote_code=True, cache_dir="/data1/wxd/HuangYuxuan/Projects/cxk")
model = ORTModelForCausalLM.from_pretrained("./glm_onnx", config=config,
)
Feature request
I tried to load the chatglm2 onnx model, but unfortunately got this reply. KeyError: 'chatglm model type is not supported yet in NormalizedConfig. Only albert, bart, bert, blenderbot, blenderbot_small, bloom, camembert, codegen, cvt, deberta, deberta-v2, deit, distilbert, donut-swin, electra, gpt2, gpt-bigcode, gpt_neo, gpt_neox, llama, gptj, imagegpt, longt5, marian, mbart, mt5, m2m_100, nystromformer, opt, pegasus, pix2struct, poolformer, regnet, resnet, roberta, speech_to_text, splinter, t5, trocr, whisper, vision-encoder-decoder, vit, xlm-roberta, yolos, mpt, gpt_bigcode are supported. If you want to support chatglm please propose a PR or open up an issue.'
Motivation
load chatglm2 onnx model
Your contribution
NO