Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
You can not select more than 25 topics
Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
ychao
8f10c40fc0
|
1 year ago | |
|---|---|---|
| README.md | 1 year ago | |
| arguments.py | 1 year ago | |
| callbacks.py | 1 year ago | |
| config.json | 1 year ago | |
| configuration_Mind.py | 1 year ago | |
| finetune.py | 1 year ago | |
| finetune_lora.sh | 1 year ago | |
| finetune_lora_v1.sh | 1 year ago | |
| modeling_mind.py | 1 year ago | |
| modeling_trainer_4GPU_v1.py | 1 year ago | |
| pcmind_test.py | 1 year ago | |
| pcmind_test_client.py | 1 year ago | |
| preprocess_utils.py | 1 year ago | |
| special_tokens_map.json | 1 year ago | |
| tokenization_baichuan.py | 1 year ago | |
| tokenizer.model | 1 year ago | |
| tokenizer_config.json | 1 year ago | |
| trainer.py | 1 year ago | |
README.md
PCMind_7B_huggingface
基于huggingface transformer的脑海7B模型推理和微调
pcmind_test.py 基于huggingface transformer的脑海7B模型加载和推理测试
pcmind_test_client.py 基于vllm加载脑海7B模型，openai api方式客户端与大模型交互
vllm服务端加载：python -m vllm.entrypoints.openai.api_server --model ckpt_7B_pytorch --host 127.0.0.1 --port 8000 --trust-remote-code
modeling_trainer_4GPU_v1.py 全参数微调
finetune_lora.sh 脑海7B模型lora微调
基于huggingface transformer的脑海7B模型推理和微调
Python Shell
Contributors (2)
