EADST

Save Hugging Face Model with One Bin

max_shard_size (int or str, optional, defaults to "10GB") — Only applicable for models. The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size lower than this size. If expressed as a string, needs to be digits followed by a unit (like "5MB").

Based on the introduction, one bin model can be saved by changing the "max_shard_size".

LlamaForCausalLM.save_pretrained(base_model, output_dir, max_shard_size="100GB") # save one bin if the model is less than 100GB

Reference

PreTrainedModel

About Me
XD
Goals determine what you are going to be.
Category
标签云
Quantization diffusers PyCharm 签证 Ubuntu LeetCode printf Git Linux Translation torchinfo 继承 HaggingFace 论文速读 Miniforge 版权 CAM Windows CUDA Plotly Search Qwen Template Sklearn Augmentation Bin Tensor QWEN Password Vmess Nginx Base64 Paper Cloudreve PDF 论文 Pillow SQL SAM NLTK Qwen2.5 阿里云 FlashAttention Attention ChatGPT Dataset Bert Magnet Gemma XGBoost Mixtral 证件照 BF16 tar logger Input ONNX LaTeX TTS VPN CEIR Hilton TSV 搞笑 Permission Land Proxy Logo UNIX GGML PDB scipy FP32 SPIE Math llama.cpp WebCrawler Datetime git Google IndexTTS2 关于博主 SQLite PIP Markdown LoRA Bitcoin GPTQ TensorRT Knowledge ResNet-50 v0.dev OpenCV 强化学习 Disk 多进程 Algorithm Python SVR Clash AI API Paddle Agent 净利润 git-lfs Distillation BTC Website Heatmap OpenAI 公式 Shortcut DeepStream Card Rebuttal NLP CSV Github Tiktoken Django Jupyter GPT4 递归学习法 RGB Ptyhon Video Color Pytorch Quantize CC UI transformers 腾讯云 Random Michelin CLAP 图形思考法 icon uWSGI 音频 Hungarian Domain LLAMA InvalidArgumentError Use DeepSeek 飞书 Interview 算法题 uwsgi ModelScope CV COCO Pandas Diagram Animate mmap tqdm Food Pickle C++ MD5 Hotel Conda Review 财报 OCR EXCEL Numpy hf 报税 v2ray Vim PyTorch Image2Text CTC Qwen2 云服务器 Tracking Breakpoint Zip Data FP64 Anaconda 第一性原理 多线程 Bipartite VGG-16 Firewall GIT RAR YOLO GoogLeNet Streamlit FastAPI FP16 XML LLM Freesound TensorFlow Crawler NameSilo Web Jetson Transformers WAN 域名 Claude Excel Baidu Plate 顶会 Docker BeautifulSoup 图标 Llama Statistics JSON HuggingFace Safetensors News VSCode FP8
站点统计

本站现有博文327篇,共被浏览833400

本站已经建立2538天!

热门文章
文章归档
回到顶部