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

本站现有博文333篇,共被浏览922249

本站已经建立2628天!

热门文章
文章归档
回到顶部