pytorch-loss:标签平滑,amsoftmax,散焦,三重损失,lovasz-softmax。 也许有用

上传者: 42111465 | 上传时间: 2023-03-21 11:04:16 | 文件大小: 93KB | 文件类型: ZIP
火炬损失 我实现的标签平滑,amsoftmax,焦点损耗,双焦点损耗,三重态损耗,giou损耗,亲和力损耗,pc_softmax_cross_entropy,ohem损耗(基于行硬挖掘损失的softmax),大利润- softmax(bmvc2019),lovasz-softmax-loss和dice-loss(广义的软骰子损失和批处理软骰子损失)。 也许这对我的未来工作很有用。 还尝试实现swish,hard-swish(hswish)和mish激活功能。 此外,添加了基于cuda的一键式功能(支持标签平滑)。 新添加一个“指数移动平均线(EMA)”运算符。 添加卷积运算,例如coord-conv2d和dynamic-conv2d(dy-conv2d)。 一些运算符是使用pytorch cuda扩展实现的,因此您需要先对其进行编译: $ python setup.py

文件下载

资源详情

[{"title":"( 58 个子文件 93KB ) pytorch-loss:标签平滑,amsoftmax,散焦,三重损失,lovasz-softmax。 也许有用","children":[{"title":"pytorch-loss-master","children":[{"title":"amsoftmax.py <span style='color:#111;'> 1.56KB </span>","children":null,"spread":false},{"title":"dual_focal_loss.py <span style='color:#111;'> 1.20KB </span>","children":null,"spread":false},{"title":"triplet_loss.py <span style='color:#111;'> 1024B </span>","children":null,"spread":false},{"title":"hswish.py <span style='color:#111;'> 3.98KB </span>","children":null,"spread":false},{"title":"affinity_loss.py <span style='color:#111;'> 4.43KB </span>","children":null,"spread":false},{"title":"conv_ops.py <span style='color:#111;'> 3.04KB </span>","children":null,"spread":false},{"title":"one_hot.py <span style='color:#111;'> 2.64KB </span>","children":null,"spread":false},{"title":"pytorch_loss","children":[{"title":"amsoftmax.py <span style='color:#111;'> 1.43KB </span>","children":null,"spread":false},{"title":"frelu.py <span style='color:#111;'> 557B </span>","children":null,"spread":false},{"title":"taylor_softmax.py <span style='color:#111;'> 3.58KB </span>","children":null,"spread":false},{"title":"dual_focal_loss.py <span style='color:#111;'> 1.20KB </span>","children":null,"spread":false},{"title":"triplet_loss.py <span style='color:#111;'> 1024B </span>","children":null,"spread":false},{"title":"hswish.py <span style='color:#111;'> 3.98KB </span>","children":null,"spread":false},{"title":"test.py <span style='color:#111;'> 2.29KB </span>","children":null,"spread":false},{"title":"affinity_loss.py <span style='color:#111;'> 4.43KB </span>","children":null,"spread":false},{"title":"conv_ops.py <span style='color:#111;'> 3.04KB </span>","children":null,"spread":false},{"title":"one_hot.py <span style='color:#111;'> 2.59KB </span>","children":null,"spread":false},{"title":"soft_dice_loss.py <span style='color:#111;'> 6.96KB </span>","children":null,"spread":false},{"title":"label_smooth.py <span style='color:#111;'> 8.50KB </span>","children":null,"spread":false},{"title":"lovasz_softmax.py <span style='color:#111;'> 7.65KB </span>","children":null,"spread":false},{"title":"mish.py <span style='color:#111;'> 3.86KB </span>","children":null,"spread":false},{"title":"ohem_loss.py <span style='color:#111;'> 2.47KB </span>","children":null,"spread":false},{"title":"__init__.py <span style='color:#111;'> 1.16KB </span>","children":null,"spread":false},{"title":"focal_loss.py <span style='color:#111;'> 8.08KB </span>","children":null,"spread":false},{"title":"dice_loss.py <span style='color:#111;'> 5.64KB </span>","children":null,"spread":false},{"title":"large_margin_softmax.py <span style='color:#111;'> 9.52KB </span>","children":null,"spread":false},{"title":"swish.py <span style='color:#111;'> 3.67KB </span>","children":null,"spread":false},{"title":"generalized_iou_loss.py <span style='color:#111;'> 1.34KB </span>","children":null,"spread":false},{"title":"ema.py <span style='color:#111;'> 2.90KB </span>","children":null,"spread":false},{"title":"pc_softmax.py <span style='color:#111;'> 5.77KB </span>","children":null,"spread":false}],"spread":false},{"title":"soft_dice_loss.py <span style='color:#111;'> 6.96KB </span>","children":null,"spread":false},{"title":"label_smooth.py <span style='color:#111;'> 8.50KB </span>","children":null,"spread":false},{"title":"lovasz_softmax.py <span style='color:#111;'> 7.75KB </span>","children":null,"spread":false},{"title":"mish.py <span style='color:#111;'> 3.86KB </span>","children":null,"spread":false},{"title":"LICENSE <span style='color:#111;'> 1.04KB </span>","children":null,"spread":false},{"title":"focal_loss.py <span style='color:#111;'> 8.08KB </span>","children":null,"spread":false},{"title":"setup.py <span style='color:#111;'> 1.25KB </span>","children":null,"spread":false},{"title":".gitignore <span style='color:#111;'> 1.18KB </span>","children":null,"spread":false},{"title":"dice_loss.py <span style='color:#111;'> 5.64KB </span>","children":null,"spread":false},{"title":"large_margin_softmax.py <span style='color:#111;'> 9.52KB </span>","children":null,"spread":false},{"title":"swish.py <span style='color:#111;'> 3.67KB </span>","children":null,"spread":false},{"title":"README.md <span style='color:#111;'> 2.76KB </span>","children":null,"spread":false},{"title":"csrc","children":[{"title":"cumsum.hpp <span style='color:#111;'> 4.64KB </span>","children":null,"spread":false},{"title":"swish_kernel.cu <span style='color:#111;'> 8.28KB </span>","children":null,"spread":false},{"title":"focal_kernel.cu <span style='color:#111;'> 6.57KB </span>","children":null,"spread":false},{"title":"one_hot_kernel.cu <span style='color:#111;'> 11.00KB </span>","children":null,"spread":false},{"title":"lsr_kernel.cu <span style='color:#111;'> 10.78KB </span>","children":null,"spread":false},{"title":"lovasz_softmax_old.cu <span style='color:#111;'> 24.05KB </span>","children":null,"spread":false},{"title":"mish_kernel.cu <span style='color:#111;'> 4.27KB </span>","children":null,"spread":false},{"title":"large_margin_kernel.cu <span style='color:#111;'> 19.96KB </span>","children":null,"spread":false},{"title":"soft_dice_kernel_v2.cu <span style='color:#111;'> 15.83KB </span>","children":null,"spread":false},{"title":"soft_dice_kernel.cu <span style='color:#111;'> 9.77KB </span>","children":null,"spread":false},{"title":"common.hpp <span style='color:#111;'> 1.11KB </span>","children":null,"spread":false},{"title":"ohem_label_kernel.cu <span style='color:#111;'> 9.45KB </span>","children":null,"spread":false},{"title":"lovasz_softmax.cu <span style='color:#111;'> 28.50KB </span>","children":null,"spread":false}],"spread":false},{"title":"generalized_iou_loss.py <span style='color:#111;'> 1.34KB </span>","children":null,"spread":false},{"title":"ema.py <span style='color:#111;'> 2.90KB </span>","children":null,"spread":false},{"title":"pc_softmax.py <span style='color:#111;'> 5.77KB </span>","children":null,"spread":false}],"spread":false}],"spread":true}]

评论信息

免责申明

【只为小站】的资源来自网友分享,仅供学习研究,请务必在下载后24小时内给予删除,不得用于其他任何用途,否则后果自负。基于互联网的特殊性,【只为小站】 无法对用户传输的作品、信息、内容的权属或合法性、合规性、真实性、科学性、完整权、有效性等进行实质审查;无论 【只为小站】 经营者是否已进行审查,用户均应自行承担因其传输的作品、信息、内容而可能或已经产生的侵权或权属纠纷等法律责任。
本站所有资源不代表本站的观点或立场,基于网友分享,根据中国法律《信息网络传播权保护条例》第二十二条之规定,若资源存在侵权或相关问题请联系本站客服人员,zhiweidada#qq.com,请把#换成@,本站将给予最大的支持与配合,做到及时反馈和处理。关于更多版权及免责申明参见 版权及免责申明