论文相关补充知识

已于 2022-03-15 11:21:46 修改
阅读量592 收藏
点赞数
分类专栏： pytorch知识 文章标签： python pytorch
于 2022-03-15 11:20:04 首次发布
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/kking_edc/article/details/123498021
版权

str的使用

实验中对于T2T模型使用了str(model)，输出结果如下：

MyT_t_24(
  (model): T2T_ViT(
    (tokens_to_token): T2T_module(
      (soft_split0): Unfold(kernel_size=(7, 7), dilation=1, padding=(2, 2), stride=(4, 4))
      (soft_split1): Unfold(kernel_size=(3, 3), dilation=1, padding=(1, 1), stride=(2, 2))
      (soft_split2): Unfold(kernel_size=(3, 3), dilation=1, padding=(1, 1), stride=(2, 2))
      (attention1): Token_transformer(
        (norm1): LayerNorm((147,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=147, out_features=192, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=64, out_features=64, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((64,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=64, out_features=64, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=64, out_features=64, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (attention2): Token_transformer(
        (norm1): LayerNorm((576,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=576, out_features=192, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=64, out_features=64, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((64,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=64, out_features=64, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=64, out_features=64, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (project): Linear(in_features=576, out_features=512, bias=True)
    )
    (pos_drop): Dropout(p=0.0, inplace=False)
    (blocks): ModuleList(
      (0): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (1): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (2): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (3): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (4): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (5): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (6): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (7): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (8): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (9): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (10): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (11): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (12): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (13): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (14): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (15): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (16): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (17): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (18): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (19): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (20): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (21): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (22): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
      (23): Block(
        (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (attn): Attention(
          (qkv): Linear(in_features=512, out_features=1536, bias=False)
          (attn_drop): Dropout(p=0.0, inplace=False)
          (proj): Linear(in_features=512, out_features=512, bias=True)
          (proj_drop): Dropout(p=0.0, inplace=False)
        )
        (drop_path): Identity()
        (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
        (mlp): Mlp(
          (fc1): Linear(in_features=512, out_features=1536, bias=True)
          (act): GELU()
          (fc2): Linear(in_features=1536, out_features=512, bias=True)
          (drop): Dropout(p=0.0, inplace=False)
        )
      )
    )
    (norm): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
    (head): Linear(in_features=512, out_features=1000, bias=True)
  )
)

可以看出输出了整个模型的信息

你回到了你的家
关注
  • 0
    点赞
  • 踩
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
T2T VIT 学习笔记（附代码）
PLANTTHESON的博客
01-18 1272
T2T-ViT是一种基于Transformer的视觉模型，用于图像分类任务。它通过将图像分割成小的图块，并使用Transformer模型对这些图块进行编码和处理，从而实现对图像的分类。
【视觉 Transformer】超详细解读 T2T-ViT 模型
m0_63642362的博客
01-23 2765
给 ViT 加个 Token to Token(T2T) 来引入局部先验，解决 ViT 训练依赖大数据(JFT-300M)问题 ~
BiFormer 实验记录
allrubots的博客
05-22 1486
试验记录
【正则化】DropPath/drop_path用法
热门推荐
qq_43426908的博客
12-01 1万+
  DropPath/drop_path 是一种正则化手段，其效果是将深度学习模型中的多分支结构随机”删除“，python中实现如下所示： def drop_path(x, drop_prob: float = 0., training: bool = False): if drop_prob == 0. or not training: return x keep_prob = 1 - drop_prob shape = (x.shape[0],) + (1,) *
t2t模型启动web服务
啊噗啊噗
05-15 1018
t2t中是自带api实现web服务的，参考地址https://github.com/tensorflow/tensor2tensor/tree/master/tensor2tensor/serving 按照给出的步骤进行即可， 1、首先将模型进行export，这里不需要安装任何依赖 t2t-exporter --model=${MODEL} \ --hparams_set=$...
对知识推理的认识的相关论文
sky_tar的博客
04-21 1150
​ 论文一（） 中文引用格式: 官赛萍,靳小龙,贾岩涛,王元卓,程学旗.面向知识图谱的知识推理研究进展.软件学报,2018,29(10):29662994. http://www.jos.org.cn/1000-9825/5551.htm 1.1 知识推理的简介 面向知识图谱的知识推理旨在基于已有的知识图谱事实，推理新的事实或识别错误知识。例如，在DBpedia 中已知三元组(X,birthPlace,Y),可以在很大程度上推理出缺失的三元组(X,nationality,Y)。 主要分为：
AlphaFold2论文及其补充文档
10-05
2021年DeepMind公司发表在Nature上的获得最佳论文奖的论文的补充文档，论文好找，补充文档不好找，个人上传上来。 zip文件包含alphafoldv2版本的正文和补充文档 Paper Name：AlphaFold2: Highly accurate protein ...
物联网论文及相关知识
03-04
物联网，英文名Internet of Things（IoT），是21世纪科技发展的重要里程碑，它将物理世界与...通过对相关论文的深入学习，我们可以了解到物联网的最新动态、技术趋势和挑战，为参与这个领域的创新和开发提供有力支持。
议论文文体知识.pptx
10-08
【议论文文体知识详解】 议论文，作为一种重要的文体，主要用来对特定问题或事件进行分析评论，表达作者的观点、立场和主张。它被称为说理文，与记叙文、说明文共同构成了表达文体的主要类别。 **议论文与其他文体...
系分论文指导课老师补充PPT.rar
01-02
以下是这些文档可能涵盖的一些核心知识点： 1. **系统分析师的角色与职责**： - 系统分析师作为项目中的关键角色，负责分析业务需求，设计系统解决方案，并进行技术评估。 - 解释系统分析师如何在项目管理、需求...
《常识知识图谱概述》综述论文
12-23
常识性知识的来源旨在支持自然语言理解、计算机视觉和知识图的应用程序。这些源包含相互补充的知识，这使得它们的融合成为可能。然而，由于它们不同的关注点、建模方法和稀疏重叠，这样的融合并非微不足道。
T2T Transformer 笔记
自然语言处理大菜鸟--HsingWang
12-08 4064
讨论：https://www.jiqizhixin.com/articles/2017-06-28-5https://ricardokleinklein.github.io/2017/11/16/Attention-is-all-you-need.html1. Mutli GPU 和 Single 配置的区别https://github.com/tensorflow/tensor2tensor/i
python __str__()魔法函数使用
yutaixin的博客
10-27 999
python 中__str__(）魔法函数的使用说明代码 说明 定义一个类：StudentModel1,演示__str__魔法函数的使用，具体步骤和功能详见代码 代码 class StudentModel: def __init__(self,name='',age=0,score=0,id=0): self.name = name self.age = age self.score...
python__str__()用法 举例django创建模型
Doraemon_meow_meow的博客
05-10 473
class Users(models.Model): UserName = models.CharFild(max_length = 50) #魔术方法，当对该类对象进行字符串操作时自动触发 #要求该魔术方法，必须返回一个字符串类型的值 def __str__(self): return self.UserName ...
DeepMind重新设计高性能ResNet！无需激活归一化层
阿木寺的博客
02-07 1090
点击上方“CVer”，选择加"星标"置顶重磅干货，第一时间送达本文转载自：AI人工智能初学者Characterizing signal propagation to clo...
Swin Transformer 代码学习笔记(目标检测)
athrunsunny的博客
05-23 8707
为了方便将img_scale设置为[(224,224)]，此时的数据集中的图片会进行resize，并且将短边padding成32的倍数，按最长边与224的比例缩放最短边，即 x=短边 / (500/224)，x即为缩放后的实际长度，比如输入时500*287（w*h）的图片，缩放后的短边长为287 * 224 / 500 = 129，由于输入的图像尺寸需要是32的倍数，此时需要向上取整，所以padding后的短边长为160。接下来把这部分代码拆分出来解读。主体位于SwinTransformer类中。
MMLAB系列：MMCLS训练结果验证与测试
qq_52053775的博客
08-28 1002
运行demo/image_demo文件，指定模型参数运行。
IDENTITY属性的使用详解
Paul_roux
06-25 1万+
IDENTITY属性的使用1.创建查看IDENTITY 创建 IF OBJECT_ID('test','U') IS NOT NULL DROP TABLE test GO CREATE TABLE test ( id int IDENTITY(1, 1), --如果直接写IDENTITY则默认为（1， 1） c1 ch
使用Python的Turtle库绘制动态风车
最新发布
爱写代码的小朋友
07-20 395
python绘制风车
本科毕业论文设计撰写规范与模板
本科毕业论文设计是培养学生综合运用本学科的基本理论、专业知识和基本技能，提高分析和解决实际问题的能力，完成初步培养从事科学研究工作和专业工程技术或实务工作基本训练的重要环节。为统一和规范我校本科毕业...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值