张德长

导航

< 2025年3月 >
23 24 25 26 27 28 1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31 1 2 3 4 5

统计

11 2022 档案

Python学习笔记3
摘要:Python学习笔记3 pip的路径问题 pip安装路径问题 D:\ProgramFile\python-3.6.6-amd64\Lib\site.py # for distutils.commands.install # These values are initialized by the ge 阅读全文

posted @ 2022-11-19 17:36 张德长 阅读(346) 评论(0) 推荐(0) 编辑

基于TensorFlow和Python的机器学习(笔记4)
摘要:基于TensorFlow和Python的机器学习(笔记4) loss MSE =Mean Squared Error 均方差 Entropy熵 Cross Entropy交叉熵 熵越大,越不稳定,惊喜度越高 activation sigmoid 将每个值映射到0-1范围内 softmax 将每个值映 阅读全文

posted @ 2022-11-14 18:42 张德长 阅读(68) 评论(0) 推荐(0) 编辑

基于TensorFlow和Python的机器学习(笔记3)
摘要:基于TensorFlow和Python的机器学习(笔记3) # region 中文情感分析任务 参数配置 task_name = 'CC' # Chinese_classificationdo_train = Truedo_eval = Truedata_dir = '../GLUE/glue_da 阅读全文

posted @ 2022-11-12 17:12 张德长 阅读(81) 评论(0) 推荐(0) 编辑

Bert_Doc BERT文档中英文对照版
摘要:Bert_Doc BERT文档中英文对照版 README.md https://github.com/google-research/bert ... BERT BERT模型的全称是:BidirectionalEncoder Representations from Transformer。从名字中 阅读全文

posted @ 2022-11-11 14:28 张德长 阅读(276) 评论(0) 推荐(0) 编辑

BERT论文中文版:用于语言理解的深度双向Transformer的预训练
摘要:BERT论文中文版:用于语言理解的深度双向Transformer的预训练 Jacob Devlin Ming-Wei Chang Kenton Lee kristina Toutanova Google AI Language {jacobdevlin, mingweichang, kentonl, 阅读全文

posted @ 2022-11-11 12:36 张德长 阅读(796) 评论(0) 推荐(0) 编辑

BERT模型源码解析
摘要:BERT模型源码解析 modeling.py 目录 属性 类 class BertConfig(object) BERT模型配置参数类 class BertModel(object) BERT模型类 函数 def gelu(x) 格鲁激活函数 def get_activation(activatio 阅读全文

posted @ 2022-11-11 12:25 张德长 阅读(831) 评论(0) 推荐(0) 编辑

基于TensorFlow和Python的机器学习(笔记2)
摘要:基于TensorFlow和Python的机器学习(笔记2) LSTM模型摘要 Model: "sequential_1" _________________________________________________________________ Layer (type) Output Sha 阅读全文

posted @ 2022-11-02 16:21 张德长 阅读(111) 评论(0) 推荐(0) 编辑

点击右上角即可分享
微信分享提示