预训练模型需要加载的权重文件

日萌社

人工智能AI：Keras PyTorch MXNet TensorFlow PaddlePaddle 深度学习实战（不定时更新）

---

链接：https://pan.baidu.com/s/13OtaUv6j4x8dD7cgD4sL5g 
提取码：7tze 

---

1.tensorflow下内置的keras加载的预训练模型加载权重文件的路径
	C:\Users\Administrator\.keras
2.NLP中预训练bert模型加载权重文件的路径
	如果每次运行都重新下载“bert-base-chinese”的话，执行如下操作
		window：cd C:/Users/Administrator/.cache/torch/hub/huggingface_pytorch-transformers_master
		linux：cd /root/.cache/torch/hub/huggingface_pytorch-transformers_master
		activate 虚拟环境名
		pip install .

C:\Users\Administrator\.keras
	datasets
		cifar-10-batches-py
		cifar-100-python
		fashion-mnist
		auto-mpg.data
		boston_housing.npz
		cifar-10-batches-py.tar.gz
		cifar-100-python.tar.gz
		fashion-mnist.rar
		imdb.npz
		imdb_word_index.json
		mnist.npz
		nietzsche.txt
		
	models
		inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5
		mobilenet_v2_weights_tf_dim_ordering_tf_kernels_1.0_224_no_top.h5
		vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5

	keras.json

C:\Users\Administrator\.cache\torch
	hub
		huggingface_pytorch-transformers_master
		transformers
	transformers