在resnet50_weights_tf上获取URL失败

Question

我必须在无法上网的地方训练一个模型。

base_cnn = resnet.ResNet50(
weights="imagenet", input_shape=target_shape + (3,), include_top=False
)

随后，培训失败:

Traceback (most recent call last):
base_cnn = resnet.ResNet50(
return ResNet(stack_fn, False, True, 'resnet50', include_top, weights,
weights_path = data_utils.get_file(
raise Exception(error_msg.format(origin, e.errno, e.reason))
Exception: URL fetch failure on https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels_notop.h5: None -- [Errno 101] Network is unreachable

是否有一种方法可以从驱动器加载权重，而不是获取URL?

Answer 1

权重可以下载为:

from tensorflow.keras.applications import resnet
base_cnn = resnet.ResNet50(
weights="imagenet", input_shape=target_shape + (3,), include_top=False
)
base_cnn.save("weights.h5")

然后加载保存的权重:

from tensorflow.keras.models import load_model
base_cnn=load_model('weights.h5')