python的h5py库保存模型 RuntimeError: Unable to create attribute (object header message is too large)

最新推荐文章于 2023-05-07 20:10:48 发布

chutongz

最新推荐文章于 2023-05-07 20:10:48 发布

阅读量7.6k

点赞数

分类专栏：原创文章标签： h5py python keras tensorflow

本文链接：https://blog.csdn.net/weixin_41864878/article/details/82876614

版权

在使用h5py库保存包含大量列的模型时，由于HDF5文件头大小限制（64kb）导致RuntimeError。为了解决这个问题，可以只保存权重，将模型结构以json格式单独保存。在ModelCheckpoint回调中设置save_weights_only=True，并更改文件路径的后缀为.h5。这样可以确保每个epoch或最佳模型的权重被正确保存，而不会遇到元数据存储空间不足的问题。

摘要由CSDN通过智能技术生成

在用h5py库函数保存模型时抛错，保存不了

Traceback (most recent call last):
  File "INQ.py", line 258, in <module>
    verbose=2)
  File "/home/chutz/anaconda3/lib/python3.5/site-packages/keras/engine/training.py", line 1705, in fit
    validation_steps=validation_steps)
  File "/home/chutz/anaconda3/lib/python3.5/site-packages/keras/engine/training.py", line 1256, in _fit_loop
    callbacks.on_epoch_end(epoch, epoch_logs)
  File "/home/chutz/anaconda3/lib/python3.5/site-packages/keras/callbacks.py", line 77, in on_epoch_end
    callback.on_epoch_end(epoch, logs)
  File "/home/chutz/anaconda3/lib/python3.5/site-packages/keras/callbacks.py", line 458, in on_epoch_end
    self.model.save(filepath, overwrite=True)
  File "/home/chutz/anaconda3/lib/python3.5/site-packages/keras/engine/topology.py", line 2591, in save
    save_model(self, filepath, overwrite, include_optimizer)
  Fi

最低0.47元/天解锁文章

chutongz

关注

0
点赞
踩
2

收藏

觉得还不错? 一键收藏
0
评论
python的h5py库保存模型 RuntimeError: Unable to create attribute (object header message is too large)

在用h5py库函数保存模型时抛错，保存不了Traceback (most recent call last): File "INQ.py", line 258, in &lt;module&gt; verbose=2) File "/home/chutz/anaconda3/lib/python3.5/site-packages/keras/engine/training....
复制链接

扫一扫

专栏目录