tensorflow - TPU本地文件系统不存在？

我编写了一个NN模型，该模型分析图像并最后提取8个浮点数。该模型在我的计算机上运行良好（但运行缓慢），因此我在TPU云上尝试了BAM！我有一个错误：

I1008 12：58：47.077905 140221679261440 tf_logging.py:115]从training_loop记录的错误：文件系统方案'[local]'未实现（文件：'/ home / gcloud_iba / Data / CGTR / model / GA_subset / model。 ckpt-0_temp_e840841d93124a67b54074b1c0fd7ae4'） [[{{node save / SaveV2}} = SaveV2 [dtypes = [DT_FLOAT，DT_FLOAT，DT_FLOAT，DT_FLOAT，DT_FLOAT，...，DT_FLOAT，DT_FLOAT，DT_FLOAT，DT_FLOAT，DT_INT64 / caer]，工作：0 /任务：0 /设备：CPU：0“]（保存/共享文件名，保存/ SaveV2 /张量名，保存/ SaveV2 / shape_and_slices，batch_normalization / beta / Read / ReadVariableOp，batch_normalization / beta / Momentum / Read_1 / ReadVariableOp，batch_normalization / gamma / Read / ReadVariableOp，batch_normalization / gamma / Momentum / Read_1 / ReadVariableOp，batch_normalization / moving_mean / Read / ReadVariableOp，batch_normalization / moving_variance / Read / ReadVariableOp，batch_normalization_1 / beta / Read / ReadVariableOp / Beta_mentation_1 ，batch_normalization_1 / gamma / Read / ReadVariableOp，batch_normalization_1 / gamma / Momentum / Read_1 / ReadVariableOp，batch_normalization_1 / moving_mean / Read / ReadVariableOp，batch_normalization_1 / moving_variance / Read / ReadVariableOp，conv2d / kernel / Read / Read / ReadVMoment / ReadVariab leOp，conv2d_1 /内核/ Read / ReadVariableOp，conv2d_1 /内核/动量/ Read_1 / ReadVariableOp，conv2d_2 /内核/ Read / ReadVariableOp，conv2d_2 /内核/动量/ Read_1 / ReadVariableOp，conv2d_3 / kernel / Readnel Momentum / Read_1 / ReadVariableOp，conv2d_4 / kernel / Read / ReadVariableOp，conv2d_4 / kernel / Momentum / Read_1 / ReadVariableOp，conv2d_5 / kernel / Read / ReadVariableOp，conv2d_5 / kernel / Momentum / Read_1 / ReadVariableOp conv2d_6 / kernel / Momentum / Read_1 / ReadVariableOp，conv2d_7 / kernel / Read / ReadVariableOp，conv2d_7 / kernel / Momentum / Read_1 / ReadVariableOp，conv2d_8 / kernel / Read / ReadVariableOp，conv2d_8 / kernel / Mov1d Read / ReadVariableOp，conv2d_9 / kernel / Momentum / Read_1 / ReadVariableOp，densed / bias / Read / ReadVariableOp，densed / bias / Momentum / Read_1 / ReadVariableOp，densed / kernel / Read / Read / ReadVariableOp，densed / kernel / Momentum / Read_1 / ReadVariable density_1 / bias / Read / ReadVariableOp，密实_1 / bias / Momentum / Read_1 / ReadVariable运算，密集_1 /内核/读取/ ReadVariableOp，密集_1 /内核/动量/读取_1 / ReadVariableOp，密集_2 /偏置/读取/ ReadVariableOp，密集_2 /偏置/动量/读取_1 / ReadVariableOp，密集_2 /内核/读取/ ReadVariableOp，密集_2 /内核/ Momentum / Read_1 / ReadVariableOp，dense_3 / bias / Read / ReadVariableOp，densage_3 / bias / Momentum / Read_1 / ReadVariableOp，densed_3 / kernel / Read / ReadVariableOp，densed_3 / kernel / Momentum / Read_1 / ReadVariableOp，global_step / Read / ReadV]

我检查了TPU是否可以访问硬盘驱动器，并且工作正常（我还有另一段代码可以读取另一模型的相同数据集）。我还没有缓存数据（但是），但是我做了一些预取。除此之外，我看不到什么不起作用？

感谢您提供的任何帮助！

Pi-r

TPU本地文件系统不存在？

2 个答案: