Created by: zachary-zheng

• 版本、环境信息 1）PaddleHub和PaddlePaddle版本：PaddleHub1.7.1，Paddle1.8.1 2）系统环境：Windows，python版本3.7.7
• 复现信息：如为报错，请给出复现环境、复现步骤 你好， 我根据如下link(https://github.com/PaddlePaddle/PaddleHub/tree/release/v1.7/demo/image_classification)的README.MD 进行如下代码，出现如下报错。 我尝试了解决方案（1）在下面代码基础上修改使用module = hub.Module(name="resnet_v2_50_imagenet")， 没有报错。 （2）在下面代码基础上将放置该脚本文件（finetuning.py）的文件夹名称由Fine_tuning_efficientnetb0改为Fine_tuning, 没有报错。 因此我很迷惑，下面代码报错的原因是什么？ 烦劳解答，谢谢。 code:

#coding:utf-8
import paddlehub as hub
import numpy as np

#加载预训练模型
# module = hub.Module(name="resnet_v2_50_imagenet")
module = hub.Module(name="efficientnetb0_imagenet")
inputs, outputs, program = module.context(trainable=True)
#下载数据集并使用ImageClassificationReader读取数据
dataset = hub.dataset.Flowers()
data_reader = hub.reader.ImageClassificationReader(
    image_width=module.get_expected_image_width(),
    image_height=module.get_expected_image_height(),
    images_mean=module.get_pretrained_images_mean(),
    images_std=module.get_pretrained_images_std(),
    dataset=dataset)
#选择优化策略和运行配置
strategy = hub.DefaultFinetuneStrategy(
    learning_rate=1e-4,
    optimizer_name="adam",
    regularization_coeff=1e-3)

config = hub.RunConfig(use_cuda=True, checkpoint_dir="cv_finetune_EfficientnetB0_2020-6-19-1", use_data_parallel=True, num_epoch=1, batch_size=8, strategy=strategy)
#构建网络并创建分类迁移任务进行Fine-tune
feature_map = outputs["feature_map"]
feed_list = [inputs["image"].name]

task = hub.ImageClassifierTask(
    data_reader=data_reader,
    feed_list=feed_list,
    feature=feature_map,
    num_classes=dataset.num_labels,
    config=config)

task.finetune_and_eval()
data = [r"C:\Users\admin\.paddlehub\dataset\flower_photos\daisy\525271784_013ddccd1b_m.jpg"] 
task.predict(data=data, return_result=True)  
print(task.predict(data=data, return_result=True))

Running error as following:

E:\Python\Python37-64\python.exe "E:/Python/Pycharm/my work/AI_study/Learning_from_Github/04 paddlepaddle-learn/03 Paddle_hub/peach_class/my_finegtuning/Fine_tuning_efficientnetb0/finetuning.py"
[2020-06-20 09:33:31,950] [    INFO] - Installing efficientnetb0_imagenet module
[2020-06-20 09:33:32,011] [    INFO] - Module efficientnetb0_imagenet already installed in C:\Users\admin\.paddlehub\modules\efficientnetb0_imagenet
[2020-06-20 09:33:32,586] [    INFO] - 311 pretrained paramaters loaded by PaddleHub
[2020-06-20 09:33:32,587] [    INFO] - Dataset C:\Users\admin\.paddlehub\dataset\flower_photos already cached.
[2020-06-20 09:33:32,612] [    INFO] - Dataset label map = {'roses': 0, 'tulips': 1, 'daisy': 2, 'sunflowers': 3, 'dandelion': 4}
[2020-06-20 09:33:32,612] [    INFO] - Checkpoint dir: cv_finetune_EfficientnetB0_2020-6-19-1
[2020-06-20 09:33:37,947] [    INFO] - Strategy with slanted triangle learning rate, L2 regularization, 
E:\Python\Python37-64\lib\site-packages\paddle\fluid\executor.py:1093: UserWarning: There are no operators in the program to be executed. If you pass Program manually, please use fluid.program_guard to ensure the current Program is being used.
  warnings.warn(error_info)
[2020-06-20 09:33:39,553] [    INFO] - Try loading checkpoint from cv_finetune_EfficientnetB0_2020-6-19-1\ckpt.meta
[2020-06-20 09:33:39,554] [    INFO] - PaddleHub model checkpoint not found, start from scratch...
[2020-06-20 09:33:39,693] [    INFO] - PaddleHub finetune start
[2020-06-20 09:33:43,610] [   TRAIN] - step 10 / 364: loss=0.27500 acc=0.27500 [step/sec: 2.70]
[2020-06-20 09:33:46,474] [   TRAIN] - step 20 / 364: loss=0.43750 acc=0.43750 [step/sec: 3.49]
[2020-06-20 09:33:49,364] [   TRAIN] - step 30 / 364: loss=0.57500 acc=0.57500 [step/sec: 3.46]
[2020-06-20 09:33:52,333] [   TRAIN] - step 40 / 364: loss=0.67500 acc=0.67500 [step/sec: 3.37]
[2020-06-20 09:33:55,201] [   TRAIN] - step 50 / 364: loss=0.67500 acc=0.67500 [step/sec: 3.49]
[2020-06-20 09:33:58,057] [   TRAIN] - step 60 / 364: loss=0.75000 acc=0.75000 [step/sec: 3.50]
[2020-06-20 09:34:00,934] [   TRAIN] - step 70 / 364: loss=0.76250 acc=0.76250 [step/sec: 3.48]
[2020-06-20 09:34:03,913] [   TRAIN] - step 80 / 364: loss=0.76250 acc=0.76250 [step/sec: 3.36]
[2020-06-20 09:34:06,765] [   TRAIN] - step 90 / 364: loss=0.72500 acc=0.72500 [step/sec: 3.51]
[2020-06-20 09:34:09,681] [   TRAIN] - step 100 / 364: loss=0.78750 acc=0.78750 [step/sec: 3.43]
[2020-06-20 09:34:09,681] [    INFO] - Evaluation on dev dataset start
share_vars_from is set, scope is ignored.
[2020-06-20 09:34:14,278] [    EVAL] - [dev dataset evaluation result] loss=0.84859 acc=0.84859 [step/sec: 12.69]
[2020-06-20 09:34:14,278] [    EVAL] - best model saved to cv_finetune_EfficientnetB0_2020-6-19-1\best_model [best acc=0.84859]
[2020-06-20 09:34:18,592] [   TRAIN] - step 110 / 364: loss=0.76250 acc=0.76250 [step/sec: 3.48]
[2020-06-20 09:34:21,455] [   TRAIN] - step 120 / 364: loss=0.75000 acc=0.75000 [step/sec: 3.49]
[2020-06-20 09:34:24,333] [   TRAIN] - step 130 / 364: loss=0.90000 acc=0.90000 [step/sec: 3.48]
[2020-06-20 09:34:27,203] [   TRAIN] - step 140 / 364: loss=0.80000 acc=0.80000 [step/sec: 3.48]
[2020-06-20 09:34:30,085] [   TRAIN] - step 150 / 364: loss=0.88750 acc=0.88750 [step/sec: 3.47]
[2020-06-20 09:34:32,956] [   TRAIN] - step 160 / 364: loss=0.73750 acc=0.73750 [step/sec: 3.48]
[2020-06-20 09:34:35,829] [   TRAIN] - step 170 / 364: loss=0.73750 acc=0.73750 [step/sec: 3.48]
[2020-06-20 09:34:38,713] [   TRAIN] - step 180 / 364: loss=0.90000 acc=0.90000 [step/sec: 3.47]
[2020-06-20 09:34:41,580] [   TRAIN] - step 190 / 364: loss=0.83750 acc=0.83750 [step/sec: 3.49]
[2020-06-20 09:34:44,452] [   TRAIN] - step 200 / 364: loss=0.80000 acc=0.80000 [step/sec: 3.48]
[2020-06-20 09:34:44,452] [    INFO] - Evaluation on dev dataset start
[2020-06-20 09:34:48,141] [    EVAL] - [dev dataset evaluation result] loss=0.89807 acc=0.89807 [step/sec: 13.16]
[2020-06-20 09:34:48,141] [    EVAL] - best model saved to cv_finetune_EfficientnetB0_2020-6-19-1\best_model [best acc=0.89807]
[2020-06-20 09:34:51,665] [   TRAIN] - step 210 / 364: loss=0.82500 acc=0.82500 [step/sec: 3.50]
[2020-06-20 09:34:54,527] [   TRAIN] - step 220 / 364: loss=0.90000 acc=0.90000 [step/sec: 3.49]
[2020-06-20 09:34:57,395] [   TRAIN] - step 230 / 364: loss=0.78750 acc=0.78750 [step/sec: 3.49]
[2020-06-20 09:35:00,277] [   TRAIN] - step 240 / 364: loss=0.80000 acc=0.80000 [step/sec: 3.47]
[2020-06-20 09:35:03,154] [   TRAIN] - step 250 / 364: loss=0.85000 acc=0.85000 [step/sec: 3.48]
[2020-06-20 09:35:06,016] [   TRAIN] - step 260 / 364: loss=0.85000 acc=0.85000 [step/sec: 3.49]
[2020-06-20 09:35:08,880] [   TRAIN] - step 270 / 364: loss=0.87500 acc=0.87500 [step/sec: 3.49]
[2020-06-20 09:35:11,767] [   TRAIN] - step 280 / 364: loss=0.88750 acc=0.88750 [step/sec: 3.47]
[2020-06-20 09:35:14,631] [   TRAIN] - step 290 / 364: loss=0.81250 acc=0.81250 [step/sec: 3.49]
[2020-06-20 09:35:17,484] [   TRAIN] - step 300 / 364: loss=0.87500 acc=0.87500 [step/sec: 3.51]
[2020-06-20 09:35:17,484] [    INFO] - Evaluation on dev dataset start
[2020-06-20 09:35:21,191] [    EVAL] - [dev dataset evaluation result] loss=0.93229 acc=0.93229 [step/sec: 13.09]
[2020-06-20 09:35:21,191] [    EVAL] - best model saved to cv_finetune_EfficientnetB0_2020-6-19-1\best_model [best acc=0.93229]
[2020-06-20 09:35:24,724] [   TRAIN] - step 310 / 364: loss=0.83750 acc=0.83750 [step/sec: 3.49]
[2020-06-20 09:35:27,577] [   TRAIN] - step 320 / 364: loss=0.81250 acc=0.81250 [step/sec: 3.51]
[2020-06-20 09:35:30,458] [   TRAIN] - step 330 / 364: loss=0.93750 acc=0.93750 [step/sec: 3.47]
[2020-06-20 09:35:33,320] [   TRAIN] - step 340 / 364: loss=0.86250 acc=0.86250 [step/sec: 3.49]
[2020-06-20 09:35:36,195] [   TRAIN] - step 350 / 364: loss=0.86250 acc=0.86250 [step/sec: 3.48]
[2020-06-20 09:35:39,043] [   TRAIN] - step 360 / 364: loss=0.87500 acc=0.87500 [step/sec: 3.51]
[2020-06-20 09:35:40,383] [    INFO] - Evaluation on dev dataset start
[2020-06-20 09:35:44,070] [    EVAL] - [dev dataset evaluation result] loss=0.92708 acc=0.92708 [step/sec: 13.17]
[2020-06-20 09:35:44,071] [    INFO] - Load the best model from cv_finetune_EfficientnetB0_2020-6-19-1\best_model
[2020-06-20 09:35:44,443] [    INFO] - Evaluation on test dataset start
[2020-06-20 09:35:48,072] [    EVAL] - [test dataset evaluation result] loss=0.92021 acc=0.92021 [step/sec: 13.14]
[2020-06-20 09:35:48,072] [    INFO] - Saving model checkpoint to cv_finetune_EfficientnetB0_2020-6-19-1\step_365
E:\Python\Python37-64\lib\site-packages\paddle\fluid\executor.py:1070: UserWarning: The following exception is not an EOF exception.
  "The following exception is not an EOF exception.")
Traceback (most recent call last):
  File "E:/Python/Pycharm/my work/AI_study/Learning_from_Github/04 paddlepaddle-learn/03 Paddle_hub/peach_class/my_finegtuning/Fine_tuning_efficientnetb0/finetuning.py", line 35, in <module>
    task.finetune_and_eval()
  File "E:\Python\Python37-64\lib\site-packages\paddlehub\finetune\task\base_task.py", line 868, in finetune_and_eval
    return self.finetune(do_eval=True)
  File "E:\Python\Python37-64\lib\site-packages\paddlehub\finetune\task\base_task.py", line 901, in finetune
    self.save_checkpoint()
  File "E:\Python\Python37-64\lib\site-packages\paddlehub\finetune\task\base_task.py", line 821, in save_checkpoint
    self.exe, dirname=model_saved_dir, main_program=self.main_program)
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\io.py", line 647, in save_persistables
    filename=filename)
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\io.py", line 301, in save_vars
    filename=filename)
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\io.py", line 356, in save_vars
    executor.run(save_program)
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\executor.py", line 1071, in run
    six.reraise(*sys.exc_info())
  File "E:\Python\Python37-64\lib\site-packages\six.py", line 703, in reraise
    raise value
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\executor.py", line 1066, in run
    return_merged=return_merged)
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\executor.py", line 1154, in _run_impl
    use_program_cache=use_program_cache)
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\executor.py", line 1229, in _run_program
    fetch_var_name)
paddle.fluid.core_avx.EnforceNotMet: 

--------------------------------------------
C++ Call Stacks (More useful to developers):
--------------------------------------------
Windows not support stack backtrace yet.

------------------------------------------
Python Call Stacks (More useful to users):
------------------------------------------
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\framework.py", line 2610, in append_op
    attrs=kwargs.get("attrs", None))
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\io.py", line 327, in save_vars
    attrs={'file_path': os.path.normpath(save_file_path)})
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\io.py", line 301, in save_vars
    filename=filename)
  File "E:\Python\Python37-64\lib\site-packages\paddle\fluid\io.py", line 647, in save_persistables
    filename=filename)
  File "E:\Python\Python37-64\lib\site-packages\paddlehub\finetune\task\base_task.py", line 821, in save_checkpoint
    self.exe, dirname=model_saved_dir, main_program=self.main_program)
  File "E:\Python\Python37-64\lib\site-packages\paddlehub\finetune\task\base_task.py", line 901, in finetune
    self.save_checkpoint()
  File "E:\Python\Python37-64\lib\site-packages\paddlehub\finetune\task\base_task.py", line 868, in finetune_and_eval
    return self.finetune(do_eval=True)
  File "E:/Python/Pycharm/my work/AI_study/Learning_from_Github/04 paddlepaddle-learn/03 Paddle_hub/peach_class/my_finegtuning/Fine_tuning_efficientnetb0/finetuning.py", line 35, in <module>
    task.finetune_and_eval()

----------------------
Error Message Summary:
----------------------
UnavailableError: Cannot open cv_finetune_EfficientnetB0_2020-6-19-1\step_365\@HUB_efficientnetb0_imagenet@_blocks.0._bn1_offset_beta1_pow_acc_0 to save variables.
  [Hint: Expected static_cast<bool>(fout) == true, but received static_cast<bool>(fout):0 != true:1.] at (D:/1.8.1/paddle\paddle/fluid/operators/save_op.h:85)
  [operator < save > error]

Process finished with exit code 1