当目录不为空时，fastai 抛出“训练集为空”错误答案

【问题标题】：fastai throwing a "training set empty" error when the directories are not empty当目录不为空时，fastai 抛出“训练集为空”错误
【发布时间】：2020-08-20 15:14:56
【问题描述】：

我正在尝试构建一个包含训练集和测试集的分类器。然而，尽管训练集和测试集都不是空的，FastAI 还是抛出了UserWarning: Your training set is empty 错误。这是我的终端命令的输出，用于查看每个文件夹有多少项目：

%cd Food-101/images/train/
!ls -1 | wc -l

%cd ../test
!ls -1 | wc -l

%cd ../../

/content/Food-101/images/train
75750
/content/Food-101/images/test
25250
/content/Food-101

这是我的代码：

import pandas as pd

from fastai import *
from fastai.vision import *
from fastai.callbacks.hooks import *
from pathlib import Path
from numba import vectorize
from subprocess import call, run
import os, git, glob, shutil

file_parse = r'/([^/]+)_\d+\.(png|jpg|jpeg)$'

np.random.seed(42)
data = ImageDataBunch.from_folder(path, train= path + '/images/train', test=path + '/images/test', valid_pct=0.2, ds_tfms=get_transforms(), size=224)
data.normalize(imagenet_stats)

这是错误：

/usr/local/lib/python3.6/dist-packages/fastai/data_block.py:458: UserWarning: Your training set is empty. If this is by design, pass `ignore_empty=True` to remove this warning.
  warn("Your training set is empty. If this is by design, pass `ignore_empty=True` to remove this warning.")
/usr/local/lib/python3.6/dist-packages/fastai/data_block.py:461: UserWarning: Your validation set is empty. If this is by design, use `split_none()`
                 or pass `ignore_empty=True` when labelling to remove this warning.
  or pass `ignore_empty=True` when labelling to remove this warning.""")
---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
/usr/local/lib/python3.6/dist-packages/fastai/data_block.py in get_label_cls(self, labels, label_cls, label_delim, **kwargs)
    264         if label_delim is not None:             return MultiCategoryList
--> 265         try: it = index_row(labels,0)
    266         except: raise Exception("""Can't infer the type of your targets. 

7 frames
IndexError: index 0 is out of bounds for axis 0 with size 0

During handling of the above exception, another exception occurred:

Exception                                 Traceback (most recent call last)
/usr/local/lib/python3.6/dist-packages/fastai/data_block.py in get_label_cls(self, labels, label_cls, label_delim, **kwargs)
    265         try: it = index_row(labels,0)
    266         except: raise Exception("""Can't infer the type of your targets. 
--> 267 It's either because your data source is empty or because your labelling function raised an error.""")
    268         if isinstance(it, (float, np.float32)): return FloatList
    269         if isinstance(try_int(it), (str, Integral)):  return CategoryList

Exception: Can't infer the type of your targets. 
It's either because your data source is empty or because your labelling function raised an error.

我不明白为什么在两个目录都清楚地填充了图像时会引发此错误。

【问题讨论】：

标签： python exception deep-learning google-colaboratory fast-ai

【解决方案1】：

您能否确保将您的 path 变量设置为 /content/Food-101/ ？

【讨论】：

【解决方案2】：

我解决了这个问题：

data = ImageDataBunch.from_folder('/content/Food-101/images', train='/content/Food-101/images/train', test='/content/Food-101/images/test', valid_pct=0.2, ds_tfms=get_transforms(), size=224)

【讨论】：