Created by: ljx111

两个数据增强操作RandomDistort 、ExpandImage加上后报错，请问一下这是什么问题导致的？ 以下是部分配置信息：

TrainReader:
  inputs_def:
    fields: ['image', 'im_info', 'im_id', 'gt_bbox', 'gt_class', 'is_crowd']
  dataset:
    !PolypsDataSet
    image_dir: image_train
    anno_path: label_train__.txt
    dataset_dir: dataset/polyps/data
  sample_transforms:
  - !DecodeImage
    to_rgb: true
  - !NormalizeImage
    is_channel_first: false
    is_scale: true
    mean: [0.485,0.456,0.406]
    std: [0.229, 0.224,0.225]
  - !ResizeImage
    target_size: 512
    max_size: 1333
    interp: 1
    use_cv2: true
  - !RandomFlipImage
    prob: 0.5
  - !RandomDistort ###
    brightness_lower: 0.75
    brightness_upper: 1.5
    contrast_lower: 0.75
    contrast_upper: 1.5
    saturation_lower: 0.5
    saturation_upper: 1.5
    hue_lower: -18
    hue_upper: 18
    brightness_prob: 0.5
    contrast_prob: 0.5
    saturation_prob: 0.5 
    hue_prob: 0.5
    count: 4
    is_order: False
  - !ExpandImage ###
    max_ratio: 1.25
    mean: [123.675, 116.28, 103.53]
    prob: 0.5
  - !RandomCrop
    aspect_ratio: [.5, 2.]
    thresholds: [.0, .1, .3, .5, .7, .9]
    scaling: [.3, 1.]
  - !MixupImage
    alpha: 1.5
    beta: 1.5
  - !Permute
    to_bgr: false
    channel_first: true
  batch_transforms:
  - !PadBatch
    pad_to_stride: 32
    use_padded_im_info: false
  batch_size: 16
  shuffle: true
  worker_num: 2
  use_process: false


以下是报名信息：
2020-06-14 23:50:30,323-INFO: places would be ommited when DataLoader is not iterable
W0614 23:50:31.137933  1011 device_context.cc:237] Please NOTE: device: 0, CUDA Capability: 70, Driver API Version: 10.1, Runtime API Version: 9.0
W0614 23:50:31.141782  1011 device_context.cc:245] device: 0, cuDNN Version: 7.3.
2020-06-14 23:50:32,759-INFO: Load model and fuse batch norm if have from http://paddle-imagenet-models-name.bj.bcebos.com/ResNet101_pretrained.tar...
2020-06-14 23:50:32,759-INFO: Found /home/aistudio/.cache/paddle/weights/ResNet101_pretrained
2020-06-14 23:50:32,767-INFO: Loading parameters from /home/aistudio/.cache/paddle/weights/ResNet101_pretrained...
2020-06-14 23:50:32,768-WARNING: /home/aistudio/.cache/paddle/weights/ResNet101_pretrained.pdparams not found, try to load model file saved with [ save_params, save_persistables, save_vars ]
2020-06-14 23:50:32,768-WARNING: /home/aistudio/.cache/paddle/weights/ResNet101_pretrained.pdparams not found, try to load model file saved with [ save_params, save_persistables, save_vars ]
2020-06-14 23:50:32,778-WARNING: variable file [ /home/aistudio/.cache/paddle/weights/ResNet101_pretrained/fc_0.w_0 /home/aistudio/.cache/paddle/weights/ResNet101_pretrained/fc_0.b_0 ] not used
2020-06-14 23:50:32,778-WARNING: variable file [ /home/aistudio/.cache/paddle/weights/ResNet101_pretrained/fc_0.w_0 /home/aistudio/.cache/paddle/weights/ResNet101_pretrained/fc_0.b_0 ] not used
2020-06-14 23:50:33,142-INFO: places would be ommited when DataLoader is not iterable
I0614 23:50:33.165014  1011 parallel_executor.cc:440] The Program will be executed on CUDA using ParallelExecutor, 1 cards are used, so 1 programs are executed in parallel.
I0614 23:50:33.210036  1011 build_strategy.cc:365] SeqOnlyAllReduceOps:0, num_trainers:1
I0614 23:50:33.329967  1011 parallel_executor.cc:307] Inplace strategy is enabled, when build_strategy.enable_inplace = True
I0614 23:50:33.363153  1011 parallel_executor.cc:375] Garbage collection strategy is enabled, when FLAGS_eager_delete_tensor_gb = 0
2020-06-14 23:50:33,391-INFO: fail to map op [RandomDistort_67107d] with error: Cannot handle this data type and stack:
Traceback (most recent call last):
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/PIL/Image.py", line 2645, in fromarray
    mode, rawmode = _fromarray_typemap[typekey]
KeyError: ((1, 1, 3), '<f4')

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/aistudio/work/PaddleDetection/ppdet/data/reader.py", line 45, in __call__
    data = f(data, ctx)
  File "/home/aistudio/work/PaddleDetection/ppdet/data/transform/operators.py", line 597, in __call__
    im = Image.fromarray(im)
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/PIL/Image.py", line 2647, in fromarray
    raise TypeError("Cannot handle this data type")
TypeError: Cannot handle this data type

2020-06-14 23:50:33,392-WARNING: recv endsignal from outq with errmsg[consumer[consumer-1ea-1] failed to map with error:[Cannot handle this data type]]
2020-06-14 23:50:33,392-INFO: fail to map op [RandomDistort_67107d] with error: Cannot handle this data type and stack:
Traceback (most recent call last):
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/PIL/Image.py", line 2645, in fromarray
    mode, rawmode = _fromarray_typemap[typekey]
KeyError: ((1, 1, 3), '<f4')

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/aistudio/work/PaddleDetection/ppdet/data/reader.py", line 45, in __call__
    data = f(data, ctx)
  File "/home/aistudio/work/PaddleDetection/ppdet/data/transform/operators.py", line 597, in __call__
    im = Image.fromarray(im)
  File "/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/PIL/Image.py", line 2647, in fromarray
    raise TypeError("Cannot handle this data type")
TypeError: Cannot handle this data type