AttributeError: module 'tensorflow' has no attribute 'placeholder'

[YenHong818]

Suddenly have this problem. Isit tensorflow problem? How can I solve this? I install tensorflow 1.14 temporarily to run the code, but it is very slow while training. Please Help

/content/drive/My Drive/FYP2
2020-03-30 07:50:00.328488: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1
Using TensorFlow backend.
Weights:  coco
Dataset:  dataset
Logs:  /content/drive/My Drive/FYP2/logs

Configurations:
BACKBONE                       resnet101
BACKBONE_STRIDES               [4, 8, 16, 32, 64]
BATCH_SIZE                     2
BBOX_STD_DEV                   [0.1 0.1 0.2 0.2]
COMPUTE_BACKBONE_SHAPE         None
DETECTION_MAX_INSTANCES        3
DETECTION_MIN_CONFIDENCE       0.5
DETECTION_NMS_THRESHOLD        0.9
FPN_CLASSIF_FC_LAYERS_SIZE     1024
GPU_COUNT                      1
GRADIENT_CLIP_NORM             5.0
IMAGES_PER_GPU                 2
IMAGE_CHANNEL_COUNT            3
IMAGE_MAX_DIM                  512
IMAGE_META_SIZE                15
IMAGE_MIN_DIM                  512
IMAGE_MIN_SCALE                0
IMAGE_RESIZE_MODE              square
IMAGE_SHAPE                    [512 512   3]
LEARNING_MOMENTUM              0.9
LEARNING_RATE                  0.005
LOSS_WEIGHTS                   {'rpn_class_loss': 1.0, 'rpn_bbox_loss': 1.0, 'mrcnn_class_loss': 1.0, 'mrcnn_bbox_loss': 1.0, 'mrcnn_mask_loss': 1.0}
MASK_POOL_SIZE                 14
MASK_SHAPE                     [28, 28]
MAX_GT_INSTANCES               6
MEAN_PIXEL                     [53.129, 53.129, 53.129]
MINI_MASK_SHAPE                (56, 56)
NAME                           lesion
NUM_CLASSES                    3
POOL_SIZE                      7
POST_NMS_ROIS_INFERENCE        1000
POST_NMS_ROIS_TRAINING         2500
PRE_NMS_LIMIT                  6000
ROI_POSITIVE_RATIO             0.33
RPN_ANCHOR_RATIOS              [0.5, 1, 2]
RPN_ANCHOR_SCALES              (32, 64, 128, 256, 512)
RPN_ANCHOR_STRIDE              1
RPN_BBOX_STD_DEV               [0.1 0.1 0.2 0.2]
RPN_NMS_THRESHOLD              0.7
RPN_TRAIN_ANCHORS_PER_IMAGE    256
STEPS_PER_EPOCH                400
TOP_DOWN_PYRAMID_SIZE          256
TRAIN_BN                       False
TRAIN_ROIS_PER_IMAGE           400
USE_MINI_MASK                  True
USE_RPN_ROIS                   True
VALIDATION_STEPS               50
WEIGHT_DECAY                   0.0001


Traceback (most recent call last):
  File "lesion.py", line 447, in <module>
    model_dir=args.logs)
  File "/content/drive/My Drive/FYP2/mrcnn/model.py", line 1837, in __init__
    self.keras_model = self.build(mode=mode, config=config)
  File "/content/drive/My Drive/FYP2/mrcnn/model.py", line 1856, in build
    shape=[None, None, config.IMAGE_SHAPE[2]], name="input_image")
  File "/usr/local/lib/python3.6/dist-packages/keras/engine/input_layer.py", line 178, in Input
    input_tensor=tensor)
  File "/usr/local/lib/python3.6/dist-packages/keras/legacy/interfaces.py", line 91, in wrapper
    return func(*args, **kwargs)
  File "/usr/local/lib/python3.6/dist-packages/keras/engine/input_layer.py", line 87, in __init__
    name=self.name)
  File "/usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py", line 541, in placeholder
    x = tf.placeholder(dtype, shape=shape, name=name)
AttributeError: module 'tensorflow' has no attribute 'placeholder'

[tomklaver1995]

Same issue since today.. Worked 3 days ago suddenly cannot run it anymore:

---

AttributeError Traceback (most recent call last) in () 1 model = modellib.MaskRCNN(mode="training", config=config, ----> 2 model_dir=MODEL_DIR) 3 print(MODEL_DIR)

5 frames /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py in placeholder(shape, ndim, dtype, sparse, name) 539 x = tf.sparse_placeholder(dtype, shape=shape, name=name) 540 else: --> 541 x = tf.placeholder(dtype, shape=shape, name=name) 542 x._keras_shape = shape 543 x._uses_learning_phase = False

AttributeError: module 'tensorflow' has no attribute 'placeholder'

[Rakesh-Raushan]

Facing the same error: This seems to occur because of the changes in Tensorflow 2.0 which no more has placeholders and works more in a pythonic way.

This page can be used as a guide to do the needful for the issues we are facing!

https://www.tensorflow.org/guide/migrate

AttributeError Traceback (most recent call last) in () ----> 1 model = modellib.MaskRCNN(mode="inference", model_dir=MODEL_DIR, config=config) 2 3 # Load weights trained on MS-COCO 4 model.load_weights(COCO_MODEL_PATH, by_name=True)

5 frames /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py in placeholder(shape, ndim, dtype, sparse, name) 539 x = tf.sparse_placeholder(dtype, shape=shape, name=name) 540 else: --> 541 x = tf.placeholder(dtype, shape=shape, name=name) 542 x._keras_shape = shape 543 x._uses_learning_phase = False

AttributeError: module 'tensorflow' has no attribute 'placeholder'

[KowsalyaR97]

Downgrade the tensorflow version as 1 using the following command %tensorflow_version 1.x

[NeyoxDrago]

try using :: import tensorflow.compat.v1 as tf

it will work.

[ahalota]

Is there an extra step needed after adding this line?

try using :: import tensorflow.compat.v1 as tf

it will work.

I made that change in all .py files inside of

• ./mrcnn
• ./build/lib/mrcnn/

But I think my error message is referencing an instance in /path/to/conda/env/lib/python3.7/site-packages/mask_rcnn-2.1-py3.7.egg/mrcnn/model.py

I can't figure out a one-step way to make the replacement and have the .ipynb file notice it.

[rakesh-kumar-k]

Is there an extra step needed after adding this line?

try using :: import tensorflow.compat.v1 as tf it will work.

I made that change in all .py files inside of

• ./mrcnn
• ./build/lib/mrcnn/

But I think my error message is referencing an instance in /path/to/conda/env/lib/python3.7/site-packages/mask_rcnn-2.1-py3.7.egg/mrcnn/model.py

I can't figure out a one-step way to make the replacement and have the .ipynb file notice it.

did you solve?

[Nishanth2708]

I solved the issue by,

-Navigating to, "/usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py".

• Then, replace tensorflow import with i -import tensorflow.compat.v1 as tf -tf.compat.v1.disable_eager_execution()
• Now, on line 442 of tensorflow_backend.py: change the hyperparameter dtype from None to tf.float32

Let me know if error still persists...!

[lichq]

How can i solve this problem: AttributeError: module 'tensorflow' has no attribute 'log' tensorflow 1.14.0 keras 2.2.4

[RaghadZenki]

@lichq try tf.math.log instead of tf.log

[joeinus134131]

Saya memecahkan masalah dengan,

-Navigasi ke, "/usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py".

• Kemudian, ganti impor tensorflow dengan i -import tensorflow.compat.v1 sebagai tf -tf.compat.v1.disable_eager_execution()
• Sekarang, pada baris 442 dari tensorflow_backend.py: ubah hyperparameter dtype dari None menjadi tf.float32

Beri tahu saya jika kesalahan masih berlanjut...!

why on line 442 in my tensorflow there is no syntax dtype = None?