DeepLab2 is a TensorFlow library for deep labeling, aiming to provide a unified and state-of-the-art TensorFlow codebase for dense pixel labeling tasks.
while testing the configuration i'm getting this error:
python deeplab2/trainer/evaluator_test.py
Traceback (most recent call last):
File "deeplab2/trainer/evaluator_test.py", line 33, in
from deeplab2.model import deeplab
File "/home/mus/Desktop/Seg/deeplab2/model/deeplab.py", line 27, in
from deeplab2.model import builder
File "/home/mus/Desktop/Seg/deeplab2/model/builder.py", line 33, in
from deeplab2.model.pixel_encoder import moat
File "/home/mus/Desktop/Seg/deeplab2/model/pixel_encoder/moat.py", line 32, in
from deeplab2.model.layers.moat_blocks import MBConvBlock
File "/home/mus/Desktop/Seg/deeplab2/model/layers/moat_blocks.py", line 146, in
class MBConvBlock(tf.keras.layers.Layer):
File "/home/mus/Desktop/Seg/deeplab2/model/layers/moat_blocks.py", line 204, in MBConvBlock
def build(self, input_shape: list[int]) -> None:
TypeError: 'type' object is not subscriptable
I' m using ubunto 20.04
cuda 11.2
tensorflow 2.6.0
keras 2.6.0
gcc 8
while testing the configuration i'm getting this error:
I' m using ubunto 20.04 cuda 11.2 tensorflow 2.6.0 keras 2.6.0 gcc 8