search

GeorgeSeif / Semantic-Segmentation-Suite

Semantic Segmentation Suite in TensorFlow. Implement, train, and test new Semantic Segmentation models easily!
2.51k stars 880 forks source link

One Python syntax error and 15 undefined names #103

Closed cclauss closed 6 years ago

cclauss commented 6 years ago

flake8 testing of https://github.com/GeorgeSeif/Semantic-Segmentation-Suite on Python 3.7.0

$ flake8 . --count --select=E901,E999,F821,F822,F823 --show-source --statistics

./predict.py:38:1: E999 SyntaxError: can't assign to operator
+network, _ = model_builder.build_model(args.model, net_input=net_input,
^
./models/ICNet.py:97:29: F821 undefined name 'resnet_v2'
        with slim.arg_scope(resnet_v2.resnet_arg_scope(weight_decay=weight_decay)):
                            ^
./models/ICNet.py:98:40: F821 undefined name 'resnet_v2'
            logits_32, end_points_32 = resnet_v2.resnet_v2_50(inputs_4, is_training=is_training, scope='resnet_v2_50')
                                       ^
./models/ICNet.py:99:40: F821 undefined name 'resnet_v2'
            logits_16, end_points_16 = resnet_v2.resnet_v2_50(inputs_2, is_training=is_training, scope='resnet_v2_50')
                                       ^
./models/ICNet.py:100:38: F821 undefined name 'resnet_v2'
            logits_8, end_points_8 = resnet_v2.resnet_v2_50(inputs_1, is_training=is_training, scope='resnet_v2_50')
                                     ^
./models/ICNet.py:105:29: F821 undefined name 'resnet_v2'
        with slim.arg_scope(resnet_v2.resnet_arg_scope(weight_decay=weight_decay)):
                            ^
./models/ICNet.py:106:40: F821 undefined name 'resnet_v2'
            logits_32, end_points_32 = resnet_v2.resnet_v2_101(inputs_4, is_training=is_training, scope='resnet_v2_101')
                                       ^
./models/ICNet.py:107:40: F821 undefined name 'resnet_v2'
            logits_16, end_points_16 = resnet_v2.resnet_v2_101(inputs_2, is_training=is_training, scope='resnet_v2_101')
                                       ^
./models/ICNet.py:108:38: F821 undefined name 'resnet_v2'
            logits_8, end_points_8 = resnet_v2.resnet_v2_101(inputs_1, is_training=is_training, scope='resnet_v2_101')
                                     ^
./models/ICNet.py:113:29: F821 undefined name 'resnet_v2'
        with slim.arg_scope(resnet_v2.resnet_arg_scope(weight_decay=weight_decay)):
                            ^
./models/ICNet.py:114:40: F821 undefined name 'resnet_v2'
            logits_32, end_points_32 = resnet_v2.resnet_v2_152(inputs_4, is_training=is_training, scope='resnet_v2_152')
                                       ^
./models/ICNet.py:115:40: F821 undefined name 'resnet_v2'
            logits_16, end_points_16 = resnet_v2.resnet_v2_152(inputs_2, is_training=is_training, scope='resnet_v2_152')
                                       ^
./models/ICNet.py:116:38: F821 undefined name 'resnet_v2'
            logits_8, end_points_8 = resnet_v2.resnet_v2_152(inputs_1, is_training=is_training, scope='resnet_v2_152')
                                     ^
./models/ICNet.py:126:16: F821 undefined name 'PyramidPoolingModule'
    block_32 = PyramidPoolingModule(end_points_32['pool3'], feature_map_shape=feature_map_shape, pooling_type=pooling_type)
               ^
./models/ICNet.py:128:33: F821 undefined name 'psp_32'
    out_16, block_16 = CFFBlock(psp_32, end_points_16['pool3'])
                                ^
./models/ICNet.py:137:44: F821 undefined name 'out_final'
    net = tf.concat([out_16, out_8, out_4, out_final])
                                           ^
1     E999 SyntaxError: can't assign to operator
15    F821 undefined name 'resnet_v2'
16
GeorgeSeif commented 6 years ago

Nice catch for the predict.py function!

ICNet is currently not supported, but I've kept the old code in there in case I integrate it later. Leaving the old stuff for now until I return to it

Thanks! :)