ValueError: Variable highway/highway_network/layer_0/kernel already exists, disallowed. Did you mean to set reuse=True in VarScope?

[yogesh-nikose]

File "/home/u19872/.conda/envs/bert01/lib/python3.6/runpy.py", line 193, in _run_module_as_main "main", mod_spec) File "/home/u19872/.conda/envs/bert01/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/u19872/bert01/bi-att-flow/basic/cli.py", line 126, in tf.app.run() File "/home/u19872/.conda/envs/bert01/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 48, in run _sys.exit(main(_sys.argv[:1] + flags_passthrough)) File "/home/u19872/bert01/bi-att-flow/basic/cli.py", line 123, in main m(config) File "/home/u19872/bert01/bi-att-flow/basic/main.py", line 23, in main _train(config) File "/home/u19872/bert01/bi-att-flow/basic/main.py", line 82, in _train models = get_multi_gpu_models(config) File "/home/u19872/bert01/bi-att-flow/basic/model.py", line 19, in get_multi_gpu_models model = Model(config, scope, rep=gpu_idx == 0) File "/home/u19872/bert01/bi-att-flow/basic/model.py", line 58, in init self._build_forward() File "/home/u19872/bert01/bi-att-flow/basic/model.py", line 128, in _build_forward xx = highway_network(xx, config.highway_num_layers, True, wd=config.wd, is_train=self.is_train) File "/home/u19872/bert01/bi-att-flow/my/tensorflow/nn.py", line 222, in highway_network input_keep_prob=input_keep_prob, is_train=is_train) File "/home/u19872/bert01/bi-att-flow/my/tensorflow/nn.py", line 210, in highway_layer gate = linear([arg], d, bias, bias_start=bias_start, scope='gate', wd=wd, input_keep_prob=input_keep_prob, is_train=is_train) File "/home/u19872/bert01/bi-att-flow/my/tensorflow/nn.py", line 91, in linear flat_out = _linear(flat_args, output_size, bias) File "/home/u19872/bert01/bi-att-flow/my/tensorflow/nn.py", line 61, in _linear initializer=kernel_initializer) File "/home/u19872/.conda/envs/bert01/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py", line 1065, in get_variable use_resource=use_resource, custom_getter=custom_getter) File "/home/u19872/.conda/envs/bert01/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py", line 962, in get_variable use_resource=use_resource, custom_getter=custom_getter) File "/home/u19872/.conda/envs/bert01/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py", line 367, in get_variable validate_shape=validate_shape, use_resource=use_resource) File "/home/u19872/.conda/envs/bert01/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py", line 352, in _true_getter use_resource=use_resource) File "/home/u19872/.conda/envs/bert01/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py", line 664, in _get_single_variable name, "".join(traceback.format_list(tb)))) ValueError: Variable highway/highway_network/layer_0/kernel already exists, disallowed. Did you mean to set reuse=True in VarScope? Originally defined at:

File "/home/u19872/bert01/bi-att-flow/my/tensorflow/nn.py", line 61, in _linear initializer=kernel_initializer) File "/home/u19872/bert01/bi-att-flow/my/tensorflow/nn.py", line 91, in linear flat_out = _linear(flat_args, output_size, bias) File "/home/u19872/bert01/bi-att-flow/my/tensorflow/nn.py", line 208, in highway_layer trans = linear([arg], d, bias, bias_start=bias_start, scope='trans', wd=wd, input_keep_prob=input_keep_prob, is_train=is_train)