search

Determined22 / zh-NER-TF

A very simple BiLSTM-CRF model for Chinese Named Entity Recognition 中文命名实体识别 (TensorFlow)
2.33k stars 937 forks source link

Incompatible shapes: [64,98] vs. [64,2997] #117

Open Joe-zsc opened 4 years ago

Joe-zsc commented 4 years ago

when I am training my own dataset(english), a bug occurred my dataset have 24 tags

发生异常: InvalidArgumentError Incompatible shapes: [64,98] vs. [64,2997] [[Node: add_1 = Add[T=DT_INT32, _device="/job:localhost/replica:0/task:0/gpu:0"](add, _recv_labels_0/_15)]] [[Node: rnn/All/_67 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_1046_rnn/All", tensor_type=DT_BOOL, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

Caused by op 'add_1', defined at: File "D:\Anaconda3\envs\tf\lib\runpy.py", line 193, in _run_module_as_main "main", mod_spec) File "D:\Anaconda3\envs\tf\lib\runpy.py", line 85, in _run_code exec(code, run_globals) File "c:\Users\Think.vscode\extensions\ms-python.python-2020.5.80290\pythonFiles\lib\python\debugpy\no_wheels\debugpy__main.py", line 45, in cli.main() File "c:\Users\Think.vscode\extensions\ms-python.python-2020.5.80290\pythonFiles\lib\python\debugpy\no_wheels\debugpy/..\debugpy\server\cli.py", line 430, in main run() File "c:\Users\Think.vscode\extensions\ms-python.python-2020.5.80290\pythonFiles\lib\python\debugpy\no_wheels\debugpy/..\debugpy\server\cli.py", line 267, in run_file runpy.run_path(options.target, run_name=compat.force_str("main__")) File "D:\Anaconda3\envs\tf\lib\runpy.py", line 263, in run_path pkg_name=pkg_name, script_name=fname) File "D:\Anaconda3\envs\tf\lib\runpy.py", line 96, in _run_module_code mod_name, mod_spec, pkg_name, script_name) File "D:\Anaconda3\envs\tf\lib\runpy.py", line 85, in _run_code exec(code, run_globals) File "f:\Experiments\NERexperiments\zh-NER-TF-master\main.py", line 78, in model.build_graph() File "f:\Experiments\NERexperiments\zh-NER-TF-master\model.py", line 39, in build_graph self.loss_op() File "f:\Experiments\NERexperiments\zh-NER-TF-master\model.py", line 96, in loss_op sequence_lengths=self.sequence_lengths) File "D:\Anaconda3\envs\tf\lib\site-packages\tensorflow\contrib\crf\python\ops\crf.py", line 155, in crf_log_likelihood transition_params) File "D:\Anaconda3\envs\tf\lib\site-packages\tensorflow\contrib\crf\python\ops\crf.py", line 91, in crf_sequence_score unary_scores = crf_unary_score(tag_indices, sequence_lengths, inputs) File "D:\Anaconda3\envs\tf\lib\site-packages\tensorflow\contrib\crf\python\ops\crf.py", line 182, in crf_unary_score flattened_tag_indices = array_ops.reshape(offsets + tag_indices, [-1]) File "D:\Anaconda3\envs\tf\lib\site-packages\tensorflow\python\ops\math_ops.py", line 821, in binary_op_wrapper return func(x, y, name=name) File "D:\Anaconda3\envs\tf\lib\site-packages\tensorflow\python\ops\gen_math_ops.py", line 73, in add result = _op_def_lib.apply_op("Add", x=x, y=y, name=name) File "D:\Anaconda3\envs\tf\lib\site-packages\tensorflow\python\framework\op_def_library.py", line 768, in apply_op op_def=op_def) File "D:\Anaconda3\envs\tf\lib\site-packages\tensorflow\python\framework\ops.py", line 2336, in create_op original_op=self._default_original_op, op_def=op_def) File "D:\Anaconda3\envs\tf\lib\site-packages\tensorflow\python\framework\ops.py", line 1228, in init self._traceback = _extract_stack()

InvalidArgumentError (see above for traceback): Incompatible shapes: [64,98] vs. [64,2997] [[Node: add_1 = Add[T=DT_INT32, _device="/job:localhost/replica:0/task:0/gpu:0"](add, _recv_labels_0/_15)]] [[Node: rnn/All/_67 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_1046_rnn/All", tensor_type=DT_BOOL, _device="/job:localhost/replica:0/task:0/cpu:0"]()]] File "F:\Experiments\NERexperiments\zh-NER-TF-master\model.py", line 209, in run_one_epoch feed_dict=feed_dict) File "F:\Experiments\NERexperiments\zh-NER-TF-master\model.py", line 161, in train self.run_one_epoch(sess, train, dev, self.tag2label, epoch, saver) File "F:\Experiments\NERexperiments\zh-NER-TF-master\main.py", line 88, in model.train(train=train_data, dev=test_data) # use test_data as the dev_data to see overfitting phenomena