'albert_base_ner_checkpoints/label2id.pkl 这个文件在哪

我用albert_large_zh 做了详细的对比试验　和 bert 。在NER任务上的表现是不如bert base的，　分开来看　loc per 可以做到93%左右　org 最高做到89%左右　bert 在更少的训练步数下　loc per 做到95%及以上，　org 可以做到91%. 这其中　两个模型都对照使用了　1.基础模型 2.基础模型＋crf 3.基础模型+bilstm+crf
结果都是做不过. 您这个结果中　我只看到了eval_accuracy，　请问您有没有具体看过没一种标签的准确率呢？可以放出来看看吗　谢谢

Hi,我改了一些label,讓程式變成sequence labeling的task，但是經過訓練及測試，發現預測的label檔案和token_test檔案長度不同(有缺字的情況)，不知道是什麼原因，所以來詢問版大，謝謝。


另外我有檢查過不是max_seq_length的問題

请问data的数据怎么做

您好 ， 我之前做过bert 的这个NER, 想问一下，在bert换成albert 的过程中，有没有提别需要注意改动的模型的地方，还是只要换模型文件就可以用了呢，谢谢

如果应用到专业领域，请问是否需要自己标注部分数据才能训练呢

W0428 10:19:42.609030 140066161387328 deprecation.py:323] From /data/glusterfs_sharing_04_v2/11117720/albert-chinese-ner-master/tf_metrics.py:142: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.where in 2.0, which has the same broadcast rule as np.where
ERROR:tensorflow:Error recorded from evaluation_loop: index 7 is out of bounds for axis 0 with size 7
E0428 10:19:42.708933 140066161387328 error_handling.py:75] Error recorded from evaluation_loop: index 7 is out of bounds for axis 0 with size 7
INFO:tensorflow:evaluation_loop marked as finished
I0428 10:19:42.709069 140066161387328 error_handling.py:101] evaluation_loop marked as finished
WARNING:tensorflow:Reraising captured error
W0428 10:19:42.709154 140066161387328 error_handling.py:135] Reraising captured error
Traceback (most recent call last):
File "albert_ner.py", line 847, in
tf.app.run()
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_core/python/platform/app.py", line 40, in run
_run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/absl/app.py", line 299, in run
_run_main(main, args)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/absl/app.py", line 250, in _run_main
sys.exit(main(argv))
File "albert_ner.py", line 787, in main
result = estimator.evaluate(input_fn=eval_input_fn, steps=eval_steps, checkpoint_path=filename)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 3056, in evaluate
rendezvous.raise_errors()
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/error_handling.py", line 136, in raise_errors
six.reraise(typ, value, traceback)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/six.py", line 703, in reraise
raise value
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 3051, in evaluate
name=name)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 480, in evaluate
name=name)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 522, in _actual_eval
return _evaluate()
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 504, in _evaluate
self._evaluate_build_graph(input_fn, hooks, checkpoint_path))
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1511, in _evaluate_build_graph
self._call_model_fn_eval(input_fn, self.config))
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1547, in _call_model_fn_eval
features, labels, ModeKeys.EVAL, config)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 2857, in _call_model_fn
config)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1149, in _call_model_fn
model_fn_results = self._model_fn(features=features, **kwargs)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 3126, in _model_fn
features, labels, is_export_mode=is_export_mode)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 1663, in call_without_tpu
return self._call_model_fn(features, labels, is_export_mode=is_export_mode)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 2005, in _call_model_fn
return estimator_spec.as_estimator_spec()
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 379, in as_estimator_spec
host_call_ret = _OutfeedHostCall.create_cpu_hostcall(host_calls)
File "/root/miniconda3/envs/tf1.15/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 2076, in create_cpu_hostcall
ret[name] = host_fn(*tensors)
File "albert_ner.py", line 531, in metric_fn
precision = tf_metrics.precision(label_ids, predictions, 7, [2, 3, 4, 5, 6, 7], average="macro")
File "/data/glusterfs_sharing_04_v2/11117720/albert-chinese-ner-master/tf_metrics.py", line 47, in precision
cm, pos_indices, average=average)
File "/data/glusterfs_sharing_04_v2/11117720/albert-chinese-ner-master/tf_metrics.py", line 196, in metrics_from_confusion_matrix
cm_mask[idx, :] = 1
IndexError: index 7 is out of bounds for axis 0 with size 7

有无这块的例子

想问下，跑的时候出现了训练刚开始就结束，显示train_loop marked as finished 的问题，提示说可能是input 是空， 不过感觉不是那个问题好像， 数据应该是正常的。谢谢

已经把tensorflow版本更改为1.15.0,但是还是出现这个问题呢

当把模型换成albert_tiny版本的时候，出现报错了：ValueError: Shape of variable bert/embeddings/LayerNorm/beta:0 ((312,)) doesn't match with shape of tensor bert/embeddings/LayerNorm/beta ([768]) from checkpoint reader.

ValueError: Dimension size must be evenly divisible by 1408 but is 122880 for 'loss/Reshape_1' (op: 'Reshape') with input shapes: [8192,15], [3] and with input tensors computed as partial shapes: input[1] = [?,128,11].

在 这行代码： estimator.train(input_fn=train_input_fn, max_steps=num_train_steps) 出错
在是什么原因造成的？看到类别数是动态的，没有写死？
报错内容如下：
W0226 11:32:03.210905 15596 error_handling.py:135] Reraising captured error
Traceback (most recent call last):
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\framework\ops.py", line 1607, in _create_c_op
c_op = c_api.TF_FinishOperation(op_desc)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Dimension size must be evenly divisible by 1408 but is 24576 for 'loss/Reshape_1' (op: 'Reshape') with input shapes: [4096,6], [3] and with input tensors computed as partial shapes: input[1] = [?,128,11].

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "F:/IdeaProjects/albert-chinese-ner/albert_ner.py", line 865, in
tf.app.run()
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\platform\app.py", line 40, in run
_run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\absl\app.py", line 299, in run
_run_main(main, args)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\absl\app.py", line 250, in _run_main
sys.exit(main(argv))
File "F:/IdeaProjects/albert-chinese-ner/albert_ner.py", line 748, in main
estimator.train(input_fn=train_input_fn, max_steps=num_train_steps)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\tpu\tpu_estimator.py", line 3035, in train
rendezvous.raise_errors()
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\tpu\error_handling.py", line 136, in raise_errors
six.reraise(typ, value, traceback)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\six.py", line 703, in reraise
raise value
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\tpu\tpu_estimator.py", line 3030, in train
saving_listeners=saving_listeners)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\estimator.py", line 370, in train
loss = self._train_model(input_fn, hooks, saving_listeners)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\estimator.py", line 1161, in _train_model
return self._train_model_default(input_fn, hooks, saving_listeners)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\estimator.py", line 1191, in _train_model_default
features, labels, ModeKeys.TRAIN, self.config)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\tpu\tpu_estimator.py", line 2857, in _call_model_fn
config)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\estimator.py", line 1149, in _call_model_fn
model_fn_results = self._model_fn(features=features, **kwargs)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\tpu\tpu_estimator.py", line 3126, in _model_fn
features, labels, is_export_mode=is_export_mode)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\tpu\tpu_estimator.py", line 1663, in call_without_tpu
return self._call_model_fn(features, labels, is_export_mode=is_export_mode)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_estimator\python\estimator\tpu\tpu_estimator.py", line 1994, in _call_model_fn
estimator_spec = self._model_fn(features=features, **kwargs)
File "F:/IdeaProjects/albert-chinese-ner/albert_ner.py", line 493, in model_fn
num_labels, use_one_hot_embeddings)
File "F:/IdeaProjects/albert-chinese-ner/albert_ner.py", line 452, in create_model
logits = tf.reshape(logits, [-1, FLAGS.max_seq_length, 11])
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\ops\array_ops.py", line 131, in reshape
result = gen_array_ops.reshape(tensor, shape, name)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\ops\gen_array_ops.py", line 8114, in reshape
"Reshape", tensor=tensor, shape=shape, name=name)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\framework\op_def_library.py", line 794, in _apply_op_helper
op_def=op_def)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\util\deprecation.py", line 507, in new_func
return func(*args, **kwargs)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\framework\ops.py", line 3357, in create_op
attrs, op_def, compute_device)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\framework\ops.py", line 3426, in _create_op_internal
op_def=op_def)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\framework\ops.py", line 1770, in init
control_input_ops)
File "C:\Users\root\Anaconda3\envs\albert\lib\site-packages\tensorflow_core\python\framework\ops.py", line 1610, in _create_c_op
raise ValueError(str(e))
ValueError: Dimension size must be evenly divisible by 1408 but is 24576 for 'loss/Reshape_1' (op: 'Reshape') with input shapes: [4096,6], [3] and with input tensors computed as partial shapes: input[1] = [?,128,11].

Hello, I am doing experiments on the MSRA NER dataset using albert. But after 3 epochs, the f1 score can only reach 82 or so. I use the conlleval.py script to calculate the metrics. How many epochs do you usually run for finetuning on NER dataset? Looking forward to your reply.

如题，多谢~，没有用tpu，用gpu跑的，batch 16 结果oom了，想不用多进程，8g 的gpu

测试结果同样：

[CLS]
B-LOC
I-LOC
O
B-LOC
I-LOC
I-PER
O
O
O
O
O
O
O
O
O
[SEP]
[CLS]

上述结果怎么显示出来？

您好，我的环境是tensorflow1.15.0，python 3.6，运行会报错ModuleNotFoundError: No module named 'tensorflow.python.types'，请问这个该怎么解决呢

有没有必要加上CRF呢？效果会更好一点吗？

tensorflow的版本为1.15.0，请问为什么报AttributeError: module 'tensorflow' has no attribute 'Optimizer'？谢谢

请问这个模型的predict代码提供了吗？如何进行predict呢？

你好，我用同样的参数跑了一下albert-chinese-ner和上一个bert-chinese-ner，但是跑出来的结果后者好于前者，论文上不是说albert比bert好一些吗？为什么不是这样呢？下面是我跑出来的结果：

albert-chinese-ner跑的命令和结果如下：
用的预训练向量地址：https://storage.googleapis.com/albert_zh/albert_base_zh.zip
命令：
python3 albert_ner.py --task_name ner --do_train true --do_eval true --data_dir data --vocab_file albert_config/vocab.txt --bert_config_file ../../albert_base_zh/albert_config_base.json --max_seq_length 128 --train_batch_size 64 --learning_rate 2e-5 --num_train_epochs 3 --output_dir albert_base_ner_checkpoints
结果：
INFO:tensorflow:***** Eval results albert_base_ner_checkpoints/model.ckpt-2374 *****
INFO:tensorflow: eval_f = 0.9456498
INFO:tensorflow: eval_precision = 0.95313764
INFO:tensorflow: eval_recall = 0.93832487
INFO:tensorflow: global_step = 2374
INFO:tensorflow: loss = 11.592924

bert-chinese-ner跑的命令和结果如下：
用的预训练向量地址：https://storage.googleapis.com/bert_models/2018_11_03/chinese_L-12_H-768_A-12.zip
命令：
python3 BERT_NER.py --task_name ner --do_train true --do_eval true --data_dir data --vocab_file vocab.txt --bert_config_file checkpoint/bert_config.json --init_checkpoint checkpoint/bert_model.ckpt --max_seq_length 128 --train_batch_size 64 --learning_rate 2e-5 --num_train_epochs 3 --output_dir output/result_dir/
结果：
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 2374: output/result_dir/model.ckpt-2374
INFO:tensorflow:evaluation_loop marked as finished
INFO:tensorflow:***** Eval results *****
INFO:tensorflow: eval_f = 0.96657985
INFO:tensorflow: eval_precision = 0.9670561
INFO:tensorflow: eval_recall = 0.96620923
INFO:tensorflow: global_step = 2374
INFO:tensorflow: loss = 8.526184

训练时间bert也比albert快，训练时每一个examples/sec耗时如下
albert耗时平均: INFO:tensorflow:examples/sec: 66.2794
bert耗时平均: INFO:tensorflow:examples/sec: 58.3886

为什么会出现这样的结果呢？

新数据已经改为示例数据格式，labels名称和数量更换了，文件txt也是utf-8格式，但是依然会报错，百思不得其解：
ERROR:tensorflow:Error recorded from evaluation_loop: 'utf-8' codec can't decode bytes in position 2844-2845: invalid continuation byte
E1008 14:17:13.590271 139627806422848 error_handling.py:75] Error recorded from evaluation_loop: 'utf-8' codec can't decode bytes in position 2844-2845: invalid continuation byte
INFO:tensorflow:evaluation_loop marked as finished
I1008 14:17:13.590479 139627806422848 error_handling.py:101] evaluation_loop marked as finished
WARNING:tensorflow:Reraising captured error
W1008 14:17:13.590565 139627806422848 error_handling.py:135] Reraising captured error
Traceback (most recent call last):
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1365, in _do_call
return fn(*args)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1350, in _run_fn
target_list, run_metadata)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1443, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.OutOfRangeError: End of sequence
[[{{node IteratorGetNext}}]]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "albert_ner.py", line 848, in
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/platform/app.py", line 40, in run
_run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/absl/app.py", line 312, in run
_run_main(main, args)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/absl/app.py", line 258, in _run_main
sys.exit(main(argv))
File "albert_ner.py", line 788, in main
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 3056, in evaluate
rendezvous.raise_errors()
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/error_handling.py", line 136, in raise_errors
six.reraise(typ, value, traceback)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/six.py", line 719, in reraise
raise value
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/tpu/tpu_estimator.py", line 3051, in evaluate
name=name)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 480, in evaluate
name=name)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 522, in _actual_eval
return _evaluate()
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 511, in _evaluate
output_dir=self.eval_dir(name))
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 1619, in _evaluate_run
config=self._session_config)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/training/evaluation.py", line 272, in _evaluate_once
session.run(eval_ops, feed_dict)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 754, in run
run_metadata=run_metadata)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1259, in run
run_metadata=run_metadata)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1360, in run
raise six.reraise(*original_exc_info)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/six.py", line 719, in reraise
raise value
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1345, in run
return self._sess.run(*args, **kwargs)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1418, in run
run_metadata=run_metadata)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1176, in run
return self._sess.run(*args, **kwargs)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 956, in run
run_metadata_ptr)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1180, in _run
feed_dict_tensor, options, run_metadata)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1359, in _do_run
run_metadata)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1378, in _do_call
message = error_interpolation.interpolate(message, self._graph)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/framework/error_interpolation.py", line 454, in interpolate
common_prefix = traceback_files_common_prefix(tagged_ops)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/framework/error_interpolation.py", line 358, in traceback_files_common_prefix
for frame in op.traceback:
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/framework/ops.py", line 2309, in traceback
return tf_stack.convert_stack(self._traceback)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/util/tf_stack.py", line 199, in convert_stack
return tuple(_tuple_generator())
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/site-packages/tensorflow_core/python/util/tf_stack.py", line 189, in _tuple_generator
line = linecache.getline(filename, lineno, frame.globals)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/linecache.py", line 16, in getline
lines = getlines(filename, module_globals)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/linecache.py", line 47, in getlines
return updatecache(filename, module_globals)
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/linecache.py", line 137, in updatecache
lines = fp.readlines()
File "/home/mingzhen/anaconda3/envs/albert/lib/python3.6/codecs.py", line 321, in decode
(result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf-8' codec can't decode bytes in position 2844-2845: invalid continuation byte

hello,在自己运行的时候出现
Traceback (most recent call last):
File "albert_ner.py", line 845, in
tf.app.run()
File "C:\Users\pangpang\Anaconda3\lib\site-packages\tensorflow\python\platform\app.py", line 126, in run
_sys.exit(main(argv))
File "albert_ner.py", line 671, in main
is_per_host = tf.contrib.tpu.InputPipelineConfig.PER_HOST_V2
AttributeError: module 'tensorflow.contrib.tpu' has no attribute 'InputPipelineConfig'这个问题，我的tensorflow版本是1.8.0，是tensorflow版本的问题所以才报错的吗？

tensorflow-gpu 1.15，ner任务使用了自行训练的albert，config.json内容如下 "attention_probs_dropout_prob": 0.1,
"directionality": "bidi",
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"max_position_embeddings": 512,
"num_attention_heads": 12,
"num_hidden_layers": 12,
"pooler_fc_size": 768,
"pooler_num_attention_heads": 12,
"pooler_num_fc_layers": 3,
"pooler_size_per_head": 128,
"pooler_type": "first_token_transform",
"type_vocab_size": 2,
"vocab_size": 20974
运行程序是提示AttributeError: 'BertConfig' object has no attribute 'embedding_size'，请问是什么原因？

convert_single_example这个方法里出现labels、tokens长度不一致的情况。一直报错。。

想要获取输出标签前那个用于判定的分数矩阵，类似predict_prob，是在代码哪层呢？
另外想要用sklearn里的东西，怎么将这个Albert做出个接口呢？
Many thanks in advance!!

使用自己的语料训练报错
Dimensions must be equal, but are 45 and 11 for 'loss/mul' (op: 'Mul') with input shapes: [?,128,45], [?,128,11].
是否需要把输入语料处理成小于max_seq_length的句子。

你好，我想问一下博主您是什么硬件配置训练这个模型用了多久，我的配置一般，对比下时间上看性价比如何

tf.Optimizer 是TF2.0才有的吧， optimization_finetuning.py 第87行 class AdamWeightDecayOptimizer(tf.Optimizer) 是怎么回事？ 报module 'tensorflow' has no attribute 'Optimizer'

NerProcessor类下get_labels方法的"X"标签的作用是什么？

584行
d = tf.data.Dataset.from_tensor_slices({
"input_ids":
tf.constant(
all_input_ids, shape=[num_examples, seq_length],
dtype=tf.int32),
"input_mask":
tf.constant(
all_input_mask,
shape=[num_examples, seq_length],
dtype=tf.int32),
"segment_ids":
tf.constant(
all_segment_ids,
shape=[num_examples, seq_length],
dtype=tf.int32),
"label_ids":
tf.constant(all_label_ids, shape=[num_examples], dtype=tf.int32),
})
shape=[num_examples, seq_length]，shape=[num_examples] 应该有问题，这里总是跑不通

您好，我发现您的这套代码，无法让loss值下降啊，即使使用您提供的数据集，loss值也保持在十几二十几的