使用Hugging face获取bert预训练模型,用于文本分类
“我爱你宝贝”文本处理
#导入Hugging face提供的transformers预训练模型库
from transformers import AutoTokenizer,TFAutoModel
#引入编码器和编码器模型
tokenizer = AutoTokenizer.from_pretrained("bert-base-chinese")
model = TFAutoModel.from_pretrained("bert-base-chinese")
#使用encoding获取token
input_ids = tokenizer.encode('我爱你宝贝',return_tensors='tf')
print(input_ids)
print('\n')
#获取三个不同的token表示
inputs = tokenizer('我爱你宝贝',return_tensors='tf')
print(inputs)
print('\n')
#我爱你宝贝的embedding值
embedding = model(input_ids)
print(embedding)
其中embedding层的理解:添加链接描述
输出结果:
tf.Tensor([[ 101 2769 4263 872 2140 6564 102]], shape=(1, 7), dtype=int32)
{'input_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[ 101, 2769, 4263, 872, 2140, 6564, 102]])>, 'token_type_ids': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[0, 0, 0, 0, 0, 0, 0]])>, 'attention_mask': <tf.Tensor: shape=(1, 7), dtype=int32, numpy=array([[1, 1, 1, 1, 1, 1, 1]])>}
TFBaseModelOutputWithPoolingAndCrossAttentions(last_hidden_state=<tf.Tensor: shape=(1, 7, 768), dtype=float32, numpy=
array([[[ 0.22551674, 0.2334234 , -0.0965183 , ..., 0.23391467,
0.12448315, -0.44088104],
[ 0.09874093, -0.05070332, 0.1096081 , ..., -0.63226104,
-0.36233944, -0.0175628 ],
[ 0.781981 , -0.41929916, -1.0031092 , ..., 0.3007508 ,
0.51924443, -0.3144598 ],
...,
[ 0.45716444, -0.89420587, -0.7758167 , ..., -0.3758897 ,
0.59490716, -0.5254687 ],
[ 0.5230568 , 0.04452747, -0.6349134 , ..., -0.1745303 ,
-0.36457115, -0.2131085 ],
[-0.37484476, -0.09532013, -0.22620994, ..., 0.09939153,
0.02245922, -0.5798783 ]]], dtype=float32)>, pooler_output=<tf.Tensor: shape=(1, 768), dtype=float32, numpy=
array([[ 0.99951375, 0.9999585 , 0.9992242 , 0.9521314 , 0.6772908 ,
0.95550525, -0.8690266 , -0.8966604 , 0.9862441 , -0.99944973,
0.9999984 , 0.99952954, -0.6903689 , -0.94052505, 0.999907 ,
-0.99993294, -0.48965782, 0.99795926, 0.98542386, -0.41674206,
0.99983186, -0.99999464, -0.9364351 , -0.00638309, 0.30202952,
0.9982603 , 0.95836174, -0.85390073, -0.999881 , 0.99924105,
0.9449711 , 0.999661 , 0.77207 , -0.9999498 , -0.99956626,
0.83075035, 0.52835643, 0.98546475, -0.42761192, -0.88479286,
-0.95826435, -0.7290919 , -0.57517326, -0.9991556 , -0.9649854 ,
0.8852386 , -1. , -0.9999767 , -0.0638053 , 0.99986666,
-0.9731981 , -0.99992514, 0.9107537 , -0.6539564 , -0.59388995,
0.989578 , -0.99978274, 0.9870353 , 1. , 0.8631872 ,
0.9998498 , -0.969598 , 0.13232271, -0.999937 , 0.99999726,
-0.99966043, -0.9971533 , 0.3563019 , 0.99992895, 0.9999971 ,
-0.9614805 , 0.9980497 , 0.99999785, 0.69701064, 0.9113608 ,
0.9994441 , -0.9994306 , 0.79707325, -0.9999976 , 0.30055138,
0.9999992 , 0.9980529 , -0.98713857, 0.9572292 , -0.9946065 ,
-0.999979 , -0.9990661 , 0.9999576 , -0.43093085, 0.99787974,
0.9981298 , -0.99936944, -0.999999 , 0.999266 , -0.9997233 ,
-0.9996603 , -0.20400661, 0.9990691 , -0.29508907, -0.8908759 ,
-0.86387736, 0.7382464 , -0.99878937, -0.9980564 , 0.9502611 ,
0.9989777 , 0.8173981 , -0.9997744 , 0.99990267, 0.78615505,
-1. , -0.95035785, -0.9999165 , -0.8321738 , -0.99752355,
0.9999309 , 0.81204945, -0.1566339 , 0.99993634, -0.9978586 ,
0.97382927, -0.998567 , -0.8548095 , -0.3403496 , 0.9902874 ,
0.9999813 , 0.9991445 , -0.9954823 , 0.99946445, 0.99999285,
0.9831586 , 0.9952379 , -0.9989788 , 0.989922 , 0.7232165 ,
-0.97291386, -0.53481036, -0.9572743 , 0.99999964, 0.9920743 ,
0.9897406 , -0.9922357 , 0.9994006 , -0.9986608 , 0.9999755 ,
-0.9999974 , 0.9997875 , -0.99999416, -0.9994994 , 0.99797654,
0.95532805, 0.9999995 , -0.8909783 , 0.9999773 , -0.99596953,
-0.9998977 , 0.9993657 , 0.40027827, 0.9921006 , -0.9999804 ,
0.98359376, -0.03588968, -0.5064751 , -0.32479793, -1. ,
0.9999594 , -0.90203744, 0.9999972 , 0.9954126 , -0.9955841 ,
-0.9808943 , -0.9968977 , 0.66703343, -0.99971366, -0.96350837,
0.99084705, -0.2077221 , 0.99952465, -0.7488135 , -0.9851071 ,
0.87106735, -0.88655657, -0.9999118 , 0.98245955, -0.7387749 ,
0.90645766, 0.28922993, 0.62658584, 0.984331 , 0.8896224 ,
-0.85062736, 0.9999931 , 0.6364938 , 0.9975679 , 0.9993347 ,
0.1677256 , -0.7637223 , -0.9097002 , -0.99999416, -0.89005023,
0.9999735 , -0.7679188 , -0.99959093, 0.8759577 , -0.9999887 ,
0.8598826 , 0.673738 , 0.19542514, -0.9997235 , -0.9999812 ,
0.9999633 , -0.9919162 , -0.999416 , 0.13023168, -0.4842012 ,
-0.04413388, -0.9995727 , 0.80254287, 0.9825549 , -0.1383966 ,
0.9795868 , -0.8842249 , -0.99779236, 0.9992252 , -0.9800692 ,
0.7213256 , 0.9732129 , 0.99999905, 0.98257124, -0.89109504,
-0.2989888 , 0.99994105, 0.62901574, -1. , 0.9641136 ,
-0.9954345 , -0.48818251, 0.9999681 , -0.99599165, 0.8507665 ,
0.9999939 , 0.9715773 , 1. , 0.23156907, -0.99945706,
-0.9995524 , 0.99999917, 0.96865034, 0.9999502 , -0.99948746,
-0.99917036, 0.42960355, -0.73817587, -0.9999985 , -0.999202 ,
-0.3500754 , 0.99888885, 0.99999183, 0.36791554, -0.9992521 ,
-0.9908179 , -0.9998418 , 0.99999774, -0.98939997, 0.99997765,
0.9902801 , -0.99194014, -0.9611848 , 0.55937403, -0.8523637 ,
-0.9996996 , 0.79383516, -0.9998934 , -0.9933539 , -0.99996805,
0.9318411 , -0.99970216, -0.9999995 , 0.96765715, 0.9999861 ,
0.8565581 , -0.9999471 , 0.9998005 , 0.99741864, -0.7784989 ,
-0.99968934, 0.8392054 , -0.9999996 , 0.9999998 , -0.9994854 ,
0.93166095, -0.8005207 , -0.9949263 , -0.41130367, 0.9995809 ,
0.99713916, -0.9951959 , -0.18618147, -0.99500746, -0.99870086,
-0.7509314 , 0.92250127, -0.80903697, 0.49754295, -0.9628656 ,
-0.9807618 , 0.891106 , -0.9792823 , -0.9998097 , 0.7365432 ,
0.99999714, -0.9586837 , 1. , 0.9671251 , 0.9999995 ,
0.936821 , -0.9994404 , 0.999575 , 0.70637655, -0.860012 ,
-0.99922866, -0.9715462 , 0.94944346, 0.49256256, -0.18733643,
-0.99984694, 0.99992305, 0.9888529 , 0.93001896, 0.8485084 ,
-0.41235867, -0.25817376, 0.96832037, -0.9961326 , 0.99835235,
-0.9997181 , -0.9917206 , 0.9997797 , 0.9999836 , 0.998336 ,
0.63495827, -0.9512613 , 0.9934805 , -0.9985411 , 0.99970275,
-0.9998071 , 0.9992197 , -0.9934266 , 0.1454727 , -0.92392623,
-0.99633276, 0.999995 , 0.9878752 , -0.42107308, 0.9998178 ,
-0.9878549 , 0.9945993 , 0.9835186 , 0.9953428 , 0.96659714,
0.98338646, 0.99998194, -0.99612033, -0.9935541 , -0.92737514,
-0.99712515, -0.9990756 , -0.9999986 , 0.58737713, -0.9984052 ,
-0.9961222 , -0.6528244 , 0.7096545 , 0.95586807, -0.38046107,
0.74068713, 0.473657 , 0.6141841 , -0.20215689, 0.5548849 ,
0.98438865, -0.9971496 , -0.9946532 , -0.99999297, -0.9993225 ,
0.94200975, 0.9998952 , -0.9998909 , 0.99921596, -0.9999831 ,
-0.9985282 , 0.99882054, -0.79864633, -0.8544362 , 0.99968654,
-0.9999791 , 0.92368674, 0.9997159 , 1. , 0.9998181 ,
0.9999495 , -0.87195814, -0.9999577 , -0.9986795 , -0.99996364,
-0.99999803, -0.999907 , 0.92731327, 0.75620824, -0.99999505,
-0.7328234 , 0.99145705, 0.999997 , 0.988811 , -0.99976283,
-0.77284473, -0.9995287 , -0.99966085, 0.9989632 , -0.9446032 ,
-0.99988997, 0.99617827, -0.05445317, 0.9999809 , -0.7030514 ,
0.92149156, 0.8852448 , 0.9638466 , 0.99250394, -0.9999869 ,
0.6906826 , 0.9999937 , 0.92754287, -0.9999963 , -0.9679839 ,
-0.78732204, -0.99995714, -0.70257646, 0.77674127, 0.9999505 ,
-0.9999676 , -0.833869 , -0.99814373, 0.92392987, 0.99506813,
0.99963397, 0.9997788 , 0.87824357, 0.8918311 , 0.9900744 ,
0.32406652, 0.99987346, 0.4580068 , -0.998568 , 0.99954724,
-0.66389877, 0.63000214, -0.9998998 , 0.99870896, 0.8527977 ,
0.99994916, 0.9983771 , 0.02654447, -0.9966444 , -0.9209089 ,
0.9962621 , 0.9999987 , -0.99568087, -0.9748604 , -0.99959505,
-0.9998028 , -0.99858516, -0.9630988 , -0.23769446, -0.99604124,
-0.9997537 , 0.71335334, 0.8423182 , 0.9999992 , 0.9999803 ,
0.99913895, -0.979068 , -0.9597613 , 0.9927908 , 0.33670604,
0.9602235 , -0.857165 , -0.99999964, -0.9990654 , -0.9995447 ,
0.9999045 , 0.23595741, -0.77770877, -0.8659783 , 0.09423243,
0.93034405, -0.99995977, -0.63107926, -0.99041086, 0.8261902 ,
0.9999982 , -0.9980706 , 0.9996412 , -0.99944156, 0.8404289 ,
0.74455214, 0.9685205 , 0.99934334, -0.3510783 , 0.22488539,
-0.74096036, 0.9169256 , 0.91471493, 0.99873304, -0.98465353,
0.56764007, 0.9996666 , -0.9769553 , 0.99997365, 0.42663586,
0.8431777 , 0.85645 , 0.9999884 , 0.7255782 , 0.9998816 ,
0.9856115 , 0.9999716 , 0.9999753 , -0.99225116, 0.753374 ,
0.24214263, -0.9915894 , -0.81609374, 0.94391257, 0.99974936,
0.53414357, -0.99619836, -0.99977595, 0.9915747 , 0.9998634 ,
1. , -0.78952557, 0.9991585 , -0.58179784, 0.9320144 ,
0.91929686, 0.83287185, 0.2685785 , 0.78830725, 0.99521255,
0.999661 , -0.9999786 , -0.9999019 , -1. , 0.9999993 ,
0.99958885, -0.7078647 , -0.9999997 , 0.9996776 , -0.4789279 ,
0.80940926, 0.99401075, 0.4332002 , -0.9423578 , 0.91720015,
-0.999896 , 0.23369907, 0.85708565, 0.93265694, 0.47120795,
0.9996671 , -0.9998968 , -0.09277818, 0.99999547, -0.49733144,
0.9999703 , 0.47026387, -0.9984143 , 0.999415 , -0.9957429 ,
-0.99995595, -0.47652856, 0.9999315 , 0.99897647, -0.0714012 ,
-0.660234 , 0.9999095 , -0.99962586, 0.9999663 , -0.99998087,
0.35069603, -0.99974513, 0.9999557 , -0.9897224 , -0.9954708 ,
-0.92215693, 0.45113984, 0.97567034, -0.863757 , 0.9999685 ,
0.0522829 , -0.943007 , 0.45604616, -0.948004 , -0.99760824,
-0.98861593, 0.5688474 , -0.9999684 , 0.7760274 , -0.3933425 ,
-0.9369243 , -0.9757494 , -0.99997884, 0.99997 , -0.7008528 ,
-0.99214804, 0.9999752 , -0.98768944, -0.99999917, 0.9772849 ,
-0.99743456, 0.12691952, 0.9924299 , 0.66166747, 0.31455383,
-0.9999932 , 0.71131855, 0.9999092 , -0.9988719 , -0.9272292 ,
-0.95572054, -0.9974266 , 0.9393952 , 0.999705 , 0.87913877,
-0.55481017, 0.88223374, 0.9862479 , 0.8830329 , 0.50108755,
0.68632865, -0.99755585, -0.9988226 , -0.9862296 , -0.9964491 ,
-0.99990624, -0.9999877 , 0.9999996 , 0.9997305 , 0.9999743 ,
-0.35132548, -0.7829424 , 0.95596844, 0.9905246 , -0.99990946,
-0.8540817 , 0.8835654 , 0.8612701 , -0.2369774 , -0.99967813,
-0.40206575, -0.999984 , -0.98034614, 0.60247403, -0.96062344,
0.59596944, 0.999995 , 0.99994415, -0.999774 , -0.99862605,
-0.9982397 , -0.99851394, 0.9999655 , 0.99967384, 0.99990207,
-0.93322766, -0.7911631 , 0.9989075 , -0.94033885, 0.00478262,
-0.99875724, -0.99710125, -0.9999525 , 0.965285 , -0.99473554,
-0.99982005, 0.9963372 , 0.99995404, 0.81072396, -0.9999799 ,
-0.62058717, 0.9999424 , 0.9989505 , 0.99999917, 0.43074998,
0.99993634, -0.99659646, 0.9925494 , -0.9946671 , 0.99999356,
-0.9999964 , 0.9999916 , 0.9999731 , 0.9993531 , 0.9986887 ,
-0.9972332 , 0.91004014, -0.9830419 , -0.6273834 , 0.9469267 ,
-0.64788556, -0.99702114, 0.30507144, 0.9915664 , -0.97477734,
0.99999803, 0.52740014, 0.4617958 , 0.88484925, 0.20282666,
0.999112 , 0.6436424 , -0.9994612 , 0.976302 , 0.99895203,
0.99344224, 0.99999726, 0.9824398 , 0.9999667 , -0.9545235 ,
-0.9966018 , 0.99578434, -0.77346295, 0.37438408, -0.9998523 ,
0.99999285, 0.9999461 , -0.9998808 , -0.9971094 , 0.4471147 ,
0.93076485, 0.9999655 , 0.99595094, 0.99771625, 0.9032935 ,
-0.06329969, 0.99881697, -0.9992911 , 0.9953064 , -0.99379414,
-0.99307436, 0.99998784, -0.9926064 , 0.99973637, -0.994869 ,
0.99974704, -0.996418 , 0.98057634, 0.9941156 , 0.9678984 ,
-0.98769575, 0.99999505, 0.49111098, -0.99909806, -0.99963087,
-0.99700135, -0.97970676, 0.80996 ]], dtype=float32)>, past_key_values=None, hidden_states=None, attentions=None, cross_attentions=None)