search

stanleylsx / entity_extractor_by_ner

基于Tensorflow2.3开发的NER模型，都是CRF范式，包含Bilstm(IDCNN)-CRF、Bert-Bilstm(IDCNN)-CRF、Bert-CRF，可微调预训练模型，可对抗学习，用于命名实体识别，配置后可直接运行。
393 stars 74 forks source link

ValueError: Tensor conversion requested dtype int32 for Tensor with dtype int64: <tf.Tensor 'targets:0' shape=(32, 300) dtype=int64> #11

Closed dongkcs closed 3 years ago

dongkcs commented 3 years ago

用下载的代码源码改为训练模式训练，报错 ValueError: in user code:

/home/ml/entity_extractor_by_ner/engines/model.py:34 call  *

ValueError: Tensor conversion requested dtype int32 for Tensor with dtype int64: <tf.Tensor 'targets:0' shape=(32, 300) dtype=int64>

stanleylsx commented 3 years ago

用下载的代码源码改为训练模式训练，报错 ValueError: in user code:

/home/ml/entity_extractor_by_ner/engines/model.py:34 call  *

ValueError: Tensor conversion requested dtype int32 for Tensor with dtype int64: <tf.Tensor 'targets:0' shape=(32, 300) dtype=int64>

已修复

guo-xiaomeng commented 1 year ago

同样遇到这个问题，没有解决。。。