Comments (7)
我也遇到了类似问题,正在排查原因。
我是在多核CPU上运行的,GPU环境不方便升级到CUDA9.0。
from bert-bilstm-crf-ner.
我遇到了类似问题,清除output里面的输出,重新运行时得到了解决
from bert-bilstm-crf-ner.
我也遇到了nan,并且我的test集人名效果奇差,不知道是哪里的问题 (我并没有改动数据):
INFO:tensorflow:Finished evaluation at 2019-01-17-15:01:59
INFO:tensorflow:Saving dict for global step 19560: eval_f = nan, eval_precision = nan, eval_recall = nan, global_step = 19560, loss = 30.386103
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 19560: /home/wei/BERT-BiLSTM-CRF-NER/output/model.ckpt-19560
INFO:tensorflow:evaluation_loop marked as finished
INFO:tensorflow:***** Predict results *****
INFO:tensorflow: eval_f = nan
INFO:tensorflow: eval_precision = nan
INFO:tensorflow: eval_recall = nan
INFO:tensorflow: global_step = 19560
INFO:tensorflow: loss = 30.386103
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Running infer on CPU
INFO:tensorflow:*** Features ***
INFO:tensorflow: name = input_ids, shape = (?, 128)
INFO:tensorflow: name = input_mask, shape = (?, 128)
INFO:tensorflow: name = label_ids, shape = (?, 128)
INFO:tensorflow: name = segment_ids, shape = (?, 128)
shape of input_ids (?, 128)
INFO:tensorflow:**** Trainable Variables ****
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /home/wei/BERT-BiLSTM-CRF-NER/output/model.ckpt-19560
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
processed 214542 tokens with 7450 phrases; found: 197725 phrases; correct: 7061.
accuracy: 10.90%; precision: 3.57%; recall: 94.78%; FB1: 6.88
LOC: precision: 95.70%; recall: 95.06%; FB1: 95.38 3441
ORG: precision: 92.09%; recall: 91.37%; FB1: 91.73 2149
PER: precision: 0.93%; recall: 98.30%; FB1: 1.84 192135
有人遇到过吗?
from bert-bilstm-crf-ner.
不知道为啥,我使用多种数据运行,都没出现过NON,能给出其他方法的baseline 再来比较么?
from bert-bilstm-crf-ner.
我也遇到了类似问题,正在排查原因。
我是在多核CPU上运行的,GPU环境不方便升级到CUDA9.0。
请问你最终解决了这个问题么,测试结果NaN
from bert-bilstm-crf-ner.
我遇到了类似问题,清除output里面的输出,重新运行时得到了解决
请问你清除了哪些output的文件
from bert-bilstm-crf-ner.
我也遇到了nan,并且我的test集人名效果奇差,不知道是哪里的问题 (我并没有改动数据):
INFO:tensorflow:Finished evaluation at 2019-01-17-15:01:59
INFO:tensorflow:Saving dict for global step 19560: eval_f = nan, eval_precision = nan, eval_recall = nan, global_step = 19560, loss = 30.386103
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 19560: /home/wei/BERT-BiLSTM-CRF-NER/output/model.ckpt-19560
INFO:tensorflow:evaluation_loop marked as finished
INFO:tensorflow:***** Predict results *****
INFO:tensorflow: eval_f = nan
INFO:tensorflow: eval_precision = nan
INFO:tensorflow: eval_recall = nan
INFO:tensorflow: global_step = 19560
INFO:tensorflow: loss = 30.386103
INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Running infer on CPU
INFO:tensorflow:*** Features ***
INFO:tensorflow: name = input_ids, shape = (?, 128)
INFO:tensorflow: name = input_mask, shape = (?, 128)
INFO:tensorflow: name = label_ids, shape = (?, 128)
INFO:tensorflow: name = segment_ids, shape = (?, 128)
shape of input_ids (?, 128)
INFO:tensorflow:**** Trainable Variables ****
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /home/wei/BERT-BiLSTM-CRF-NER/output/model.ckpt-19560
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
processed 214542 tokens with 7450 phrases; found: 197725 phrases; correct: 7061.
accuracy: 10.90%; precision: 3.57%; recall: 94.78%; FB1: 6.88
LOC: precision: 95.70%; recall: 95.06%; FB1: 95.38 3441
ORG: precision: 92.09%; recall: 91.37%; FB1: 91.73 2149
PER: precision: 0.93%; recall: 98.30%; FB1: 1.84 192135
有人遇到过吗?
@newwaylw 请问你解决了这个问题么?
from bert-bilstm-crf-ner.
Related Issues (20)
- 文本分类模型提供服务如何输出多个标签的结果
- Assign requires shapes of both tensors to match. lhs shape= [1] rhs shape= [13] [[node save/Assign_201 (defined at /usr/local/lib/python3.5/dist-packages/bert_base-0.0.9-py3.5.egg/bert_base/server/graph.py:291) ]]
- 如何进行测试呢?
- 英文预测
- 开启服务后,怎么查看服务对输入字符串的tokenize结果?
- shape of input_ids (?, 202)
- 为什么我换了学习率,最终的结果都是0呢? batch_size设大一点就OOM? HOT 4
- 文本分类返回label 1和-1 HOT 2
- adam_filter后的问题
- Errors may have originated from an input operation. 为什么会出现这个问题呢,都是按照您的步骤来的啊
- 显存几乎沾满,利用率却很低 HOT 2
- Graph is finalized and cannot be modified HOT 1
- 怎么评估模型
- 救命,为什么我换什么标注数据集,预测序列都是O,PRF都是0
- tensorflow-gpu2.6.0没法运行,并且找不到tensorflow1.12.0 HOT 2
- 为什么总是删除上一轮产出的文件,导致还没来得及进行下一步就提示找不到训练数据集了
- 训练过程具体是怎么样的
- 在训练的时候显示 Failed to find any matching files for chinese_L-12_H-768_A-12//bert_model.ckpt
- 请问预测结果和文件是要选最后输出的吗?还是自己在各个score文件中找得分最高的
- 训练时出错
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from bert-bilstm-crf-ner.