Comments (9)
凭直觉看应该是在预测DRCD_test.json的时候网络没有被initialize。
init_checkpoint
在单独做predict的时候是不起作用的。
接下来:
- 你需要查看一下
${MODEL_DIR}/model_ckpt
目录下是否有xlnet的model文件。 - 查看log中是否包含类似
Restoring parameters from
这样的信息,如果没有就说明model没有被正确加载进去。
from chinese-xlnet.
from chinese-xlnet.
目前 ${MODEL_DIR}/model_ckpt
目錄底下的文件
請問該如何解決model没有被正确加載進去的問題..
from chinese-xlnet.
上面那个Estimator的问题可以忽略。你需要再看一下log中是否包含Restoring parameters from
这样的信息。
from chinese-xlnet.
from chinese-xlnet.
看似weights应该是读进去了,方便的话可以把log和prediction.json发给我看看, [email protected]
from chinese-xlnet.
已寄出,十分感謝協助!!
from chinese-xlnet.
由于命名原因,dev和test生成的tf_records/features.pkl都会命名成XLNet/proc_data/spiece.model.slen-256.qlen-64.eval.features.pkl
。
你能否改变一下--output_dir=
到其他的目录,重新预测一遍DRCD_test.json,这样可以重新生成features和tf_records,看看会有什么样的结果。
from chinese-xlnet.
成功了!
test set
EM: 84.42
F1: 92.85
十分感謝!!
from chinese-xlnet.
Related Issues (20)
- train.py HOT 1
- 你好,我用 pytorch 版本的 XLNet-base进行测试生成,未 fine-tuning,发现效果贼差,不知道怎么回事? HOT 7
- 正在训练的时候就报错,重新尝试了几次都是这个错误,不知道是代码原因还是数据原因,跪求解决 HOT 2
- 如何对chinese xlnet 蒸馏?产生小模型 HOT 1
- 相对于官方版本,中文版的xlnet对算法上有改动吗,如果有的话改动在什么地方呢? HOT 2
- 预训练时设置的mem_len=384但是下载的pytorch模型里mem_len=null HOT 4
- XLNet其实不能稳压RoBERTa吧? HOT 2
- 如何做预测 HOT 2
- 在huggingface.co的chinese-xlnet-mid预训练模型做生成任务,没有结果 HOT 2
- 你好,我使用 pytorch 版本的 XLNet 跑 baseline 二分类,效果非常差 HOT 3
- 有没有比过GPU (train_gpu.py)和TPU (train.py)版本的预训练效果 HOT 2
- 关于分词上的一点问题 HOT 5
- Performance issues in the program HOT 5
- Performance issue in src/data_utils.py (by P3) HOT 7
- 想在自己领域数据集上进行二次pretrain,正确的操作方式是什么呢? HOT 6
- 请问大佬,关于中文XLNet自回归的问题 HOT 4
- ValueError: not enough values to unpack (expected 2, got 1) HOT 2
- Feature: cls_index (data type: int64) is required but could not be found HOT 4
- 请教一下有适合的CPU推理加速的框架么? HOT 1
- ChnSentiCorp 生成的模型怎么使用?
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 chinese-xlnet.