Comments (14)
I cannot help you with the information you provided. please always patse your commsnds.
from bertsum.
problem1:
the commend for train classifier is:
python train.py -mode train -encoder classifier -dropout 0.1 -bert_data_path /home/test/WangHN/BertSum-master/bert_data/cnndm -model_path /home/test/WangHN/BertSum-master/models/bert_classifier -lr 2e-3 -visible_gpus 0,1,2 -gpu_ranks 0,1,2 -world_size 3 -report_every 50 -save_checkpoint_steps 1000 -batch_size 3000 -decay_method noam -train_steps 50000 -accum_count 2 -log_file /home/test/WangHN/BertSum-master/logs/bert_classifier -use_interval true -warmup_steps 10000
for test classifier is:
python train.py -mode validate -bert_data_path /home/test/WangHN/BertSum-master/bert_data/cnndm -model_path /home/test/WangHN/BertSum-master/models/bert_classifier -visible_gpus 1 -gpu_ranks 0 -batch_size 30000 -log_file /home/test/WangHN/BertSum-master/logs/Evaluation/bert_classifier -result_path /home/test/WangHN/BertSum-master/results/classifier/cnndm -test_all -block_trigram true
the log file's results are what I provide above , the rouge score is low, this is also true in another (rnn and transformer)
problem2:
for the summary results, how can I find the original article ,for example ref.83.txt and can.83.txt's original article?
from bertsum.
could you run -mode lead to see if the data is correct
from bertsum.
how to do this
from bertsum.
python train.py -mode lead -bert_data_path ../../../bertabs/bert_data/cnndm -visible_gpus -1 -batch_size 30000 -log_file cnndm_lead -result_path ../results/cnndm_lead -block_trigram false
from bertsum.
dont have the mode lead.
I also think the problem may be model training, need I retraining the model?
from bertsum.
please checkout to dev branch and run lead
from bertsum.
sorry,where is dev branch
from bertsum.
please google for "check out branches on github"
from bertsum.
I run mode -lead ,got this results, .
from bertsum.
apparently the data is wrong. how do you process the data?
from bertsum.
I process the data by the commend you provide,
Put the CNN story file and the daily mail story file under one file,using the Stanford CoreNLP version:3.7.0
then got the sentence splitting and token, just follow the step 3,4,5 , then got the bert_data file.
from bertsum.
please past some parts of the json file in got from format_to_lines
from bertsum.
or could you rerun the preprocessing process and calculate lead
from bertsum.
Related Issues (20)
- How to implement Bert + CNN
- How to enter a sentence for prediction
- Question for models/trainer.py#L325 ? HOT 3
- How to train TransformerExt baseline?
- Package Requirements - versions? HOT 1
- Hi i wonder if i want to do a multi-classification task, what should i change?
- How to continue training from previous checkpoints? HOT 4
- I got xent of 3-4 after 30000 training steps, is this normal? HOT 1
- Can someone please explain the validation metrics for me?
- Format of the .story files
- Help needed when processing my own dataset for testing.
- Questions about imbalance sentence distribution of label 1 and label 0 when training
- [CLS] similar context vector on Evaluation
- FileNotFoundError HOT 1
- Extractive Setting? HOT 2
- I also got slightly Lower Rouge score for the same code
- Order inconsistency of output candidate file with original test.json when testing bertSum Extractive HOT 1
- expected mask dtype to be Bool but got Long
- default batch_size is 3000, I don't quite understand, why so huge? HOT 1
- Training in Colab (CNN/DM)
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 bertsum.