thudm/cogqa issues and pull requests

#46 - 感觉这代码没法复现

Issue - State: open - Opened by BA826974 7 months ago

#45 - System 1 with roberta?

Issue - State: open - Opened by truongnh112 over 1 year ago

#44 - Bump redis from 3.2.1 to 4.4.4

Pull Request - State: open - Opened by dependabot[bot] over 1 year ago
Labels: dependencies

#43 - Bump redis from 3.2.1 to 4.5.3

Pull Request - State: closed - Opened by dependabot[bot] over 1 year ago - 1 comment
Labels: dependencies

#42 - 文件夹里enwiki-20171001-pages-meta-current-withlinks-abstracts应该有什么文件？

Issue - State: open - Opened by 1024-program about 2 years ago

#41 - Bump ujson from 1.35 to 5.4.0

Pull Request - State: open - Opened by dependabot[bot] over 2 years ago
Labels: dependencies

#40 - Bump ujson from 1.35 to 5.2.0

Pull Request - State: closed - Opened by dependabot[bot] over 2 years ago - 1 comment
Labels: dependencies

#39 - Where can i get `hotpot_train_v1.1.json`?

Issue - State: closed - Opened by daniildiveev over 3 years ago - 1 comment

#38 - 请问训练速度和什么有关呀？显卡？内存？

Issue - State: open - Opened by ShaoaAllen almost 4 years ago

#37 - model1 = BertForMultiHopQuestionAnswering.from_pretrained(BERT_MODEL, cache_dir=PYTORCH_PRETRAINED_BERT_CACHE / 'distributed_{}'.format(-1))

Issue - State: open - Opened by ShaoaAllen almost 4 years ago - 2 comments

#36 - model2 = CognitiveGNN(model1.config.hidden_size) 'NoneType' object has no attribute 'config'

Issue - State: open - Opened by ShaoaAllen almost 4 years ago

#35 - 将bert换成albert时，加载输入数据时出了个错误

Issue - State: open - Opened by ShaoaAllen almost 4 years ago - 2 comments

#34 - NEG TOO LONG!

Issue - State: closed - Opened by ZuoZicheng almost 4 years ago - 1 comment

#33 - where can i find hotpot_train_v1.1_refined.json

Issue - State: closed - Opened by qwzlh almost 4 years ago - 1 comment

#32 - 不好意思一次性问了很多细节问题，希望您有时间可以解答下

Issue - State: closed - Opened by ditingdapeng about 4 years ago - 1 comment

#32 - 不好意思一次性问了很多细节问题，希望您有时间可以解答下

Issue - State: closed - Opened by ditingdapeng about 4 years ago - 1 comment

#31 - 代码中utils中的GENERAL_WD解释

Issue - State: closed - Opened by ditingdapeng about 4 years ago - 1 comment

#30 - 数据集中的答案并不是来源于该条数据的材料--问题

Issue - State: closed - Opened by ditingdapeng about 4 years ago - 1 comment

#29 - sem[x,Q,clues]在系统一和系统二中的作用

Issue - State: open - Opened by ditingdapeng about 4 years ago - 5 comments

#28 - 论文中"提取一跳节点，但是不计算语义向量"的含义

Issue - State: closed - Opened by ditingdapeng about 4 years ago - 2 comments

#27 - 请问下"hotpot_dev_fullwiki_v1_merge.json"文件从哪来？

Issue - State: closed - Opened by ditingdapeng about 4 years ago - 2 comments

#26 - 关于bert使用不加载链接，而是通过文件加载的问题

Issue - State: closed - Opened by ditingdapeng about 4 years ago - 1 comment

#25 - fullwiki data

Issue - State: closed - Opened by Two222 about 4 years ago - 1 comment

#24 - RuntimeError: CUDA error: an illegal memory access was encountered

Issue - State: open - Opened by ditingdapeng about 4 years ago - 12 comments

#23 - dump.rdb

Issue - State: open - Opened by bultiful over 4 years ago - 4 comments

#22 - Where can I get “wikipedia documents” ?

Issue - State: closed - Opened by bultiful over 4 years ago - 2 comments

#21 - where can i get the dataset file 'enwiki-20171001-pages-meta-current-withlinks-abstracts'?

Issue - State: closed - Opened by chisato-lycoreco over 4 years ago - 1 comment

#20 - use rapidfuzz instead of fuzzywuzzy

Pull Request - State: closed - Opened by maxbachmann over 4 years ago

#19 - where is the dump.pkl ?

Issue - State: closed - Opened by INNOCENT-BOY almost 5 years ago - 1 comment

#18 - Possibility of training on 1 gpu?

Issue - State: closed - Opened by dogydev almost 5 years ago - 10 comments

#17 - 关于训练结果

Issue - State: closed - Opened by zhawhjw about 5 years ago - 3 comments

#16 - 您好，关于段落抽取的问题。

Issue - State: closed - Opened by modongmoqiu about 5 years ago - 4 comments

#15 - About the framework of GNN

Issue - State: closed - Opened by eyuansu62 about 5 years ago - 2 comments

#14 - where can I find 'enwiki-20171001-pages-meta-current-withlinks-abstracts'

Issue - State: closed - Opened by slobLX about 5 years ago - 1 comment

#13 - About dump.pkl

Issue - State: closed - Opened by zycdev about 5 years ago - 2 comments

#12 - A question about the "process_train.py"

Issue - State: closed - Opened by shuxiaobo about 5 years ago - 6 comments

#11 - Where can find the file "enwiki-20171001-pages-meta-current-withlinks-abstracts" mentioned in "read_fullwiki.py"?

Issue - State: closed - Opened by Tesla-jiang about 5 years ago - 4 comments

#10 - ans_loss and hop_loss became 'nan'

Issue - State: closed - Opened by michaelmoju about 5 years ago - 1 comment

#9 - Self-cycle in gold-only cognitive graph for comparison question

Issue - State: closed - Opened by zycdev about 5 years ago - 10 comments

#8 - The low results

Issue - State: closed - Opened by JerryZeyu over 5 years ago - 12 comments

#7 - Where is the fullwiki_input_improved_by_cogqa1hop.zip?

Issue - State: closed - Opened by JerryZeyu over 5 years ago - 3 comments

#6 - Update read_fullwiki.py

Pull Request - State: closed - Opened by ronakice over 5 years ago

#5 - [About BertEmbeddingsPlus] Adding extra embedding without changing encoder

Issue - State: open - Opened by Vimos over 5 years ago

#4 - About process_train.py

Issue - State: closed - Opened by Yarkona over 5 years ago - 2 comments

#3 - Add blog post (in Chinese) link in README

Issue - State: closed - Opened by qibinc over 5 years ago

#2 - About process_train.py

Issue - State: closed - Opened by tomtang110 over 5 years ago - 3 comments

#1 - Update process_train.ipynb

Pull Request - State: closed - Opened by wunaidev over 5 years ago

GitHub / thudm/cogqa issues and pull requests