From 216efb446f150a6e6d9e8e6687e363e69af9e90b Mon Sep 17 00:00:00 2001 From: wyg <1505116161@qq.com> Date: Thu, 8 Aug 2019 14:56:03 +0800 Subject: [PATCH] [verify] add data source in readme --- reproduction/text_classification/README.md | 7 +++++++ reproduction/text_classification/train_char_cnn.py | 2 +- 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/reproduction/text_classification/README.md b/reproduction/text_classification/README.md index 8bdfb9fe..96ea7a10 100644 --- a/reproduction/text_classification/README.md +++ b/reproduction/text_classification/README.md @@ -11,6 +11,13 @@ LSTM+self_attention:论文链接[A Structured Self-attentive Sentence Embedding] AWD-LSTM:论文链接[Regularizing and Optimizing LSTM Language Models](https://arxiv.org/pdf/1708.02182.pdf) +#数据集来源 +IMDB：http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz +SST-2：https://firebasestorage.googleapis.com/v0/b/mtl-sentence-representations.appspot.com/o/data%2FSST-2.zip?alt=media&token=aabc5f6b-e466-44a2-b9b4-cf6337f84ac8 +SST：https://nlp.stanford.edu/sentiment/trainDevTestTrees_PTB.zip +yelp_full:https://drive.google.com/drive/folders/0Bz8a_Dbh9Qhbfll6bVpmNUtUcFdjYmF2SEpmZUZUcVNiMUw1TWN6RDV3a0JHT3kxLVhVR2M +yelp_polarity:https://drive.google.com/drive/folders/0Bz8a_Dbh9Qhbfll6bVpmNUtUcFdjYmF2SEpmZUZUcVNiMUw1TWN6RDV3a0JHT3kxLVhVR2M + # 数据集及复现结果汇总 使用fastNLP复现的结果vs论文汇报结果(/前为fastNLP实现，后面为论文报道,-表示论文没有在该数据集上列出结果) diff --git a/reproduction/text_classification/train_char_cnn.py b/reproduction/text_classification/train_char_cnn.py index 0b8fc535..3482de70 100644 --- a/reproduction/text_classification/train_char_cnn.py +++ b/reproduction/text_classification/train_char_cnn.py @@ -203,7 +203,7 @@ callbacks.append( def train(model,datainfo,loss,metrics,optimizer,num_epochs=100): trainer = Trainer(datainfo.datasets['train'], model, optimizer=optimizer, loss=loss(target='target'),batch_size=ops.batch_size, metrics=[metrics(target='target')], dev_data=datainfo.datasets['test'], device=[0,1,2], check_code_level=-1, - n_epochs=num_epochs) + n_epochs=num_epochs,callbacks=callbacks) print(trainer.train())