multigen

multigen copied to clipboard

I noticed that the test dataset also generates 2hops_100_directed_triple_filter.json files based on source and target files, so when I want to test any sentence with the trained model, there is...

wyupupup

HI , where is "data/conceptnet_en.txt" file in your data.zip thanks

jiangchao9987

NotImplementedError

1

when i run inference code `python3 main.py \ --train_data_file ${ROOT_PATH}/data/${DATA_TYPE}/train \ --dev_data_file ${ROOT_PATH}/data/${DATA_TYPE}/dev \ --test_data_file ${ROOT_PATH}/data/${DATA_TYPE}/test \ --graph_path 2hops_100_directed_triple_filter.json \ --output_dir ${ROOT_PATH}/models/${DATA_TYPE}/grf-${DATA_TYPE} \ --source_length 32 \ --target_length 32 \ --model_type...

xiang-xiang-zhu

arch-raven

关于测试时数据处理的问题

1

作者您好， 请问当测试时，由于看不到target，“concept_label”、“gate_labels”等这些数据应该怎么处理呢？ 谢谢！

victorup

result_eptest.txt: A plane is too large to fit in a grocery store, and a plane is too large to fit in a plane. It is too large to fit in...

cyqqq

您好，请问一个不是该模型的问题，也是您曾经在goole-bert提问过的attention mask的问题。

1

在bert源码create_attention_mask_from_input_mask中，We don't assume that `from_tensor` is a mask (although it could be). We don't actually care if we attend *from* padding tokens (only *to* padding) tokens so we create a...

Yesgo1220

你好，想请问一下在论文3.2.3部分提到的计算未访问节点分数相关细节。1、在生成输出序列的每个时间步，是否计算节点分数都会从初始节点为1其他节点为0开始，还是会在上个时间步的分数基础上更新呢？2、论文4.2部分说到“Specifically, we iterate the following process for H hops: starting from the nodes in the current sub-graph (initialized by Cx) and search for the direct neighbours of each node and...

hanhun7