camel2000
camel2000
https://github.com/CLUEbenchmark/CLUENER2020/tree/master/pytorch_version 中所述 “在dev上为F1分数为0.8076”， 是用的这个（https://github.com/CLUEbenchmark/CLUENER2020/blob/master/score.py）评价代码吗？ 上面说的到评价代码中根本没有考虑到位置信息，如果是这样测出来的，那你的指标“虚高”
> score.py文件中，不看位置不就成了多标签分类了吗？ > > https://github.com/CLUEbenchmark/CLUENER2020/blob/master/score.py 这个问题解决了吗？ 你跑baseline 跑的指标是多少啊？ readme 里面写的指标是错误的吗？
@FreyrS @linminhtoo @Wendysigh @jeanfeydy I have modified the code and tried to test it by batch, but it is found that when the batch_size is different, the output embedding is...
is this a gpu version ？I saw that you used an "extension" command, and this command caused the program to report an error
@dklurker did you solve this bug now？ i have the same problem
@dklurker thanks for remind the "Starting from scratch" ### conclusion： 1. gpu search doubled the speed compared with cpu version 2. I still failed in the expandaln step when i...
now， i can use gpu to build database and than retrieve MSA from input_DB.fasta, but for colabfold_envdb_202108 ,we should build the database from .fasta or just start from tsv2exprofiledb step...
@yanj14jy15 I think mmseq2 version 15 did not support GPU, how can you test it with GPU ?
@milot-mirdita i tried https://mmseqs.com/archive/8783404eab75833dcb865153ed2e146431649efa , but still failed at 'expandaln'. ### error: ``` terminate called after throwing an instance of 'std::logic_error' what(): basic_string::_S_construct null not valid /home/yakun_li_genbio_ai/project/msa/bin/../bin/msa_retrieve_pipeline.bash: line 72: 4133157...
``` PROJECT_DIR:~/project/msa/bin/.. use GPU: 1 createdb ~/project/msa/examples/one_query.fasta ~/project/msa/examples/demo/qdb Converting sequences Time for merging to qdb_h: 0h 0m 0s 496ms Time for merging to qdb: 0h 0m 0s 474ms Database type:...