eval报错

[PhilrainV]

在使用wandb_eval.py进行lpkt模型的eval时出现报错：
Traceback (most recent call last):
File "wandb_eval.py", line 83, in
main(params)
File "wandb_eval.py", line 57, in main
atkt_pad)
File "D:\CODE_env\Anaconda\anaconda3\envs\py375\lib\site-packages\pykt\models\evaluate_model.py", line 628, in evaluate_splitpred_question
curqin, curcin, currin, curdforget, ctrues, cpreds = predict_each_group(dtotal, dcur, dforget, curdforget, is_repeat, qidx, uid, idx, model_name, model, t, end, fout, atkt_pad)
File "D:\CODE_env\Anaconda\anaconda3\envs\py375\lib\site-packages\pykt\models\evaluate_model.py", line 677, in predict_each_group
curitin = dcur["curitin"]
KeyError: 'curitin'

通过debug发现当前学生信息dcur在"curitin"列会在循环中被drop掉导致predict_each_group()无法运行

如果注释掉：
curqin, curcin, currin, curdforget, ctrues, cpreds = predict_each_group(dtotal, dcur, dforget, curdforget, is_repeat, qidx, uid, idx, model_name, model, t, end, fout, atkt_pad)
#dcur = {"curqin": curqin, "curcin": curcin, "currin": currin, "curtin": curtin}
则可正常跑通。

[PhilrainV]

另外，from pykt.models import evaluate_splitpred_question, load_model, lpkt_evaluate_multi_ahead 中没有lpkt_evaluate_multi_ahead ，其具体函数包含在下载的完整文件下pykt-toolkit-main/pykt/models /lpkt_utils.py中

[sonyawong]

在使用wandb_eval.py进行lpkt模型的eval时出现报错： Traceback (most recent call last): File "wandb_eval.py", line 83, in main(params) File "wandb_eval.py", line 57, in main atkt_pad) File "D:\CODE_env\Anaconda\anaconda3\envs\py375\lib\site-packages\pykt\models\evaluate_model.py", line 628, in evaluate_splitpred_question curqin, curcin, currin, curdforget, ctrues, cpreds = predict_each_group(dtotal, dcur, dforget, curdforget, is_repeat, qidx, uid, idx, model_name, model, t, end, fout, atkt_pad) File "D:\CODE_env\Anaconda\anaconda3\envs\py375\lib\site-packages\pykt\models\evaluate_model.py", line 677, in predict_each_group curitin = dcur["curitin"] KeyError: 'curitin'

通过debug发现当前学生信息dcur在"curitin"列会在循环中被drop掉导致predict_each_group()无法运行

如果注释掉： curqin, curcin, currin, curdforget, ctrues, cpreds = predict_each_group(dtotal, dcur, dforget, curdforget, is_repeat, qidx, uid, idx, model_name, model, t, end, fout, atkt_pad) #dcur = {"curqin": curqin, "curcin": curcin, "currin": currin, "curtin": curtin} 则可正常跑通。

您的代码有改动吗, “pykt\models\evaluate_model.py", line 677 , in predict_each_group curitin = dcur["curitin"] KeyError: 'curitin' 跟原文件“evaluate_model.py”所在的行号不对, 我这边本地执行是能跑通的. 另外, lpkt_evaluate_multi_ahead放在了model/init.py进行了初始化, 所以from pykt.models import evaluate_splitpred_question, load_model, lpkt_evaluate_multi_ahead是能执行通的

[PhilrainV]

我的代码没有改动，就是直接在下载然后运行的