Dimensionality Mismatch While Trying to Run Prediction

[savan77]

Hi,

Thanks for this work. I was trying to test on unannoatated data using the predict model. But, when I try to run the targetid, it throws following error. I think this has been reported before but not sure how to solve this. Also, I just downloaded the code and model so they are up to date. Dimensions of lookup parameter /_0 lookup up from file ({100,400574}) do not match parameters to be populated ({100,400001}) Thanks

[zjxs1997]

similar problem. I ran python -m sesame.targetid --mode predict --model_name fn1.7-pretrained-targetid --raw_input sentences.txt and get RuntimeError: Dimensions of lookup parameter /_0 lookup up from file ({100,400574}) do not match parameters to be populated ({100,400000})

[heeh]

Same error here. Did anyone resolve this issue? RuntimeError: Dimensions of lookup parameter /_0 lookup up from file ({100,400574}) do not match parameters to be populated ({100,410050})

[Zce1112zslx]

Same error here. Did anyone resolve this issue? RuntimeError: Dimensions of lookup parameter /_0 lookup up from file ({100,400574}) do not match parameters to be populated ({100,410050})

I've also met this problem, and it seems that the glove embeddings we use are not same as their pretrained model. Retrain this model on framenet data can solve this problem, but i got 'nan gradient' during train stage. Now i decided to ask them for their version of GloVe embeddings.

[heeh]

@Zce1112zslx Thank you for figuring this out. I finally understand what's going on here.

[ankushtale]

Facing the same issue. @heeh @Zce1112zslx, Can you please drop a note when it works for you?

@swabhs @sammthomson Would really appreciate your help!

[jianliu-ml]

Same error here. Did anyone resolve this issue? RuntimeError: Dimensions of lookup parameter /_0 lookup up from file ({100,400574}) do not match parameters to be populated ({100,410050})

I've also met this problem, and it seems that the glove embeddings we use are not same as their pretrained model. Retrain this model on framenet data can solve this problem, but i got 'nan gradient' during train stage. Now i decided to ask them for their version of GloVe embeddings.

Hi, same issue! Have you obtained the GloVe embeddings?

[Zce1112zslx]

Not yet. And I can also not fixed the nan-gradient bug on retraining. I am now trying to add gradient-clip to their code. I will email you if I slove this promblem out. Thanks.

发送自 Windows 10 版邮件应用

发件人: jianliu-ml 发送时间: 2020年12月30日 17:35 收件人: swabhs/open-sesame 抄送: Zce1112zslx; Mention 主题: Re: [swabhs/open-sesame] Dimensionality Mismatch While Trying to RunPrediction (#48)

Same error here. Did anyone resolve this issue? RuntimeError: Dimensions of lookup parameter /_0 lookup up from file ({100,400574}) do not match parameters to be populated ({100,410050}) I've also met this problem, and it seems that the glove embeddings we use are not same as their pretrained model. Retrain this model on framenet data can solve this problem, but i got 'nan gradient' during train stage. Now i decided to ask them for their version of GloVe embeddings. Hi, same issue! Have you obtained the GloVe embeddings? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.