search

Leezekun / ViTST

[NeurIPS 2023] The official repo for the paper: "Time Series as Images: Vision Transformer for Irregularly Sampled Time Series"."
https://arxiv.org/abs/2303.12799
111 stars 8 forks source link

ValueError: Unrecognized model identifier in allenai/longformer-base-4096. Should contains one of 'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', 'xlm', 'roberta', 'ctrl' #3

Open KDGPR opened 1 year ago

KDGPR commented 1 year ago

When I follow the steps of 'ReadMe'. When I run ConstructImage.py. The error comes. it comes from row 209: tokenizer = AutoTokenizer.from_pretrained("allenai/longformer-base-4096") how to solve this? Thank you.

Leezekun commented 1 year ago

Have you installed the transformers package?

KDGPR commented 1 year ago

Thank you for your help. Yes, I updated my transformers yesterday, and the problem solved.A new problem comes. I can't download the model. When I run the code, the error comes: OSError: We couldn't connect to 'https://huggingface.co' to load this file, couldn't find it in the cached files and it looks like allenai/longformer-base-4096 is not the path to a directory containing a file named config.json. Checkout your internet connection or see how to run the library in offline mode at 'https://huggingface.co/docs/transformers/installation#offline-mode'.

Maybe I need Network ramming?

------------------ 原始邮件 ------------------ 发件人: "Zekun @.>; 发送时间: 2023年10月12日(星期四) 凌晨4:58 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [Leezekun/ViTST] ValueError: Unrecognized model identifier in allenai/longformer-base-4096. Should contains one of 'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', 'xlm', 'roberta', 'ctrl' (Issue #3)

Have you installed the transformers package?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

Leezekun commented 1 year ago

The tokenizer is just for analyzing the length of static features. It is not necessary. You can remove the code related to the tokenizer if you have trouble loading it.

KDGPR commented 1 year ago

OK I'll try. Thank you again my hero!!!!!!!!!!!

------------------ 原始邮件 ------------------ 发件人: "Zekun @.>; 发送时间: 2023年10月12日(星期四) 上午8:40 收件人: @.>; 抄送: @.>; @.>; 主题: Re: [Leezekun/ViTST] ValueError: Unrecognized model identifier in allenai/longformer-base-4096. Should contains one of 'bert', 'openai-gpt', 'gpt2', 'transfo-xl', 'xlnet', 'xlm', 'roberta', 'ctrl' (Issue #3)

The tokenizer is just for analyzing the length of static features. It is not necessary. You can remove the code related to the tokenizer if you have trouble loading it.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

KDGPR commented 1 year ago

I'm soory to bother you again. What need I do on the datasets? put it in Classification folder? you can download the datasets at http://www.timeseriesclassification.com/Downloads/Archives/Multivariate2018_ts.zip.

Create a folder Classification in the TSRAdara folder. Run PlotMarkers.py and ConstructDataset.py to create the images.