search

zrguo / MPLMM

[ACL 2024 Main] Official PyTorch implementation of the paper "Multimodal Prompt Learning with Missing Modalities for Sentiment Analysis and Emotion Recognition"
MIT License
33 stars 1 forks source link

model.py 的429行，在Fine-tuning on CMU-MOSI的时候报错 #2

Open wupeng1989 opened 3 months ago

wupeng1989 commented 3 months ago

你好，很感谢你的工作。 先Pre-train the model on CMU-MOSEI without prompts 后，然后执行Fine-tuning on CMU-MOSI的代码时候，在model.py的429行， h_ls = torch.cat([h_ls, batch_prompt[0]], dim=0)，提示报错Sizes of tensors must match except in dimension 0. h_ls的维度是:[50,8,60], batch_prompt[0]的维度是[8,16,60]，这个变量cat会提示报错。 期待你的回复，谢谢！

supercool706 commented 2 months ago

我也是遇到了这个问题TAT

zrguo commented 2 months ago

你好，代码整理时出现纰漏，现已更新。

version98 commented 1 month ago

是[8,16,

Hello, did you perform feature extraction on your dataset yourself? And are they still undergoing training? 您好，请问您复现的时候，是自己重新提取了数据集特征吗？

czy-a commented 2 days ago

是[8,16,

Hello, did you perform feature extraction on your dataset yourself? And are they still undergoing training? 您好，请问您复现的时候，是自己重新提取了数据集特征吗？

你好，想问一下这个数据集是需要根据缺失的mode来对原始完整数据进行缺失处理吗