search

BaoNinh2808 / Multi-Branch-Network-for-Imagery-Emotion-Prediction

Multi-Branch Network for Imagery Emotion Prediction
0 stars 0 forks source link

FileNotFoundError: [Errno 2] No such file or directory: '/content/drive/MyDrive/VA-prediction/models/SwinT_EMOTIC.pth' #1

Open RamPurush143 opened 5 months ago

RamPurush143 commented 5 months ago

Hello bro you used some pretrained models from your drive can you share me that models ?? and one more thing can you tell me where you get them??

can you reply fast i am trying to do this project.

model_body = torch.load('/content/drive/MyDrive/VA-prediction/models/body_train_lr001_b24_crossEtropy/model1.pth') print (summary(model_body, (3,128,128), device="cpu"))

model_body = torch.load("/content/drive/MyDrive/VA-prediction/models/SwinT_EMOTIC.pth", map_location=lambda storage, loc: storage) print (summary(model_body, (3,128,128), device="cpu")) isSwinT = True

reply fast

Thank you

BaoNinh2808 commented 5 months ago

https://drive.google.com/drive/folders/1H5FIVT_fwNvMDVuV9wgeytbIvmaMrtBo?usp=sharing

you can access two models through this link

RamPurush143 commented 5 months ago

Hello bro can you tell me which dataset you used and code to train the models and thank you for Providing the models

BaoNinh2808 commented 5 months ago

I also use the Emotic Dataset to train this model, but not in the task prediction VAD, instead of task classify gender, age, ... of the human in the image. Classification task is faster and easier to do. When you viewing the dataset, you will see beside V,A,D value and discrete emotion category, it also label the gender, age and some information of the human in this image. You can use this additional label for train your model and then transfer this to the task predict emotions. Although you not train with the similar task, you train with the similar data, it will help the model better in "understand" the way you work with this data --> better result.