search

av-savchenko / face-emotion-recognition

Efficient face emotion recognition in photos and videos
Apache License 2.0
694 stars 127 forks source link

Question about this work. #26

Closed sonnguyen1996 closed 1 year ago

sonnguyen1996 commented 1 year ago

Dear Andrey Savchenko,

I'm a student and going to build a small system to detect student's emotions for my thesis. After finding a solution, I found your job. But I can't run https://github.com/HSE-asavchenko/face-emotion-recognition/blob/main/src/affectnet/train_emotions.ipynb by current AFFECT dataset's version. Please correct me if I'm wrong. My question is: Can I run this workhttps://github.com/HSE-asavchenko/face-emotion-recognition/blob/main/src/affectnet/train_affectnet_march2021_pytorch.ipynb with MobileNet. Because I tend to build small applications to detect emotions from client site then send result to server.

Many thanks,

Son Nguyen.

av-savchenko commented 1 year ago

Hello! I do not have the recent version of AffectNet, so I cannot modify the code by myself. As far as I know, the images are the same as before, only their labels are organized in the csv file. Actually, that's was the reason to create the copy of my code for recent dataset that you mentioned. I believe you could easily convert this code for usage with my TensorFlow's implementation of MobileNet. Just load the labels of validation images that you need and match them with my predictions, but be sure about changes in position of labeling as it is written in train_affectnet_march2021_pytorch.ipynb. If you need to retrain the model, just change the imagedataloader and read labels from csv files with Pandas.

sonnguyen1996 commented 1 year ago

Hi, Thanks for you reply, I had try create my own csv file by myself as your suggestion in previous closed issue, but it's seem to take a long time. I will try again like what you guide above. Thank you!

av-savchenko commented 1 year ago

Closing due to inactivity