search

oarriaga / face_classification

Real-time face detection and emotion/gender classification using fer2013/imdb datasets with a keras CNN model and openCV.
MIT License
5.61k stars 1.59k forks source link

Can you share the architecture of this project? #77

Closed nk0307 closed 5 years ago

nk0307 commented 6 years ago

I am really lost in the project. Can you show me the flow so I can carry on reading it.

nishgaba-ai commented 6 years ago

The model files are already saved in the trained_models folder, it is easy to see the architecture of the models using model.summary(), I am attaching the output of model.summary() output for FER miniXception model file.

`

Reading the model

import tensorflow as tf

model_file = "fer2013_mini_XCEPTION.102-0.66.hdf5"

model = tf.keras.models.load_model(model_file) from keras import backend as K sess = K.get_session()

print(model.summary())

`

selection_106 selection_105

The images are uploaded in reverse order, so this is the whole architecture of the given model