gender-image-classifier's Introduction

Gender-Image-Classifier

Uses Keras with a TensorFlow Backend to classify facial photos as male or female.

To use this to create your own model you would do a few steps:

Download lots of facial images - I used the LFW dataset: http://vis-www.cs.umass.edu/lfw/
run the image_classifier.py script
- This will write your model to a JSON file and the weights for your model to an h5 file.

This can be used to classify images by:

Importing your model and load your weights
read images you'd like to classify into numpy array
Call the 'model.predict()' function on your numpy array representation of the image you'd like to classify.

An example of this can be found in the testing.py file.

gender-image-classifier's People

Contributors

Stargazers

Watchers

gender-image-classifier's Issues

Classifying age

Is it possible to classify the age as well or is there any dataset apart from imdb and adience to do so since both are inaccurate?

Test exception

Hi, I run testing.py but throw this exception:

Traceback (most recent call last):
  File "testing.py", line 35, in <module>
    prediction_b = model.predict(bailey)
  File "/usr/lib/python3.7/site-packages/keras/engine/training.py", line 1441, in predict
    x, _, _ = self._standardize_user_data(x)
  File "/usr/lib/python3.7/site-packages/keras/engine/training.py", line 579, in _standardize_user_data
    exception_prefix='input')
  File "/usr/lib/python3.7/site-packages/keras/engine/training_utils.py", line 135, in standardize_input_data
    'with shape ' + str(data_shape))
ValueError: Error when checking input: expected conv2d_1_input to have 4 dimensions, but got array with shape (1, 1)