Render animated network diagrams from (simple) TensorFlow models.
This code was initially developed to create renders and animations for an AWS Machine Learning Specialty Certification (MLS-C01) course. You can find that course here: https://learn.mikegchambers.com/p/aws-machine-learning-specialty-certification-course
This is an early stage commit. Documentation is sparse.
When calling create_animation
, the options are as follows:
Option | Default |
---|---|
margin | 200 |
size | [1920, 1080] |
node_size | 90 |
node_gap | 40 |
background_rgba | (32, 32, 32, 0) |
node_rgb | (248, 167, 3) |
node_stroke | 5 |
conn_rgb | (81, 181, 237) |
conn_max_width | 20 |
frame_numbers | False |
frame_numbers_title | 'epoch' |
frame_numbers_font | 'OpenSans-Regular.ttf' |
frame_numbers_size | 50 |
frame_numbers_xy | (100,100) |
frame_numbers_rgb | (50,50,50) |
gif | False |
gif_name | 'animation.gif' |
fps | 5 |
See the example notebook to get you going. More docs in the future...