Practical Machine Learning for Computer Vision by Valliappa Lakshmanan & Martin Görner & Ryan Gillard

Author:Valliappa Lakshmanan & Martin Görner & Ryan Gillard [Valliappa Lakshmanan]
Language: eng
Format: epub
Publisher: O'Reilly Media, Inc.
Published: 2021-10-24T16:00:00+00:00

VGG19 at a glance

Architecture: alternates convolutional and max pooling layers using 3x3 convolutions only

Model Parameters excl. classif. heada ImageNet accuracy 104 flowers f1 scoreb fine-tuning 104 flowers f1 scorec trained from scratch

VGG19 20M 71% 88% precision: 89%, recall: 88%

previous best for comparison:

AlexNet 3.7M 60% 39% precision: 44%, recall: 38%

a Excluding classification head from parameter counts for easier comparisons between architectures. Without the classification head, the number of parameters in the network is resolution independent. Also, in fine-tuning examples, a different classification head might be used.

b Accuracy, precision, recall, and f1 score values: higher is better.

c

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.