May 2020
Beginner to intermediate
430 pages
10h 39m
English
Content preview from Mastering Computer Vision with TensorFlow 2.x
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
Start your free trial
Baseline case
The baseline case is denoted by the following parameters of the neural network:
- Number of instances: 35,888
- Instance length: 2,304
- 28,709 train samples
- 3,589 test samples
The model iterations are as follows:
Epoch 1/5 256/256 [=======================] - 78s 306ms/step - loss: 1.8038 - acc: 0.2528 Epoch 2/5 256/256 [=======================] - 78s 303ms/step - loss: 1.6188 - acc: 0.3561 Epoch 3/5 256/256 [=======================] - 78s 305ms/step - loss: 1.4309 - acc: 0.4459 Epoch 4/5 256/256 [=======================] - 78s 306ms/step - loss: 1.2889 - acc: 0.5046 Epoch 5/5 256/256 [=======================] - 79s 308ms/step - loss: 1.1947 - acc: 0.5444
Next, we will optimize the CNN parameters to determine which parameters have ...