Timeline for GAN model predictions before training is predictable
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 17, 2020 at 14:38 | vote | accept | dee cue | ||
| Nov 17, 2020 at 12:37 | answer | added | Aray Karjauv | timeline score: 3 | |
| Nov 17, 2020 at 11:54 | comment | added | dee cue | Yes, and it only generates noises. Although, I do notice some preference of a certain pallete, and I think this is a good start. The D accuracy went constant at 0.5 after 20 epochs, and the G accuracy went up to 1.0, and both losses gradually went down. All metrics did oscillate in the middle of the training but in small amount of deviation. Did my model overfit at epoch 20? | |
| Nov 17, 2020 at 11:29 | comment | added | Aray Karjauv | Did you try to generate fake images? Since the main goal of GANs to produce synthetic samples, the main matric will be the quality of generated images. Take a look at this answer. There is a plot of G loss and D loss of DCGAN. We can see that D loss is near 0, whereas G loss has oscillations, but G was able to produce good samples. | |
| Nov 17, 2020 at 11:20 | history | edited | dee cue | CC BY-SA 4.0 |
add more context
|
| Nov 17, 2020 at 11:18 | comment | added | dee cue |
In other words, I feed real images array into the discriminator model, and it outputs values close to 1.0. Next, I feed fake images (the one that is generated by the generator) into the discriminator model, and it outputs values close to 0.5. These happens too often, and it is very predictable.
|
|
| Nov 16, 2020 at 19:29 | comment | added | Aray Karjauv | what do you mean by "a comparison of the discriminator model prediction ... against the whole GAN model prediction"? do you mean the output of your generator for fake and real images? | |
| Nov 16, 2020 at 15:41 | history | asked | dee cue | CC BY-SA 4.0 |