Update README.md
Browse files
README.md
CHANGED
@@ -54,12 +54,12 @@ The model is trained on runpod using 3090 and cost me around 15 dollors.
|
|
54 |
|
55 |
#### Hyperparameter specification
|
56 |
|
57 |
-
The model is trained for
|
58 |
|
59 |
Note that as a consequence of the weighting scheme which translates into a number of different multiply for each image,
|
60 |
the count of repeat and epoch has a quite different meaning here.
|
61 |
For example, depending on the weighting, I have around 300K images (some images are used multiple times) in an epoch,
|
62 |
-
and therefore I did not even finish an entire epoch with the
|
63 |
|
64 |
### Failures
|
65 |
|
|
|
54 |
|
55 |
#### Hyperparameter specification
|
56 |
|
57 |
+
The model is trained for 50000 steps, at batch size 4, lr 1e-6, resolution 512, and conditional dropping rate of 10%.
|
58 |
|
59 |
Note that as a consequence of the weighting scheme which translates into a number of different multiply for each image,
|
60 |
the count of repeat and epoch has a quite different meaning here.
|
61 |
For example, depending on the weighting, I have around 300K images (some images are used multiple times) in an epoch,
|
62 |
+
and therefore I did not even finish an entire epoch with the 50000 steps at batch size 4.
|
63 |
|
64 |
### Failures
|
65 |
|