iamsubrata
commited on
Commit
•
0e1ee10
1
Parent(s):
e5c0c43
Update README.md
Browse files
README.md
CHANGED
@@ -75,7 +75,7 @@ These steps establish the DataLoaders required for training and validation, maki
|
|
75 |
- Enabled mixed-precision training using `to_fp16()` for enhanced training efficiency.
|
76 |
|
77 |
**Finding Learning Rate:**
|
78 |
-
- Utilized the
|
79 |
- Discovered a suitable learning rate range (`lr`) for model training, set as `slice(1e-4, 5e-3)`.
|
80 |
|
81 |
This section highlights the model configuration, including the choice of architecture, pretrained weights, and learning rate discovery, ensuring an effective setup for model training and evaluation.
|
|
|
75 |
- Enabled mixed-precision training using `to_fp16()` for enhanced training efficiency.
|
76 |
|
77 |
**Finding Learning Rate:**
|
78 |
+
- Utilized the **`lr_find()`** method to determine an optimal learning rate.
|
79 |
- Discovered a suitable learning rate range (`lr`) for model training, set as `slice(1e-4, 5e-3)`.
|
80 |
|
81 |
This section highlights the model configuration, including the choice of architecture, pretrained weights, and learning rate discovery, ensuring an effective setup for model training and evaluation.
|