Update README.md
Browse files
README.md
CHANGED
|
@@ -38,9 +38,8 @@ The dataset for pre-training is collected from public data, most of which are we
|
|
| 38 |
|
| 39 |
## Training Detail
|
| 40 |
The training is performed thanks to a grant from [TPU Research Cloud](https://sites.research.google/trc/about/) on a TPU v4-128 node.
|
| 41 |
-
|
| 42 |
-
Accuracy: 0.48
|
| 43 |
-
Batch Size: 1024
|
| 44 |
|
| 45 |
## Copyright
|
| 46 |
The model is released under the Apache 2.0 license.
|
|
|
|
| 38 |
|
| 39 |
## Training Detail
|
| 40 |
The training is performed thanks to a grant from [TPU Research Cloud](https://sites.research.google/trc/about/) on a TPU v4-128 node.
|
| 41 |
+
|
| 42 |
+
Loss: 2.68; Accuracy: 0.48; Batch Size: 1024
|
|
|
|
| 43 |
|
| 44 |
## Copyright
|
| 45 |
The model is released under the Apache 2.0 license.
|