Commit
•
2c6ddbd
1
Parent(s):
2574b2f
Update README.md
Browse files
README.md
CHANGED
|
@@ -61,8 +61,46 @@ The following hyperparameters were used during training:
|
|
| 61 |
- lr_scheduler_type: linear
|
| 62 |
- num_epochs: 3
|
| 63 |
|
| 64 |
-
### Training
|
| 65 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
|
| 67 |
|
| 68 |
### Framework versions
|
|
|
|
| 61 |
- lr_scheduler_type: linear
|
| 62 |
- num_epochs: 3
|
| 63 |
|
| 64 |
+
### Training details
|
| 65 |
+
Here's the data presented in a table format:
|
| 66 |
+
|
| 67 |
+
| Step | Training Loss |
|
| 68 |
+
|--------|---------------|
|
| 69 |
+
| 500 | 1.423400 |
|
| 70 |
+
| 1000 | 1.233600 |
|
| 71 |
+
| 1500 | 1.184600 |
|
| 72 |
+
| 2000 | 1.125000 |
|
| 73 |
+
| 2500 | 1.113000 |
|
| 74 |
+
| 3000 | 1.070500 |
|
| 75 |
+
| 3500 | 1.063300 |
|
| 76 |
+
| 4000 | 1.031900 |
|
| 77 |
+
| 4500 | 1.017900 |
|
| 78 |
+
| 5000 | 1.008200 |
|
| 79 |
+
| 5500 | 1.002500 |
|
| 80 |
+
| 6000 | 0.973900 |
|
| 81 |
+
| 6500 | 0.907700 |
|
| 82 |
+
| 7000 | 0.920600 |
|
| 83 |
+
| 7500 | 0.905000 |
|
| 84 |
+
| 8000 | 0.900300 |
|
| 85 |
+
| 8500 | 0.888500 |
|
| 86 |
+
| 9000 | 0.892000 |
|
| 87 |
+
| 9500 | 0.881200 |
|
| 88 |
+
| 10000 | 0.890200 |
|
| 89 |
+
| 10500 | 0.881500 |
|
| 90 |
+
| 11000 | 0.876800 |
|
| 91 |
+
| 11500 | 0.861000 |
|
| 92 |
+
| 12000 | 0.854800 |
|
| 93 |
+
| 12500 | 0.819500 |
|
| 94 |
+
| 13000 | 0.818100 |
|
| 95 |
+
| 13500 | 0.827400 |
|
| 96 |
+
| 14000 | 0.806400 |
|
| 97 |
+
| 14500 | 0.811000 |
|
| 98 |
+
| 15000 | 0.815600 |
|
| 99 |
+
| 15500 | 0.818500 |
|
| 100 |
+
| 16000 | 0.804800 |
|
| 101 |
+
| 16500 | 0.827200 |
|
| 102 |
+
| 17000 | 0.808300 |
|
| 103 |
+
| 17500 | 0.807600 |
|
| 104 |
|
| 105 |
|
| 106 |
### Framework versions
|