End of training
Browse files- all_results.json +9 -9
- eval_results.json +5 -5
- train_results.json +4 -4
- trainer_state.json +0 -0
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 28.
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 2.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.9997974273270536,
|
| 4 |
+
"eval_loss": 0.001714603858999908,
|
| 5 |
+
"eval_runtime": 21.9591,
|
| 6 |
+
"eval_samples_per_second": 899.216,
|
| 7 |
+
"eval_steps_per_second": 28.143,
|
| 8 |
+
"train_loss": 0.03131400679324697,
|
| 9 |
+
"train_runtime": 2706.2965,
|
| 10 |
+
"train_samples_per_second": 328.332,
|
| 11 |
+
"train_steps_per_second": 2.564
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 28.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.9997974273270536,
|
| 4 |
+
"eval_loss": 0.001714603858999908,
|
| 5 |
+
"eval_runtime": 21.9591,
|
| 6 |
+
"eval_samples_per_second": 899.216,
|
| 7 |
+
"eval_steps_per_second": 28.143
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 2.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.03131400679324697,
|
| 4 |
+
"train_runtime": 2706.2965,
|
| 5 |
+
"train_samples_per_second": 328.332,
|
| 6 |
+
"train_steps_per_second": 2.564
|
| 7 |
}
|
trainer_state.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|