Model save
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +7 -0
- trainer_state.json +28 -0
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -40,7 +40,7 @@ The following hyperparameters were used during training:
|
|
| 40 |
- seed: 42
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
-
- num_epochs:
|
| 44 |
|
| 45 |
### Training results
|
| 46 |
|
|
|
|
| 40 |
- seed: 42
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
+
- num_epochs: 60
|
| 44 |
|
| 45 |
### Training results
|
| 46 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.42857142857142855,
|
| 4 |
+
"eval_f1": 0.33939393939393936,
|
| 5 |
+
"eval_loss": 1.7643522024154663,
|
| 6 |
+
"eval_runtime": 6.8329,
|
| 7 |
+
"eval_samples_per_second": 2.049,
|
| 8 |
+
"eval_steps_per_second": 0.293,
|
| 9 |
+
"train_loss": 2.4430296421051025,
|
| 10 |
+
"train_runtime": 13.7707,
|
| 11 |
+
"train_samples_per_second": 1.017,
|
| 12 |
+
"train_steps_per_second": 0.073
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.42857142857142855,
|
| 4 |
+
"eval_f1": 0.33939393939393936,
|
| 5 |
+
"eval_loss": 1.7643522024154663,
|
| 6 |
+
"eval_runtime": 6.8329,
|
| 7 |
+
"eval_samples_per_second": 2.049,
|
| 8 |
+
"eval_steps_per_second": 0.293
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343293613
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ada227cd5742a1b321fe3eeca873a29bf485942337687a50ee7159873bada9a9
|
| 3 |
size 343293613
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 2.4430296421051025,
|
| 4 |
+
"train_runtime": 13.7707,
|
| 5 |
+
"train_samples_per_second": 1.017,
|
| 6 |
+
"train_steps_per_second": 0.073
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 1,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"step": 1,
|
| 14 |
+
"total_flos": 1084975366164480.0,
|
| 15 |
+
"train_loss": 2.4430296421051025,
|
| 16 |
+
"train_runtime": 13.7707,
|
| 17 |
+
"train_samples_per_second": 1.017,
|
| 18 |
+
"train_steps_per_second": 0.073
|
| 19 |
+
}
|
| 20 |
+
],
|
| 21 |
+
"logging_steps": 1000,
|
| 22 |
+
"max_steps": 1,
|
| 23 |
+
"num_train_epochs": 1,
|
| 24 |
+
"save_steps": 1000,
|
| 25 |
+
"total_flos": 1084975366164480.0,
|
| 26 |
+
"trial_name": null,
|
| 27 |
+
"trial_params": null
|
| 28 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4027
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0f8ae252b4d6cc9563f4e253723ea6a22afed0965df161dddb07412ca0d4723
|
| 3 |
size 4027
|