Commit
•
4fe24a2
1
Parent(s):
bcdb08c
End of training
Browse files- README.md +3 -1
- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Dec19_00-25-43_b6c43624a1ee/events.out.tfevents.1702946597.b6c43624a1ee.6546.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +1182 -0
README.md
CHANGED
@@ -2,6 +2,8 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: google/vit-base-patch16-224-in21k
|
4 |
tags:
|
|
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- imagefolder
|
@@ -32,7 +34,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
-
- Loss: 0.
|
36 |
- Accuracy: 1.0
|
37 |
|
38 |
## Model description
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: google/vit-base-patch16-224-in21k
|
4 |
tags:
|
5 |
+
- image-classification
|
6 |
+
- vision
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- imagefolder
|
|
|
34 |
|
35 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
36 |
It achieves the following results on the evaluation set:
|
37 |
+
- Loss: 0.0186
|
38 |
- Accuracy: 1.0
|
39 |
|
40 |
## Model description
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 120.0,
|
3 |
+
"eval_accuracy": 1.0,
|
4 |
+
"eval_loss": 0.0186475720256567,
|
5 |
+
"eval_runtime": 6.0071,
|
6 |
+
"eval_samples_per_second": 1.498,
|
7 |
+
"eval_steps_per_second": 0.166,
|
8 |
+
"train_loss": 0.07813952726622422,
|
9 |
+
"train_runtime": 1030.5859,
|
10 |
+
"train_samples_per_second": 5.938,
|
11 |
+
"train_steps_per_second": 0.116
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 120.0,
|
3 |
+
"eval_accuracy": 1.0,
|
4 |
+
"eval_loss": 0.0186475720256567,
|
5 |
+
"eval_runtime": 6.0071,
|
6 |
+
"eval_samples_per_second": 1.498,
|
7 |
+
"eval_steps_per_second": 0.166
|
8 |
+
}
|
runs/Dec19_00-25-43_b6c43624a1ee/events.out.tfevents.1702946597.b6c43624a1ee.6546.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3e74b338be4e2e5598d48702393dc2043e0619b2ab84bdb59d6f6493836580c3
|
3 |
+
size 405
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 120.0,
|
3 |
+
"train_loss": 0.07813952726622422,
|
4 |
+
"train_runtime": 1030.5859,
|
5 |
+
"train_samples_per_second": 5.938,
|
6 |
+
"train_steps_per_second": 0.116
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1182 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.018274879083037376,
|
3 |
+
"best_model_checkpoint": "./drive/MyDrive/repositories/torch_example_image-classification/outputs3/checkpoint-115",
|
4 |
+
"epoch": 120.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 120,
|
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 |
+
"eval_accuracy": 1.0,
|
14 |
+
"eval_loss": 0.5751205682754517,
|
15 |
+
"eval_runtime": 6.6033,
|
16 |
+
"eval_samples_per_second": 1.363,
|
17 |
+
"eval_steps_per_second": 0.151,
|
18 |
+
"step": 1
|
19 |
+
},
|
20 |
+
{
|
21 |
+
"epoch": 2.0,
|
22 |
+
"eval_accuracy": 1.0,
|
23 |
+
"eval_loss": 0.508073091506958,
|
24 |
+
"eval_runtime": 0.1375,
|
25 |
+
"eval_samples_per_second": 65.474,
|
26 |
+
"eval_steps_per_second": 7.275,
|
27 |
+
"step": 2
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 3.0,
|
31 |
+
"eval_accuracy": 1.0,
|
32 |
+
"eval_loss": 0.46539825201034546,
|
33 |
+
"eval_runtime": 0.1277,
|
34 |
+
"eval_samples_per_second": 70.471,
|
35 |
+
"eval_steps_per_second": 7.83,
|
36 |
+
"step": 3
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 4.0,
|
40 |
+
"eval_accuracy": 1.0,
|
41 |
+
"eval_loss": 0.4014008939266205,
|
42 |
+
"eval_runtime": 0.1365,
|
43 |
+
"eval_samples_per_second": 65.949,
|
44 |
+
"eval_steps_per_second": 7.328,
|
45 |
+
"step": 4
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 5.0,
|
49 |
+
"eval_accuracy": 1.0,
|
50 |
+
"eval_loss": 0.3691883683204651,
|
51 |
+
"eval_runtime": 0.1452,
|
52 |
+
"eval_samples_per_second": 62.003,
|
53 |
+
"eval_steps_per_second": 6.889,
|
54 |
+
"step": 5
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 6.0,
|
58 |
+
"eval_accuracy": 1.0,
|
59 |
+
"eval_loss": 0.33271580934524536,
|
60 |
+
"eval_runtime": 0.1543,
|
61 |
+
"eval_samples_per_second": 58.338,
|
62 |
+
"eval_steps_per_second": 6.482,
|
63 |
+
"step": 6
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 7.0,
|
67 |
+
"eval_accuracy": 1.0,
|
68 |
+
"eval_loss": 0.29366230964660645,
|
69 |
+
"eval_runtime": 0.1402,
|
70 |
+
"eval_samples_per_second": 64.178,
|
71 |
+
"eval_steps_per_second": 7.131,
|
72 |
+
"step": 7
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 8.0,
|
76 |
+
"eval_accuracy": 1.0,
|
77 |
+
"eval_loss": 0.27746275067329407,
|
78 |
+
"eval_runtime": 0.1482,
|
79 |
+
"eval_samples_per_second": 60.711,
|
80 |
+
"eval_steps_per_second": 6.746,
|
81 |
+
"step": 8
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 9.0,
|
85 |
+
"eval_accuracy": 1.0,
|
86 |
+
"eval_loss": 0.23346243798732758,
|
87 |
+
"eval_runtime": 0.1319,
|
88 |
+
"eval_samples_per_second": 68.21,
|
89 |
+
"eval_steps_per_second": 7.579,
|
90 |
+
"step": 9
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 10.0,
|
94 |
+
"learning_rate": 1.8333333333333333e-05,
|
95 |
+
"loss": 0.4432,
|
96 |
+
"step": 10
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 10.0,
|
100 |
+
"eval_accuracy": 1.0,
|
101 |
+
"eval_loss": 0.2092142403125763,
|
102 |
+
"eval_runtime": 0.1641,
|
103 |
+
"eval_samples_per_second": 54.858,
|
104 |
+
"eval_steps_per_second": 6.095,
|
105 |
+
"step": 10
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 11.0,
|
109 |
+
"eval_accuracy": 1.0,
|
110 |
+
"eval_loss": 0.20070181787014008,
|
111 |
+
"eval_runtime": 0.1365,
|
112 |
+
"eval_samples_per_second": 65.913,
|
113 |
+
"eval_steps_per_second": 7.324,
|
114 |
+
"step": 11
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 12.0,
|
118 |
+
"eval_accuracy": 1.0,
|
119 |
+
"eval_loss": 0.16740131378173828,
|
120 |
+
"eval_runtime": 0.1357,
|
121 |
+
"eval_samples_per_second": 66.338,
|
122 |
+
"eval_steps_per_second": 7.371,
|
123 |
+
"step": 12
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 13.0,
|
127 |
+
"eval_accuracy": 1.0,
|
128 |
+
"eval_loss": 0.1546010971069336,
|
129 |
+
"eval_runtime": 0.1323,
|
130 |
+
"eval_samples_per_second": 68.042,
|
131 |
+
"eval_steps_per_second": 7.56,
|
132 |
+
"step": 13
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 14.0,
|
136 |
+
"eval_accuracy": 1.0,
|
137 |
+
"eval_loss": 0.13928303122520447,
|
138 |
+
"eval_runtime": 0.1256,
|
139 |
+
"eval_samples_per_second": 71.669,
|
140 |
+
"eval_steps_per_second": 7.963,
|
141 |
+
"step": 14
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 15.0,
|
145 |
+
"eval_accuracy": 1.0,
|
146 |
+
"eval_loss": 0.12973126769065857,
|
147 |
+
"eval_runtime": 0.1598,
|
148 |
+
"eval_samples_per_second": 56.334,
|
149 |
+
"eval_steps_per_second": 6.259,
|
150 |
+
"step": 15
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 16.0,
|
154 |
+
"eval_accuracy": 1.0,
|
155 |
+
"eval_loss": 0.12185994535684586,
|
156 |
+
"eval_runtime": 0.129,
|
157 |
+
"eval_samples_per_second": 69.791,
|
158 |
+
"eval_steps_per_second": 7.755,
|
159 |
+
"step": 16
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 17.0,
|
163 |
+
"eval_accuracy": 1.0,
|
164 |
+
"eval_loss": 0.10901562124490738,
|
165 |
+
"eval_runtime": 0.1351,
|
166 |
+
"eval_samples_per_second": 66.632,
|
167 |
+
"eval_steps_per_second": 7.404,
|
168 |
+
"step": 17
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 18.0,
|
172 |
+
"eval_accuracy": 1.0,
|
173 |
+
"eval_loss": 0.10118965059518814,
|
174 |
+
"eval_runtime": 0.1339,
|
175 |
+
"eval_samples_per_second": 67.193,
|
176 |
+
"eval_steps_per_second": 7.466,
|
177 |
+
"step": 18
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 19.0,
|
181 |
+
"eval_accuracy": 1.0,
|
182 |
+
"eval_loss": 0.09813226014375687,
|
183 |
+
"eval_runtime": 0.1408,
|
184 |
+
"eval_samples_per_second": 63.926,
|
185 |
+
"eval_steps_per_second": 7.103,
|
186 |
+
"step": 19
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 20.0,
|
190 |
+
"learning_rate": 1.6666666666666667e-05,
|
191 |
+
"loss": 0.1696,
|
192 |
+
"step": 20
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 20.0,
|
196 |
+
"eval_accuracy": 1.0,
|
197 |
+
"eval_loss": 0.08743412792682648,
|
198 |
+
"eval_runtime": 0.1305,
|
199 |
+
"eval_samples_per_second": 68.95,
|
200 |
+
"eval_steps_per_second": 7.661,
|
201 |
+
"step": 20
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 21.0,
|
205 |
+
"eval_accuracy": 1.0,
|
206 |
+
"eval_loss": 0.08119351416826248,
|
207 |
+
"eval_runtime": 0.1384,
|
208 |
+
"eval_samples_per_second": 65.021,
|
209 |
+
"eval_steps_per_second": 7.225,
|
210 |
+
"step": 21
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 22.0,
|
214 |
+
"eval_accuracy": 1.0,
|
215 |
+
"eval_loss": 0.07503622770309448,
|
216 |
+
"eval_runtime": 0.1377,
|
217 |
+
"eval_samples_per_second": 65.382,
|
218 |
+
"eval_steps_per_second": 7.265,
|
219 |
+
"step": 22
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 23.0,
|
223 |
+
"eval_accuracy": 1.0,
|
224 |
+
"eval_loss": 0.07538507133722305,
|
225 |
+
"eval_runtime": 0.1454,
|
226 |
+
"eval_samples_per_second": 61.884,
|
227 |
+
"eval_steps_per_second": 6.876,
|
228 |
+
"step": 23
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 24.0,
|
232 |
+
"eval_accuracy": 1.0,
|
233 |
+
"eval_loss": 0.06934237480163574,
|
234 |
+
"eval_runtime": 0.1277,
|
235 |
+
"eval_samples_per_second": 70.459,
|
236 |
+
"eval_steps_per_second": 7.829,
|
237 |
+
"step": 24
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 25.0,
|
241 |
+
"eval_accuracy": 1.0,
|
242 |
+
"eval_loss": 0.06419406086206436,
|
243 |
+
"eval_runtime": 0.1298,
|
244 |
+
"eval_samples_per_second": 69.338,
|
245 |
+
"eval_steps_per_second": 7.704,
|
246 |
+
"step": 25
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 26.0,
|
250 |
+
"eval_accuracy": 1.0,
|
251 |
+
"eval_loss": 0.06099466606974602,
|
252 |
+
"eval_runtime": 0.1313,
|
253 |
+
"eval_samples_per_second": 68.527,
|
254 |
+
"eval_steps_per_second": 7.614,
|
255 |
+
"step": 26
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 27.0,
|
259 |
+
"eval_accuracy": 1.0,
|
260 |
+
"eval_loss": 0.058600153774023056,
|
261 |
+
"eval_runtime": 2.7473,
|
262 |
+
"eval_samples_per_second": 3.276,
|
263 |
+
"eval_steps_per_second": 0.364,
|
264 |
+
"step": 27
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 28.0,
|
268 |
+
"eval_accuracy": 1.0,
|
269 |
+
"eval_loss": 0.05692301690578461,
|
270 |
+
"eval_runtime": 0.1269,
|
271 |
+
"eval_samples_per_second": 70.898,
|
272 |
+
"eval_steps_per_second": 7.878,
|
273 |
+
"step": 28
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 29.0,
|
277 |
+
"eval_accuracy": 1.0,
|
278 |
+
"eval_loss": 0.05317116528749466,
|
279 |
+
"eval_runtime": 0.1357,
|
280 |
+
"eval_samples_per_second": 66.346,
|
281 |
+
"eval_steps_per_second": 7.372,
|
282 |
+
"step": 29
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 30.0,
|
286 |
+
"learning_rate": 1.5000000000000002e-05,
|
287 |
+
"loss": 0.0792,
|
288 |
+
"step": 30
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 30.0,
|
292 |
+
"eval_accuracy": 1.0,
|
293 |
+
"eval_loss": 0.0506051704287529,
|
294 |
+
"eval_runtime": 0.1381,
|
295 |
+
"eval_samples_per_second": 65.164,
|
296 |
+
"eval_steps_per_second": 7.24,
|
297 |
+
"step": 30
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 31.0,
|
301 |
+
"eval_accuracy": 1.0,
|
302 |
+
"eval_loss": 0.049507126212120056,
|
303 |
+
"eval_runtime": 0.1288,
|
304 |
+
"eval_samples_per_second": 69.852,
|
305 |
+
"eval_steps_per_second": 7.761,
|
306 |
+
"step": 31
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 32.0,
|
310 |
+
"eval_accuracy": 1.0,
|
311 |
+
"eval_loss": 0.047622185200452805,
|
312 |
+
"eval_runtime": 0.1335,
|
313 |
+
"eval_samples_per_second": 67.403,
|
314 |
+
"eval_steps_per_second": 7.489,
|
315 |
+
"step": 32
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 33.0,
|
319 |
+
"eval_accuracy": 1.0,
|
320 |
+
"eval_loss": 0.04570432007312775,
|
321 |
+
"eval_runtime": 0.1542,
|
322 |
+
"eval_samples_per_second": 58.363,
|
323 |
+
"eval_steps_per_second": 6.485,
|
324 |
+
"step": 33
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 34.0,
|
328 |
+
"eval_accuracy": 1.0,
|
329 |
+
"eval_loss": 0.04416535794734955,
|
330 |
+
"eval_runtime": 0.1278,
|
331 |
+
"eval_samples_per_second": 70.446,
|
332 |
+
"eval_steps_per_second": 7.827,
|
333 |
+
"step": 34
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 35.0,
|
337 |
+
"eval_accuracy": 1.0,
|
338 |
+
"eval_loss": 0.041947636753320694,
|
339 |
+
"eval_runtime": 0.1271,
|
340 |
+
"eval_samples_per_second": 70.804,
|
341 |
+
"eval_steps_per_second": 7.867,
|
342 |
+
"step": 35
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 36.0,
|
346 |
+
"eval_accuracy": 1.0,
|
347 |
+
"eval_loss": 0.04041131213307381,
|
348 |
+
"eval_runtime": 0.1382,
|
349 |
+
"eval_samples_per_second": 65.106,
|
350 |
+
"eval_steps_per_second": 7.234,
|
351 |
+
"step": 36
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 37.0,
|
355 |
+
"eval_accuracy": 1.0,
|
356 |
+
"eval_loss": 0.03958812355995178,
|
357 |
+
"eval_runtime": 0.149,
|
358 |
+
"eval_samples_per_second": 60.405,
|
359 |
+
"eval_steps_per_second": 6.712,
|
360 |
+
"step": 37
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 38.0,
|
364 |
+
"eval_accuracy": 1.0,
|
365 |
+
"eval_loss": 0.0383819043636322,
|
366 |
+
"eval_runtime": 0.1281,
|
367 |
+
"eval_samples_per_second": 70.282,
|
368 |
+
"eval_steps_per_second": 7.809,
|
369 |
+
"step": 38
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 39.0,
|
373 |
+
"eval_accuracy": 1.0,
|
374 |
+
"eval_loss": 0.03773792088031769,
|
375 |
+
"eval_runtime": 0.1297,
|
376 |
+
"eval_samples_per_second": 69.409,
|
377 |
+
"eval_steps_per_second": 7.712,
|
378 |
+
"step": 39
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 40.0,
|
382 |
+
"learning_rate": 1.3333333333333333e-05,
|
383 |
+
"loss": 0.049,
|
384 |
+
"step": 40
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 40.0,
|
388 |
+
"eval_accuracy": 1.0,
|
389 |
+
"eval_loss": 0.03664920851588249,
|
390 |
+
"eval_runtime": 0.1332,
|
391 |
+
"eval_samples_per_second": 67.583,
|
392 |
+
"eval_steps_per_second": 7.509,
|
393 |
+
"step": 40
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 41.0,
|
397 |
+
"eval_accuracy": 1.0,
|
398 |
+
"eval_loss": 0.03701655566692352,
|
399 |
+
"eval_runtime": 0.1399,
|
400 |
+
"eval_samples_per_second": 64.34,
|
401 |
+
"eval_steps_per_second": 7.149,
|
402 |
+
"step": 41
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 42.0,
|
406 |
+
"eval_accuracy": 1.0,
|
407 |
+
"eval_loss": 0.03389488160610199,
|
408 |
+
"eval_runtime": 0.1394,
|
409 |
+
"eval_samples_per_second": 64.543,
|
410 |
+
"eval_steps_per_second": 7.171,
|
411 |
+
"step": 42
|
412 |
+
},
|
413 |
+
{
|
414 |
+
"epoch": 43.0,
|
415 |
+
"eval_accuracy": 1.0,
|
416 |
+
"eval_loss": 0.03304838761687279,
|
417 |
+
"eval_runtime": 0.1264,
|
418 |
+
"eval_samples_per_second": 71.198,
|
419 |
+
"eval_steps_per_second": 7.911,
|
420 |
+
"step": 43
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 44.0,
|
424 |
+
"eval_accuracy": 1.0,
|
425 |
+
"eval_loss": 0.03438062593340874,
|
426 |
+
"eval_runtime": 0.1325,
|
427 |
+
"eval_samples_per_second": 67.899,
|
428 |
+
"eval_steps_per_second": 7.544,
|
429 |
+
"step": 44
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 45.0,
|
433 |
+
"eval_accuracy": 1.0,
|
434 |
+
"eval_loss": 0.032368406653404236,
|
435 |
+
"eval_runtime": 0.1314,
|
436 |
+
"eval_samples_per_second": 68.518,
|
437 |
+
"eval_steps_per_second": 7.613,
|
438 |
+
"step": 45
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 46.0,
|
442 |
+
"eval_accuracy": 1.0,
|
443 |
+
"eval_loss": 0.03226124122738838,
|
444 |
+
"eval_runtime": 0.1386,
|
445 |
+
"eval_samples_per_second": 64.933,
|
446 |
+
"eval_steps_per_second": 7.215,
|
447 |
+
"step": 46
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 47.0,
|
451 |
+
"eval_accuracy": 1.0,
|
452 |
+
"eval_loss": 0.031140118837356567,
|
453 |
+
"eval_runtime": 0.1386,
|
454 |
+
"eval_samples_per_second": 64.914,
|
455 |
+
"eval_steps_per_second": 7.213,
|
456 |
+
"step": 47
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 48.0,
|
460 |
+
"eval_accuracy": 1.0,
|
461 |
+
"eval_loss": 0.030753135681152344,
|
462 |
+
"eval_runtime": 0.1285,
|
463 |
+
"eval_samples_per_second": 70.06,
|
464 |
+
"eval_steps_per_second": 7.784,
|
465 |
+
"step": 48
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 49.0,
|
469 |
+
"eval_accuracy": 1.0,
|
470 |
+
"eval_loss": 0.029377290979027748,
|
471 |
+
"eval_runtime": 0.134,
|
472 |
+
"eval_samples_per_second": 67.146,
|
473 |
+
"eval_steps_per_second": 7.461,
|
474 |
+
"step": 49
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 50.0,
|
478 |
+
"learning_rate": 1.1666666666666668e-05,
|
479 |
+
"loss": 0.0359,
|
480 |
+
"step": 50
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 50.0,
|
484 |
+
"eval_accuracy": 1.0,
|
485 |
+
"eval_loss": 0.02965053915977478,
|
486 |
+
"eval_runtime": 0.1451,
|
487 |
+
"eval_samples_per_second": 62.041,
|
488 |
+
"eval_steps_per_second": 6.893,
|
489 |
+
"step": 50
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 51.0,
|
493 |
+
"eval_accuracy": 1.0,
|
494 |
+
"eval_loss": 0.02891460806131363,
|
495 |
+
"eval_runtime": 0.1394,
|
496 |
+
"eval_samples_per_second": 64.561,
|
497 |
+
"eval_steps_per_second": 7.173,
|
498 |
+
"step": 51
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 52.0,
|
502 |
+
"eval_accuracy": 1.0,
|
503 |
+
"eval_loss": 0.028457429260015488,
|
504 |
+
"eval_runtime": 0.1579,
|
505 |
+
"eval_samples_per_second": 56.996,
|
506 |
+
"eval_steps_per_second": 6.333,
|
507 |
+
"step": 52
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 53.0,
|
511 |
+
"eval_accuracy": 1.0,
|
512 |
+
"eval_loss": 0.027989288792014122,
|
513 |
+
"eval_runtime": 2.6177,
|
514 |
+
"eval_samples_per_second": 3.438,
|
515 |
+
"eval_steps_per_second": 0.382,
|
516 |
+
"step": 53
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 54.0,
|
520 |
+
"eval_accuracy": 1.0,
|
521 |
+
"eval_loss": 0.02697157859802246,
|
522 |
+
"eval_runtime": 0.1341,
|
523 |
+
"eval_samples_per_second": 67.138,
|
524 |
+
"eval_steps_per_second": 7.46,
|
525 |
+
"step": 54
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 55.0,
|
529 |
+
"eval_accuracy": 1.0,
|
530 |
+
"eval_loss": 0.026479896157979965,
|
531 |
+
"eval_runtime": 0.138,
|
532 |
+
"eval_samples_per_second": 65.23,
|
533 |
+
"eval_steps_per_second": 7.248,
|
534 |
+
"step": 55
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 56.0,
|
538 |
+
"eval_accuracy": 1.0,
|
539 |
+
"eval_loss": 0.02658195048570633,
|
540 |
+
"eval_runtime": 0.1301,
|
541 |
+
"eval_samples_per_second": 69.158,
|
542 |
+
"eval_steps_per_second": 7.684,
|
543 |
+
"step": 56
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 57.0,
|
547 |
+
"eval_accuracy": 1.0,
|
548 |
+
"eval_loss": 0.026072878390550613,
|
549 |
+
"eval_runtime": 0.137,
|
550 |
+
"eval_samples_per_second": 65.67,
|
551 |
+
"eval_steps_per_second": 7.297,
|
552 |
+
"step": 57
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 58.0,
|
556 |
+
"eval_accuracy": 1.0,
|
557 |
+
"eval_loss": 0.026831289753317833,
|
558 |
+
"eval_runtime": 0.1385,
|
559 |
+
"eval_samples_per_second": 64.98,
|
560 |
+
"eval_steps_per_second": 7.22,
|
561 |
+
"step": 58
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 59.0,
|
565 |
+
"eval_accuracy": 1.0,
|
566 |
+
"eval_loss": 0.025475598871707916,
|
567 |
+
"eval_runtime": 0.1444,
|
568 |
+
"eval_samples_per_second": 62.335,
|
569 |
+
"eval_steps_per_second": 6.926,
|
570 |
+
"step": 59
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 60.0,
|
574 |
+
"learning_rate": 1e-05,
|
575 |
+
"loss": 0.0293,
|
576 |
+
"step": 60
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 60.0,
|
580 |
+
"eval_accuracy": 1.0,
|
581 |
+
"eval_loss": 0.0254827793687582,
|
582 |
+
"eval_runtime": 0.1362,
|
583 |
+
"eval_samples_per_second": 66.058,
|
584 |
+
"eval_steps_per_second": 7.34,
|
585 |
+
"step": 60
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 61.0,
|
589 |
+
"eval_accuracy": 1.0,
|
590 |
+
"eval_loss": 0.02456030063331127,
|
591 |
+
"eval_runtime": 0.1455,
|
592 |
+
"eval_samples_per_second": 61.856,
|
593 |
+
"eval_steps_per_second": 6.873,
|
594 |
+
"step": 61
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 62.0,
|
598 |
+
"eval_accuracy": 1.0,
|
599 |
+
"eval_loss": 0.025619812309741974,
|
600 |
+
"eval_runtime": 0.14,
|
601 |
+
"eval_samples_per_second": 64.308,
|
602 |
+
"eval_steps_per_second": 7.145,
|
603 |
+
"step": 62
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 63.0,
|
607 |
+
"eval_accuracy": 1.0,
|
608 |
+
"eval_loss": 0.024687545374035835,
|
609 |
+
"eval_runtime": 0.1341,
|
610 |
+
"eval_samples_per_second": 67.136,
|
611 |
+
"eval_steps_per_second": 7.46,
|
612 |
+
"step": 63
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 64.0,
|
616 |
+
"eval_accuracy": 1.0,
|
617 |
+
"eval_loss": 0.02405836433172226,
|
618 |
+
"eval_runtime": 0.1463,
|
619 |
+
"eval_samples_per_second": 61.502,
|
620 |
+
"eval_steps_per_second": 6.834,
|
621 |
+
"step": 64
|
622 |
+
},
|
623 |
+
{
|
624 |
+
"epoch": 65.0,
|
625 |
+
"eval_accuracy": 1.0,
|
626 |
+
"eval_loss": 0.02410367876291275,
|
627 |
+
"eval_runtime": 0.1336,
|
628 |
+
"eval_samples_per_second": 67.368,
|
629 |
+
"eval_steps_per_second": 7.485,
|
630 |
+
"step": 65
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 66.0,
|
634 |
+
"eval_accuracy": 1.0,
|
635 |
+
"eval_loss": 0.02341434732079506,
|
636 |
+
"eval_runtime": 0.1326,
|
637 |
+
"eval_samples_per_second": 67.862,
|
638 |
+
"eval_steps_per_second": 7.54,
|
639 |
+
"step": 66
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 67.0,
|
643 |
+
"eval_accuracy": 1.0,
|
644 |
+
"eval_loss": 0.023647040128707886,
|
645 |
+
"eval_runtime": 0.1325,
|
646 |
+
"eval_samples_per_second": 67.902,
|
647 |
+
"eval_steps_per_second": 7.545,
|
648 |
+
"step": 67
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 68.0,
|
652 |
+
"eval_accuracy": 1.0,
|
653 |
+
"eval_loss": 0.02281663753092289,
|
654 |
+
"eval_runtime": 0.1334,
|
655 |
+
"eval_samples_per_second": 67.464,
|
656 |
+
"eval_steps_per_second": 7.496,
|
657 |
+
"step": 68
|
658 |
+
},
|
659 |
+
{
|
660 |
+
"epoch": 69.0,
|
661 |
+
"eval_accuracy": 1.0,
|
662 |
+
"eval_loss": 0.023257458582520485,
|
663 |
+
"eval_runtime": 0.1308,
|
664 |
+
"eval_samples_per_second": 68.813,
|
665 |
+
"eval_steps_per_second": 7.646,
|
666 |
+
"step": 69
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 70.0,
|
670 |
+
"learning_rate": 8.333333333333334e-06,
|
671 |
+
"loss": 0.0256,
|
672 |
+
"step": 70
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 70.0,
|
676 |
+
"eval_accuracy": 1.0,
|
677 |
+
"eval_loss": 0.022747112438082695,
|
678 |
+
"eval_runtime": 0.1352,
|
679 |
+
"eval_samples_per_second": 66.563,
|
680 |
+
"eval_steps_per_second": 7.396,
|
681 |
+
"step": 70
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 71.0,
|
685 |
+
"eval_accuracy": 1.0,
|
686 |
+
"eval_loss": 0.022654477506875992,
|
687 |
+
"eval_runtime": 0.138,
|
688 |
+
"eval_samples_per_second": 65.228,
|
689 |
+
"eval_steps_per_second": 7.248,
|
690 |
+
"step": 71
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 72.0,
|
694 |
+
"eval_accuracy": 1.0,
|
695 |
+
"eval_loss": 0.022984912618994713,
|
696 |
+
"eval_runtime": 0.1431,
|
697 |
+
"eval_samples_per_second": 62.899,
|
698 |
+
"eval_steps_per_second": 6.989,
|
699 |
+
"step": 72
|
700 |
+
},
|
701 |
+
{
|
702 |
+
"epoch": 73.0,
|
703 |
+
"eval_accuracy": 1.0,
|
704 |
+
"eval_loss": 0.022151878103613853,
|
705 |
+
"eval_runtime": 0.137,
|
706 |
+
"eval_samples_per_second": 65.705,
|
707 |
+
"eval_steps_per_second": 7.301,
|
708 |
+
"step": 73
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 74.0,
|
712 |
+
"eval_accuracy": 1.0,
|
713 |
+
"eval_loss": 0.022022059187293053,
|
714 |
+
"eval_runtime": 0.1274,
|
715 |
+
"eval_samples_per_second": 70.646,
|
716 |
+
"eval_steps_per_second": 7.85,
|
717 |
+
"step": 74
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 75.0,
|
721 |
+
"eval_accuracy": 1.0,
|
722 |
+
"eval_loss": 0.02206052653491497,
|
723 |
+
"eval_runtime": 0.1313,
|
724 |
+
"eval_samples_per_second": 68.558,
|
725 |
+
"eval_steps_per_second": 7.618,
|
726 |
+
"step": 75
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 76.0,
|
730 |
+
"eval_accuracy": 1.0,
|
731 |
+
"eval_loss": 0.021884923800826073,
|
732 |
+
"eval_runtime": 0.1275,
|
733 |
+
"eval_samples_per_second": 70.602,
|
734 |
+
"eval_steps_per_second": 7.845,
|
735 |
+
"step": 76
|
736 |
+
},
|
737 |
+
{
|
738 |
+
"epoch": 77.0,
|
739 |
+
"eval_accuracy": 1.0,
|
740 |
+
"eval_loss": 0.021523766219615936,
|
741 |
+
"eval_runtime": 0.1281,
|
742 |
+
"eval_samples_per_second": 70.269,
|
743 |
+
"eval_steps_per_second": 7.808,
|
744 |
+
"step": 77
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 78.0,
|
748 |
+
"eval_accuracy": 1.0,
|
749 |
+
"eval_loss": 0.02103014849126339,
|
750 |
+
"eval_runtime": 0.1284,
|
751 |
+
"eval_samples_per_second": 70.078,
|
752 |
+
"eval_steps_per_second": 7.786,
|
753 |
+
"step": 78
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 79.0,
|
757 |
+
"eval_accuracy": 1.0,
|
758 |
+
"eval_loss": 0.020907729864120483,
|
759 |
+
"eval_runtime": 0.1402,
|
760 |
+
"eval_samples_per_second": 64.198,
|
761 |
+
"eval_steps_per_second": 7.133,
|
762 |
+
"step": 79
|
763 |
+
},
|
764 |
+
{
|
765 |
+
"epoch": 80.0,
|
766 |
+
"learning_rate": 6.666666666666667e-06,
|
767 |
+
"loss": 0.0234,
|
768 |
+
"step": 80
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 80.0,
|
772 |
+
"eval_accuracy": 1.0,
|
773 |
+
"eval_loss": 0.021248452365398407,
|
774 |
+
"eval_runtime": 0.138,
|
775 |
+
"eval_samples_per_second": 65.201,
|
776 |
+
"eval_steps_per_second": 7.245,
|
777 |
+
"step": 80
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 81.0,
|
781 |
+
"eval_accuracy": 1.0,
|
782 |
+
"eval_loss": 0.02118901163339615,
|
783 |
+
"eval_runtime": 1.9564,
|
784 |
+
"eval_samples_per_second": 4.6,
|
785 |
+
"eval_steps_per_second": 0.511,
|
786 |
+
"step": 81
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 82.0,
|
790 |
+
"eval_accuracy": 1.0,
|
791 |
+
"eval_loss": 0.020638220012187958,
|
792 |
+
"eval_runtime": 0.1298,
|
793 |
+
"eval_samples_per_second": 69.343,
|
794 |
+
"eval_steps_per_second": 7.705,
|
795 |
+
"step": 82
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 83.0,
|
799 |
+
"eval_accuracy": 1.0,
|
800 |
+
"eval_loss": 0.0210101418197155,
|
801 |
+
"eval_runtime": 0.1332,
|
802 |
+
"eval_samples_per_second": 67.563,
|
803 |
+
"eval_steps_per_second": 7.507,
|
804 |
+
"step": 83
|
805 |
+
},
|
806 |
+
{
|
807 |
+
"epoch": 84.0,
|
808 |
+
"eval_accuracy": 1.0,
|
809 |
+
"eval_loss": 0.020395496860146523,
|
810 |
+
"eval_runtime": 0.1435,
|
811 |
+
"eval_samples_per_second": 62.705,
|
812 |
+
"eval_steps_per_second": 6.967,
|
813 |
+
"step": 84
|
814 |
+
},
|
815 |
+
{
|
816 |
+
"epoch": 85.0,
|
817 |
+
"eval_accuracy": 1.0,
|
818 |
+
"eval_loss": 0.02046314626932144,
|
819 |
+
"eval_runtime": 0.1272,
|
820 |
+
"eval_samples_per_second": 70.782,
|
821 |
+
"eval_steps_per_second": 7.865,
|
822 |
+
"step": 85
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 86.0,
|
826 |
+
"eval_accuracy": 1.0,
|
827 |
+
"eval_loss": 0.020390955731272697,
|
828 |
+
"eval_runtime": 0.131,
|
829 |
+
"eval_samples_per_second": 68.683,
|
830 |
+
"eval_steps_per_second": 7.631,
|
831 |
+
"step": 86
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 87.0,
|
835 |
+
"eval_accuracy": 1.0,
|
836 |
+
"eval_loss": 0.02033914439380169,
|
837 |
+
"eval_runtime": 0.1316,
|
838 |
+
"eval_samples_per_second": 68.365,
|
839 |
+
"eval_steps_per_second": 7.596,
|
840 |
+
"step": 87
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 88.0,
|
844 |
+
"eval_accuracy": 1.0,
|
845 |
+
"eval_loss": 0.02000078745186329,
|
846 |
+
"eval_runtime": 0.1264,
|
847 |
+
"eval_samples_per_second": 71.187,
|
848 |
+
"eval_steps_per_second": 7.91,
|
849 |
+
"step": 88
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 89.0,
|
853 |
+
"eval_accuracy": 1.0,
|
854 |
+
"eval_loss": 0.020298130810260773,
|
855 |
+
"eval_runtime": 0.1425,
|
856 |
+
"eval_samples_per_second": 63.148,
|
857 |
+
"eval_steps_per_second": 7.016,
|
858 |
+
"step": 89
|
859 |
+
},
|
860 |
+
{
|
861 |
+
"epoch": 90.0,
|
862 |
+
"learning_rate": 5e-06,
|
863 |
+
"loss": 0.0218,
|
864 |
+
"step": 90
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 90.0,
|
868 |
+
"eval_accuracy": 1.0,
|
869 |
+
"eval_loss": 0.01956954225897789,
|
870 |
+
"eval_runtime": 0.1265,
|
871 |
+
"eval_samples_per_second": 71.16,
|
872 |
+
"eval_steps_per_second": 7.907,
|
873 |
+
"step": 90
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 91.0,
|
877 |
+
"eval_accuracy": 1.0,
|
878 |
+
"eval_loss": 0.01989126391708851,
|
879 |
+
"eval_runtime": 0.1281,
|
880 |
+
"eval_samples_per_second": 70.264,
|
881 |
+
"eval_steps_per_second": 7.807,
|
882 |
+
"step": 91
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 92.0,
|
886 |
+
"eval_accuracy": 1.0,
|
887 |
+
"eval_loss": 0.019824812188744545,
|
888 |
+
"eval_runtime": 0.138,
|
889 |
+
"eval_samples_per_second": 65.198,
|
890 |
+
"eval_steps_per_second": 7.244,
|
891 |
+
"step": 92
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 93.0,
|
895 |
+
"eval_accuracy": 1.0,
|
896 |
+
"eval_loss": 0.019585557281970978,
|
897 |
+
"eval_runtime": 0.1356,
|
898 |
+
"eval_samples_per_second": 66.353,
|
899 |
+
"eval_steps_per_second": 7.373,
|
900 |
+
"step": 93
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 94.0,
|
904 |
+
"eval_accuracy": 1.0,
|
905 |
+
"eval_loss": 0.019477464258670807,
|
906 |
+
"eval_runtime": 6.0159,
|
907 |
+
"eval_samples_per_second": 1.496,
|
908 |
+
"eval_steps_per_second": 0.166,
|
909 |
+
"step": 94
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 95.0,
|
913 |
+
"eval_accuracy": 1.0,
|
914 |
+
"eval_loss": 0.019807875156402588,
|
915 |
+
"eval_runtime": 0.1312,
|
916 |
+
"eval_samples_per_second": 68.571,
|
917 |
+
"eval_steps_per_second": 7.619,
|
918 |
+
"step": 95
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 96.0,
|
922 |
+
"eval_accuracy": 1.0,
|
923 |
+
"eval_loss": 0.019691793248057365,
|
924 |
+
"eval_runtime": 0.1276,
|
925 |
+
"eval_samples_per_second": 70.51,
|
926 |
+
"eval_steps_per_second": 7.834,
|
927 |
+
"step": 96
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 97.0,
|
931 |
+
"eval_accuracy": 1.0,
|
932 |
+
"eval_loss": 0.01933690346777439,
|
933 |
+
"eval_runtime": 0.1347,
|
934 |
+
"eval_samples_per_second": 66.808,
|
935 |
+
"eval_steps_per_second": 7.423,
|
936 |
+
"step": 97
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 98.0,
|
940 |
+
"eval_accuracy": 1.0,
|
941 |
+
"eval_loss": 0.019513830542564392,
|
942 |
+
"eval_runtime": 0.1261,
|
943 |
+
"eval_samples_per_second": 71.367,
|
944 |
+
"eval_steps_per_second": 7.93,
|
945 |
+
"step": 98
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 99.0,
|
949 |
+
"eval_accuracy": 1.0,
|
950 |
+
"eval_loss": 0.019352566450834274,
|
951 |
+
"eval_runtime": 0.1395,
|
952 |
+
"eval_samples_per_second": 64.504,
|
953 |
+
"eval_steps_per_second": 7.167,
|
954 |
+
"step": 99
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 100.0,
|
958 |
+
"learning_rate": 3.3333333333333333e-06,
|
959 |
+
"loss": 0.0208,
|
960 |
+
"step": 100
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 100.0,
|
964 |
+
"eval_accuracy": 1.0,
|
965 |
+
"eval_loss": 0.01923985406756401,
|
966 |
+
"eval_runtime": 5.8953,
|
967 |
+
"eval_samples_per_second": 1.527,
|
968 |
+
"eval_steps_per_second": 0.17,
|
969 |
+
"step": 100
|
970 |
+
},
|
971 |
+
{
|
972 |
+
"epoch": 101.0,
|
973 |
+
"eval_accuracy": 1.0,
|
974 |
+
"eval_loss": 0.01895654946565628,
|
975 |
+
"eval_runtime": 6.9547,
|
976 |
+
"eval_samples_per_second": 1.294,
|
977 |
+
"eval_steps_per_second": 0.144,
|
978 |
+
"step": 101
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 102.0,
|
982 |
+
"eval_accuracy": 1.0,
|
983 |
+
"eval_loss": 0.018784379586577415,
|
984 |
+
"eval_runtime": 0.153,
|
985 |
+
"eval_samples_per_second": 58.805,
|
986 |
+
"eval_steps_per_second": 6.534,
|
987 |
+
"step": 102
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 103.0,
|
991 |
+
"eval_accuracy": 1.0,
|
992 |
+
"eval_loss": 0.019066665321588516,
|
993 |
+
"eval_runtime": 0.1378,
|
994 |
+
"eval_samples_per_second": 65.292,
|
995 |
+
"eval_steps_per_second": 7.255,
|
996 |
+
"step": 103
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 104.0,
|
1000 |
+
"eval_accuracy": 1.0,
|
1001 |
+
"eval_loss": 0.01931922882795334,
|
1002 |
+
"eval_runtime": 0.1359,
|
1003 |
+
"eval_samples_per_second": 66.226,
|
1004 |
+
"eval_steps_per_second": 7.358,
|
1005 |
+
"step": 104
|
1006 |
+
},
|
1007 |
+
{
|
1008 |
+
"epoch": 105.0,
|
1009 |
+
"eval_accuracy": 1.0,
|
1010 |
+
"eval_loss": 0.019273685291409492,
|
1011 |
+
"eval_runtime": 0.1487,
|
1012 |
+
"eval_samples_per_second": 60.506,
|
1013 |
+
"eval_steps_per_second": 6.723,
|
1014 |
+
"step": 105
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 106.0,
|
1018 |
+
"eval_accuracy": 1.0,
|
1019 |
+
"eval_loss": 0.018980564549565315,
|
1020 |
+
"eval_runtime": 0.1426,
|
1021 |
+
"eval_samples_per_second": 63.122,
|
1022 |
+
"eval_steps_per_second": 7.014,
|
1023 |
+
"step": 106
|
1024 |
+
},
|
1025 |
+
{
|
1026 |
+
"epoch": 107.0,
|
1027 |
+
"eval_accuracy": 1.0,
|
1028 |
+
"eval_loss": 0.01911451853811741,
|
1029 |
+
"eval_runtime": 5.4956,
|
1030 |
+
"eval_samples_per_second": 1.638,
|
1031 |
+
"eval_steps_per_second": 0.182,
|
1032 |
+
"step": 107
|
1033 |
+
},
|
1034 |
+
{
|
1035 |
+
"epoch": 108.0,
|
1036 |
+
"eval_accuracy": 1.0,
|
1037 |
+
"eval_loss": 0.018641483038663864,
|
1038 |
+
"eval_runtime": 0.156,
|
1039 |
+
"eval_samples_per_second": 57.704,
|
1040 |
+
"eval_steps_per_second": 6.412,
|
1041 |
+
"step": 108
|
1042 |
+
},
|
1043 |
+
{
|
1044 |
+
"epoch": 109.0,
|
1045 |
+
"eval_accuracy": 1.0,
|
1046 |
+
"eval_loss": 0.018821991980075836,
|
1047 |
+
"eval_runtime": 0.191,
|
1048 |
+
"eval_samples_per_second": 47.114,
|
1049 |
+
"eval_steps_per_second": 5.235,
|
1050 |
+
"step": 109
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 110.0,
|
1054 |
+
"learning_rate": 1.6666666666666667e-06,
|
1055 |
+
"loss": 0.0202,
|
1056 |
+
"step": 110
|
1057 |
+
},
|
1058 |
+
{
|
1059 |
+
"epoch": 110.0,
|
1060 |
+
"eval_accuracy": 1.0,
|
1061 |
+
"eval_loss": 0.018715376034379005,
|
1062 |
+
"eval_runtime": 5.378,
|
1063 |
+
"eval_samples_per_second": 1.673,
|
1064 |
+
"eval_steps_per_second": 0.186,
|
1065 |
+
"step": 110
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 111.0,
|
1069 |
+
"eval_accuracy": 1.0,
|
1070 |
+
"eval_loss": 0.01905597560107708,
|
1071 |
+
"eval_runtime": 0.1393,
|
1072 |
+
"eval_samples_per_second": 64.614,
|
1073 |
+
"eval_steps_per_second": 7.179,
|
1074 |
+
"step": 111
|
1075 |
+
},
|
1076 |
+
{
|
1077 |
+
"epoch": 112.0,
|
1078 |
+
"eval_accuracy": 1.0,
|
1079 |
+
"eval_loss": 0.01877736486494541,
|
1080 |
+
"eval_runtime": 0.1478,
|
1081 |
+
"eval_samples_per_second": 60.896,
|
1082 |
+
"eval_steps_per_second": 6.766,
|
1083 |
+
"step": 112
|
1084 |
+
},
|
1085 |
+
{
|
1086 |
+
"epoch": 113.0,
|
1087 |
+
"eval_accuracy": 1.0,
|
1088 |
+
"eval_loss": 0.01854606531560421,
|
1089 |
+
"eval_runtime": 5.9004,
|
1090 |
+
"eval_samples_per_second": 1.525,
|
1091 |
+
"eval_steps_per_second": 0.169,
|
1092 |
+
"step": 113
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 114.0,
|
1096 |
+
"eval_accuracy": 1.0,
|
1097 |
+
"eval_loss": 0.018782339990139008,
|
1098 |
+
"eval_runtime": 0.141,
|
1099 |
+
"eval_samples_per_second": 63.82,
|
1100 |
+
"eval_steps_per_second": 7.091,
|
1101 |
+
"step": 114
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 115.0,
|
1105 |
+
"eval_accuracy": 1.0,
|
1106 |
+
"eval_loss": 0.018274879083037376,
|
1107 |
+
"eval_runtime": 6.7975,
|
1108 |
+
"eval_samples_per_second": 1.324,
|
1109 |
+
"eval_steps_per_second": 0.147,
|
1110 |
+
"step": 115
|
1111 |
+
},
|
1112 |
+
{
|
1113 |
+
"epoch": 116.0,
|
1114 |
+
"eval_accuracy": 1.0,
|
1115 |
+
"eval_loss": 0.018706435337662697,
|
1116 |
+
"eval_runtime": 0.1365,
|
1117 |
+
"eval_samples_per_second": 65.941,
|
1118 |
+
"eval_steps_per_second": 7.327,
|
1119 |
+
"step": 116
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 117.0,
|
1123 |
+
"eval_accuracy": 1.0,
|
1124 |
+
"eval_loss": 0.018547622486948967,
|
1125 |
+
"eval_runtime": 2.2842,
|
1126 |
+
"eval_samples_per_second": 3.94,
|
1127 |
+
"eval_steps_per_second": 0.438,
|
1128 |
+
"step": 117
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 118.0,
|
1132 |
+
"eval_accuracy": 1.0,
|
1133 |
+
"eval_loss": 0.018446724861860275,
|
1134 |
+
"eval_runtime": 6.0918,
|
1135 |
+
"eval_samples_per_second": 1.477,
|
1136 |
+
"eval_steps_per_second": 0.164,
|
1137 |
+
"step": 118
|
1138 |
+
},
|
1139 |
+
{
|
1140 |
+
"epoch": 119.0,
|
1141 |
+
"eval_accuracy": 1.0,
|
1142 |
+
"eval_loss": 0.018762176856398582,
|
1143 |
+
"eval_runtime": 0.1343,
|
1144 |
+
"eval_samples_per_second": 67.039,
|
1145 |
+
"eval_steps_per_second": 7.449,
|
1146 |
+
"step": 119
|
1147 |
+
},
|
1148 |
+
{
|
1149 |
+
"epoch": 120.0,
|
1150 |
+
"learning_rate": 0.0,
|
1151 |
+
"loss": 0.0197,
|
1152 |
+
"step": 120
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"epoch": 120.0,
|
1156 |
+
"eval_accuracy": 1.0,
|
1157 |
+
"eval_loss": 0.01847647689282894,
|
1158 |
+
"eval_runtime": 0.1403,
|
1159 |
+
"eval_samples_per_second": 64.13,
|
1160 |
+
"eval_steps_per_second": 7.126,
|
1161 |
+
"step": 120
|
1162 |
+
},
|
1163 |
+
{
|
1164 |
+
"epoch": 120.0,
|
1165 |
+
"step": 120,
|
1166 |
+
"total_flos": 4.7425097644130304e+17,
|
1167 |
+
"train_loss": 0.07813952726622422,
|
1168 |
+
"train_runtime": 1030.5859,
|
1169 |
+
"train_samples_per_second": 5.938,
|
1170 |
+
"train_steps_per_second": 0.116
|
1171 |
+
}
|
1172 |
+
],
|
1173 |
+
"logging_steps": 10,
|
1174 |
+
"max_steps": 120,
|
1175 |
+
"num_input_tokens_seen": 0,
|
1176 |
+
"num_train_epochs": 120,
|
1177 |
+
"save_steps": 500,
|
1178 |
+
"total_flos": 4.7425097644130304e+17,
|
1179 |
+
"train_batch_size": 200,
|
1180 |
+
"trial_name": null,
|
1181 |
+
"trial_params": null
|
1182 |
+
}
|