End of training
Browse files- README.md +13 -13
- all_results.json +18 -0
- eval_results.json +12 -0
- train_results.json +9 -0
- trainer_state.json +496 -0
README.md
CHANGED
|
@@ -5,7 +5,7 @@ base_model: bert-base-uncased
|
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
-
- datasaur-
|
| 9 |
metrics:
|
| 10 |
- precision
|
| 11 |
- recall
|
|
@@ -18,24 +18,24 @@ model-index:
|
|
| 18 |
name: Token Classification
|
| 19 |
type: token-classification
|
| 20 |
dataset:
|
| 21 |
-
name: datasaur-
|
| 22 |
-
type: datasaur-
|
| 23 |
config: default
|
| 24 |
split: validation
|
| 25 |
args: default
|
| 26 |
metrics:
|
| 27 |
- name: Precision
|
| 28 |
type: precision
|
| 29 |
-
value: 0.
|
| 30 |
- name: Recall
|
| 31 |
type: recall
|
| 32 |
-
value: 0.
|
| 33 |
- name: F1
|
| 34 |
type: f1
|
| 35 |
-
value: 0.
|
| 36 |
- name: Accuracy
|
| 37 |
type: accuracy
|
| 38 |
-
value: 0.
|
| 39 |
---
|
| 40 |
|
| 41 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -43,13 +43,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 43 |
|
| 44 |
# autotrain-radesky-lab-span-v1
|
| 45 |
|
| 46 |
-
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the datasaur-
|
| 47 |
It achieves the following results on the evaluation set:
|
| 48 |
-
- Loss: 0.
|
| 49 |
-
- Precision: 0.
|
| 50 |
-
- Recall: 0.
|
| 51 |
-
- F1: 0.
|
| 52 |
-
- Accuracy: 0.
|
| 53 |
|
| 54 |
## Model description
|
| 55 |
|
|
|
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
+
- datasaur-dev/datasaur-MTFiZjUwM2Q-ZWJiZDRmNGI
|
| 9 |
metrics:
|
| 10 |
- precision
|
| 11 |
- recall
|
|
|
|
| 18 |
name: Token Classification
|
| 19 |
type: token-classification
|
| 20 |
dataset:
|
| 21 |
+
name: datasaur-dev/datasaur-MTFiZjUwM2Q-ZWJiZDRmNGI
|
| 22 |
+
type: datasaur-dev/datasaur-MTFiZjUwM2Q-ZWJiZDRmNGI
|
| 23 |
config: default
|
| 24 |
split: validation
|
| 25 |
args: default
|
| 26 |
metrics:
|
| 27 |
- name: Precision
|
| 28 |
type: precision
|
| 29 |
+
value: 0.7853658536585366
|
| 30 |
- name: Recall
|
| 31 |
type: recall
|
| 32 |
+
value: 0.8385416666666666
|
| 33 |
- name: F1
|
| 34 |
type: f1
|
| 35 |
+
value: 0.8110831234256927
|
| 36 |
- name: Accuracy
|
| 37 |
type: accuracy
|
| 38 |
+
value: 0.9709519365375642
|
| 39 |
---
|
| 40 |
|
| 41 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 43 |
|
| 44 |
# autotrain-radesky-lab-span-v1
|
| 45 |
|
| 46 |
+
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the datasaur-dev/datasaur-MTFiZjUwM2Q-ZWJiZDRmNGI dataset.
|
| 47 |
It achieves the following results on the evaluation set:
|
| 48 |
+
- Loss: 0.2518
|
| 49 |
+
- Precision: 0.7854
|
| 50 |
+
- Recall: 0.8385
|
| 51 |
+
- F1: 0.8111
|
| 52 |
+
- Accuracy: 0.9710
|
| 53 |
|
| 54 |
## Model description
|
| 55 |
|
all_results.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.9709519365375642,
|
| 4 |
+
"eval_f1": 0.8110831234256927,
|
| 5 |
+
"eval_loss": 0.2518477439880371,
|
| 6 |
+
"eval_precision": 0.7853658536585366,
|
| 7 |
+
"eval_recall": 0.8385416666666666,
|
| 8 |
+
"eval_runtime": 3.2853,
|
| 9 |
+
"eval_samples": 909,
|
| 10 |
+
"eval_samples_per_second": 276.687,
|
| 11 |
+
"eval_steps_per_second": 34.7,
|
| 12 |
+
"total_flos": 2472027466589526.0,
|
| 13 |
+
"train_loss": 0.03293397539264553,
|
| 14 |
+
"train_runtime": 1492.2693,
|
| 15 |
+
"train_samples": 3635,
|
| 16 |
+
"train_samples_per_second": 60.897,
|
| 17 |
+
"train_steps_per_second": 7.623
|
| 18 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.9709519365375642,
|
| 4 |
+
"eval_f1": 0.8110831234256927,
|
| 5 |
+
"eval_loss": 0.2518477439880371,
|
| 6 |
+
"eval_precision": 0.7853658536585366,
|
| 7 |
+
"eval_recall": 0.8385416666666666,
|
| 8 |
+
"eval_runtime": 3.2853,
|
| 9 |
+
"eval_samples": 909,
|
| 10 |
+
"eval_samples_per_second": 276.687,
|
| 11 |
+
"eval_steps_per_second": 34.7
|
| 12 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"total_flos": 2472027466589526.0,
|
| 4 |
+
"train_loss": 0.03293397539264553,
|
| 5 |
+
"train_runtime": 1492.2693,
|
| 6 |
+
"train_samples": 3635,
|
| 7 |
+
"train_samples_per_second": 60.897,
|
| 8 |
+
"train_steps_per_second": 7.623
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,496 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8110831234256927,
|
| 3 |
+
"best_model_checkpoint": "checkpoint-v1/checkpoint-10465",
|
| 4 |
+
"epoch": 25.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 11375,
|
| 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": 0.9440037330844611,
|
| 14 |
+
"eval_f1": 0.48913043478260865,
|
| 15 |
+
"eval_loss": 0.25645163655281067,
|
| 16 |
+
"eval_precision": 0.5113636363636364,
|
| 17 |
+
"eval_recall": 0.46875,
|
| 18 |
+
"eval_runtime": 3.4617,
|
| 19 |
+
"eval_samples_per_second": 262.585,
|
| 20 |
+
"eval_steps_per_second": 32.931,
|
| 21 |
+
"step": 455
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 1.098901098901099,
|
| 25 |
+
"grad_norm": 3.66561222076416,
|
| 26 |
+
"learning_rate": 9.560439560439561e-05,
|
| 27 |
+
"loss": 0.2875,
|
| 28 |
+
"step": 500
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"eval_accuracy": 0.9238217452169856,
|
| 33 |
+
"eval_f1": 0.29100529100529104,
|
| 34 |
+
"eval_loss": 0.3291769027709961,
|
| 35 |
+
"eval_precision": 0.2956989247311828,
|
| 36 |
+
"eval_recall": 0.2864583333333333,
|
| 37 |
+
"eval_runtime": 3.3496,
|
| 38 |
+
"eval_samples_per_second": 271.38,
|
| 39 |
+
"eval_steps_per_second": 34.034,
|
| 40 |
+
"step": 910
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.197802197802198,
|
| 44 |
+
"grad_norm": 4.647622585296631,
|
| 45 |
+
"learning_rate": 9.12087912087912e-05,
|
| 46 |
+
"loss": 0.1173,
|
| 47 |
+
"step": 1000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.0,
|
| 51 |
+
"eval_accuracy": 0.9531031264582361,
|
| 52 |
+
"eval_f1": 0.5489443378119002,
|
| 53 |
+
"eval_loss": 0.19305621087551117,
|
| 54 |
+
"eval_precision": 0.43465045592705165,
|
| 55 |
+
"eval_recall": 0.7447916666666666,
|
| 56 |
+
"eval_runtime": 3.3449,
|
| 57 |
+
"eval_samples_per_second": 271.756,
|
| 58 |
+
"eval_steps_per_second": 34.082,
|
| 59 |
+
"step": 1365
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.2967032967032965,
|
| 63 |
+
"grad_norm": 0.18239367008209229,
|
| 64 |
+
"learning_rate": 8.681318681318682e-05,
|
| 65 |
+
"loss": 0.0945,
|
| 66 |
+
"step": 1500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.0,
|
| 70 |
+
"eval_accuracy": 0.957769482034531,
|
| 71 |
+
"eval_f1": 0.603448275862069,
|
| 72 |
+
"eval_loss": 0.1780281662940979,
|
| 73 |
+
"eval_precision": 0.5147058823529411,
|
| 74 |
+
"eval_recall": 0.7291666666666666,
|
| 75 |
+
"eval_runtime": 3.3629,
|
| 76 |
+
"eval_samples_per_second": 270.306,
|
| 77 |
+
"eval_steps_per_second": 33.9,
|
| 78 |
+
"step": 1820
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.395604395604396,
|
| 82 |
+
"grad_norm": 0.10920047760009766,
|
| 83 |
+
"learning_rate": 8.241758241758242e-05,
|
| 84 |
+
"loss": 0.0559,
|
| 85 |
+
"step": 2000
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 5.0,
|
| 89 |
+
"eval_accuracy": 0.9591693887074195,
|
| 90 |
+
"eval_f1": 0.6343612334801763,
|
| 91 |
+
"eval_loss": 0.19242332875728607,
|
| 92 |
+
"eval_precision": 0.549618320610687,
|
| 93 |
+
"eval_recall": 0.75,
|
| 94 |
+
"eval_runtime": 3.3708,
|
| 95 |
+
"eval_samples_per_second": 269.67,
|
| 96 |
+
"eval_steps_per_second": 33.82,
|
| 97 |
+
"step": 2275
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 5.4945054945054945,
|
| 101 |
+
"grad_norm": 0.08735374361276627,
|
| 102 |
+
"learning_rate": 7.802197802197802e-05,
|
| 103 |
+
"loss": 0.0412,
|
| 104 |
+
"step": 2500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.0,
|
| 108 |
+
"eval_accuracy": 0.9653523098460103,
|
| 109 |
+
"eval_f1": 0.7132530120481928,
|
| 110 |
+
"eval_loss": 0.16734345257282257,
|
| 111 |
+
"eval_precision": 0.6636771300448431,
|
| 112 |
+
"eval_recall": 0.7708333333333334,
|
| 113 |
+
"eval_runtime": 3.3682,
|
| 114 |
+
"eval_samples_per_second": 269.875,
|
| 115 |
+
"eval_steps_per_second": 33.846,
|
| 116 |
+
"step": 2730
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 6.593406593406593,
|
| 120 |
+
"grad_norm": 15.276603698730469,
|
| 121 |
+
"learning_rate": 7.362637362637363e-05,
|
| 122 |
+
"loss": 0.0309,
|
| 123 |
+
"step": 3000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 7.0,
|
| 127 |
+
"eval_accuracy": 0.9634857676154923,
|
| 128 |
+
"eval_f1": 0.6906474820143884,
|
| 129 |
+
"eval_loss": 0.1927553117275238,
|
| 130 |
+
"eval_precision": 0.64,
|
| 131 |
+
"eval_recall": 0.75,
|
| 132 |
+
"eval_runtime": 3.3642,
|
| 133 |
+
"eval_samples_per_second": 270.199,
|
| 134 |
+
"eval_steps_per_second": 33.886,
|
| 135 |
+
"step": 3185
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 7.6923076923076925,
|
| 139 |
+
"grad_norm": 0.018443187698721886,
|
| 140 |
+
"learning_rate": 6.923076923076924e-05,
|
| 141 |
+
"loss": 0.0231,
|
| 142 |
+
"step": 3500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.0,
|
| 146 |
+
"eval_accuracy": 0.9643023798413439,
|
| 147 |
+
"eval_f1": 0.6888361045130641,
|
| 148 |
+
"eval_loss": 0.1938297301530838,
|
| 149 |
+
"eval_precision": 0.6331877729257642,
|
| 150 |
+
"eval_recall": 0.7552083333333334,
|
| 151 |
+
"eval_runtime": 3.3614,
|
| 152 |
+
"eval_samples_per_second": 270.421,
|
| 153 |
+
"eval_steps_per_second": 33.914,
|
| 154 |
+
"step": 3640
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 8.791208791208792,
|
| 158 |
+
"grad_norm": 0.002568595577031374,
|
| 159 |
+
"learning_rate": 6.483516483516484e-05,
|
| 160 |
+
"loss": 0.0191,
|
| 161 |
+
"step": 4000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 9.0,
|
| 165 |
+
"eval_accuracy": 0.9669855342977135,
|
| 166 |
+
"eval_f1": 0.7194244604316546,
|
| 167 |
+
"eval_loss": 0.18560947477817535,
|
| 168 |
+
"eval_precision": 0.6666666666666666,
|
| 169 |
+
"eval_recall": 0.78125,
|
| 170 |
+
"eval_runtime": 3.3745,
|
| 171 |
+
"eval_samples_per_second": 269.37,
|
| 172 |
+
"eval_steps_per_second": 33.782,
|
| 173 |
+
"step": 4095
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 9.89010989010989,
|
| 177 |
+
"grad_norm": 0.022610006853938103,
|
| 178 |
+
"learning_rate": 6.043956043956044e-05,
|
| 179 |
+
"loss": 0.018,
|
| 180 |
+
"step": 4500
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 10.0,
|
| 184 |
+
"eval_accuracy": 0.9659356042930471,
|
| 185 |
+
"eval_f1": 0.7289719626168225,
|
| 186 |
+
"eval_loss": 0.20415335893630981,
|
| 187 |
+
"eval_precision": 0.6610169491525424,
|
| 188 |
+
"eval_recall": 0.8125,
|
| 189 |
+
"eval_runtime": 3.3613,
|
| 190 |
+
"eval_samples_per_second": 270.431,
|
| 191 |
+
"eval_steps_per_second": 33.915,
|
| 192 |
+
"step": 4550
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 10.989010989010989,
|
| 196 |
+
"grad_norm": 1.0466951131820679,
|
| 197 |
+
"learning_rate": 5.604395604395605e-05,
|
| 198 |
+
"loss": 0.0138,
|
| 199 |
+
"step": 5000
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 11.0,
|
| 203 |
+
"eval_accuracy": 0.9648856742883808,
|
| 204 |
+
"eval_f1": 0.7002288329519452,
|
| 205 |
+
"eval_loss": 0.22543059289455414,
|
| 206 |
+
"eval_precision": 0.6244897959183674,
|
| 207 |
+
"eval_recall": 0.796875,
|
| 208 |
+
"eval_runtime": 3.3542,
|
| 209 |
+
"eval_samples_per_second": 271.007,
|
| 210 |
+
"eval_steps_per_second": 33.988,
|
| 211 |
+
"step": 5005
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 12.0,
|
| 215 |
+
"eval_accuracy": 0.969318712085861,
|
| 216 |
+
"eval_f1": 0.7815533980582524,
|
| 217 |
+
"eval_loss": 0.2193489372730255,
|
| 218 |
+
"eval_precision": 0.7318181818181818,
|
| 219 |
+
"eval_recall": 0.8385416666666666,
|
| 220 |
+
"eval_runtime": 3.3347,
|
| 221 |
+
"eval_samples_per_second": 272.592,
|
| 222 |
+
"eval_steps_per_second": 34.186,
|
| 223 |
+
"step": 5460
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 12.087912087912088,
|
| 227 |
+
"grad_norm": 0.005279447417706251,
|
| 228 |
+
"learning_rate": 5.164835164835166e-05,
|
| 229 |
+
"loss": 0.0104,
|
| 230 |
+
"step": 5500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 13.0,
|
| 234 |
+
"eval_accuracy": 0.9643023798413439,
|
| 235 |
+
"eval_f1": 0.7242990654205607,
|
| 236 |
+
"eval_loss": 0.2286960929632187,
|
| 237 |
+
"eval_precision": 0.6567796610169492,
|
| 238 |
+
"eval_recall": 0.8072916666666666,
|
| 239 |
+
"eval_runtime": 3.3478,
|
| 240 |
+
"eval_samples_per_second": 271.521,
|
| 241 |
+
"eval_steps_per_second": 34.052,
|
| 242 |
+
"step": 5915
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 13.186813186813186,
|
| 246 |
+
"grad_norm": 0.02039416879415512,
|
| 247 |
+
"learning_rate": 4.7252747252747257e-05,
|
| 248 |
+
"loss": 0.0088,
|
| 249 |
+
"step": 6000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 14.0,
|
| 253 |
+
"eval_accuracy": 0.9682687820811946,
|
| 254 |
+
"eval_f1": 0.7553444180522564,
|
| 255 |
+
"eval_loss": 0.2258196324110031,
|
| 256 |
+
"eval_precision": 0.6943231441048034,
|
| 257 |
+
"eval_recall": 0.828125,
|
| 258 |
+
"eval_runtime": 3.3427,
|
| 259 |
+
"eval_samples_per_second": 271.935,
|
| 260 |
+
"eval_steps_per_second": 34.104,
|
| 261 |
+
"step": 6370
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 14.285714285714286,
|
| 265 |
+
"grad_norm": 0.00889052078127861,
|
| 266 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 267 |
+
"loss": 0.0052,
|
| 268 |
+
"step": 6500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 15.0,
|
| 272 |
+
"eval_accuracy": 0.9676854876341577,
|
| 273 |
+
"eval_f1": 0.7746835443037974,
|
| 274 |
+
"eval_loss": 0.23228472471237183,
|
| 275 |
+
"eval_precision": 0.7536945812807881,
|
| 276 |
+
"eval_recall": 0.796875,
|
| 277 |
+
"eval_runtime": 3.348,
|
| 278 |
+
"eval_samples_per_second": 271.503,
|
| 279 |
+
"eval_steps_per_second": 34.05,
|
| 280 |
+
"step": 6825
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 15.384615384615385,
|
| 284 |
+
"grad_norm": 0.0009669111459515989,
|
| 285 |
+
"learning_rate": 3.846153846153846e-05,
|
| 286 |
+
"loss": 0.0091,
|
| 287 |
+
"step": 7000
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 16.0,
|
| 291 |
+
"eval_accuracy": 0.9678021465235651,
|
| 292 |
+
"eval_f1": 0.762589928057554,
|
| 293 |
+
"eval_loss": 0.22262629866600037,
|
| 294 |
+
"eval_precision": 0.7066666666666667,
|
| 295 |
+
"eval_recall": 0.828125,
|
| 296 |
+
"eval_runtime": 3.3638,
|
| 297 |
+
"eval_samples_per_second": 270.226,
|
| 298 |
+
"eval_steps_per_second": 33.89,
|
| 299 |
+
"step": 7280
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 16.483516483516482,
|
| 303 |
+
"grad_norm": 0.0009670483996160328,
|
| 304 |
+
"learning_rate": 3.406593406593407e-05,
|
| 305 |
+
"loss": 0.0039,
|
| 306 |
+
"step": 7500
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 17.0,
|
| 310 |
+
"eval_accuracy": 0.9695520298646757,
|
| 311 |
+
"eval_f1": 0.7741935483870966,
|
| 312 |
+
"eval_loss": 0.2151697427034378,
|
| 313 |
+
"eval_precision": 0.7393364928909952,
|
| 314 |
+
"eval_recall": 0.8125,
|
| 315 |
+
"eval_runtime": 3.3394,
|
| 316 |
+
"eval_samples_per_second": 272.206,
|
| 317 |
+
"eval_steps_per_second": 34.138,
|
| 318 |
+
"step": 7735
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 17.582417582417584,
|
| 322 |
+
"grad_norm": 0.0011728066019713879,
|
| 323 |
+
"learning_rate": 2.9670329670329673e-05,
|
| 324 |
+
"loss": 0.006,
|
| 325 |
+
"step": 8000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 18.0,
|
| 329 |
+
"eval_accuracy": 0.9672188520765282,
|
| 330 |
+
"eval_f1": 0.7544303797468355,
|
| 331 |
+
"eval_loss": 0.26869261264801025,
|
| 332 |
+
"eval_precision": 0.7339901477832512,
|
| 333 |
+
"eval_recall": 0.7760416666666666,
|
| 334 |
+
"eval_runtime": 3.3437,
|
| 335 |
+
"eval_samples_per_second": 271.857,
|
| 336 |
+
"eval_steps_per_second": 34.094,
|
| 337 |
+
"step": 8190
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 18.681318681318682,
|
| 341 |
+
"grad_norm": 0.0027556538116186857,
|
| 342 |
+
"learning_rate": 2.5274725274725276e-05,
|
| 343 |
+
"loss": 0.0024,
|
| 344 |
+
"step": 8500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 19.0,
|
| 348 |
+
"eval_accuracy": 0.9689687354176388,
|
| 349 |
+
"eval_f1": 0.7722772277227722,
|
| 350 |
+
"eval_loss": 0.2464381903409958,
|
| 351 |
+
"eval_precision": 0.7358490566037735,
|
| 352 |
+
"eval_recall": 0.8125,
|
| 353 |
+
"eval_runtime": 3.3427,
|
| 354 |
+
"eval_samples_per_second": 271.937,
|
| 355 |
+
"eval_steps_per_second": 34.104,
|
| 356 |
+
"step": 8645
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 19.78021978021978,
|
| 360 |
+
"grad_norm": 0.000459605420473963,
|
| 361 |
+
"learning_rate": 2.0879120879120882e-05,
|
| 362 |
+
"loss": 0.0004,
|
| 363 |
+
"step": 9000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 20.0,
|
| 367 |
+
"eval_accuracy": 0.9694353709752683,
|
| 368 |
+
"eval_f1": 0.794044665012407,
|
| 369 |
+
"eval_loss": 0.2462543547153473,
|
| 370 |
+
"eval_precision": 0.7582938388625592,
|
| 371 |
+
"eval_recall": 0.8333333333333334,
|
| 372 |
+
"eval_runtime": 3.3329,
|
| 373 |
+
"eval_samples_per_second": 272.736,
|
| 374 |
+
"eval_steps_per_second": 34.205,
|
| 375 |
+
"step": 9100
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 20.87912087912088,
|
| 379 |
+
"grad_norm": 0.0002796004991978407,
|
| 380 |
+
"learning_rate": 1.6483516483516486e-05,
|
| 381 |
+
"loss": 0.0003,
|
| 382 |
+
"step": 9500
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 21.0,
|
| 386 |
+
"eval_accuracy": 0.9700186654223052,
|
| 387 |
+
"eval_f1": 0.806045340050378,
|
| 388 |
+
"eval_loss": 0.24657504260540009,
|
| 389 |
+
"eval_precision": 0.7804878048780488,
|
| 390 |
+
"eval_recall": 0.8333333333333334,
|
| 391 |
+
"eval_runtime": 3.3393,
|
| 392 |
+
"eval_samples_per_second": 272.21,
|
| 393 |
+
"eval_steps_per_second": 34.139,
|
| 394 |
+
"step": 9555
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 21.978021978021978,
|
| 398 |
+
"grad_norm": 0.0019895117729902267,
|
| 399 |
+
"learning_rate": 1.2087912087912089e-05,
|
| 400 |
+
"loss": 0.001,
|
| 401 |
+
"step": 10000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 22.0,
|
| 405 |
+
"eval_accuracy": 0.9706019598693421,
|
| 406 |
+
"eval_f1": 0.8020304568527918,
|
| 407 |
+
"eval_loss": 0.2513893246650696,
|
| 408 |
+
"eval_precision": 0.7821782178217822,
|
| 409 |
+
"eval_recall": 0.8229166666666666,
|
| 410 |
+
"eval_runtime": 3.3391,
|
| 411 |
+
"eval_samples_per_second": 272.232,
|
| 412 |
+
"eval_steps_per_second": 34.141,
|
| 413 |
+
"step": 10010
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 23.0,
|
| 417 |
+
"eval_accuracy": 0.9709519365375642,
|
| 418 |
+
"eval_f1": 0.8110831234256927,
|
| 419 |
+
"eval_loss": 0.2518477439880371,
|
| 420 |
+
"eval_precision": 0.7853658536585366,
|
| 421 |
+
"eval_recall": 0.8385416666666666,
|
| 422 |
+
"eval_runtime": 3.3453,
|
| 423 |
+
"eval_samples_per_second": 271.724,
|
| 424 |
+
"eval_steps_per_second": 34.078,
|
| 425 |
+
"step": 10465
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 23.076923076923077,
|
| 429 |
+
"grad_norm": 0.00038343301275745034,
|
| 430 |
+
"learning_rate": 7.692307692307694e-06,
|
| 431 |
+
"loss": 0.0002,
|
| 432 |
+
"step": 10500
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 24.0,
|
| 436 |
+
"eval_accuracy": 0.9704853009799347,
|
| 437 |
+
"eval_f1": 0.8050632911392406,
|
| 438 |
+
"eval_loss": 0.25856369733810425,
|
| 439 |
+
"eval_precision": 0.7832512315270936,
|
| 440 |
+
"eval_recall": 0.828125,
|
| 441 |
+
"eval_runtime": 3.3535,
|
| 442 |
+
"eval_samples_per_second": 271.062,
|
| 443 |
+
"eval_steps_per_second": 33.995,
|
| 444 |
+
"step": 10920
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 24.175824175824175,
|
| 448 |
+
"grad_norm": 0.0008269179961644113,
|
| 449 |
+
"learning_rate": 3.2967032967032968e-06,
|
| 450 |
+
"loss": 0.0002,
|
| 451 |
+
"step": 11000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 25.0,
|
| 455 |
+
"eval_accuracy": 0.9696686887540831,
|
| 456 |
+
"eval_f1": 0.7969924812030075,
|
| 457 |
+
"eval_loss": 0.2650408446788788,
|
| 458 |
+
"eval_precision": 0.7681159420289855,
|
| 459 |
+
"eval_recall": 0.828125,
|
| 460 |
+
"eval_runtime": 3.3206,
|
| 461 |
+
"eval_samples_per_second": 273.743,
|
| 462 |
+
"eval_steps_per_second": 34.331,
|
| 463 |
+
"step": 11375
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 25.0,
|
| 467 |
+
"step": 11375,
|
| 468 |
+
"total_flos": 2472027466589526.0,
|
| 469 |
+
"train_loss": 0.03293397539264553,
|
| 470 |
+
"train_runtime": 1492.2693,
|
| 471 |
+
"train_samples_per_second": 60.897,
|
| 472 |
+
"train_steps_per_second": 7.623
|
| 473 |
+
}
|
| 474 |
+
],
|
| 475 |
+
"logging_steps": 500,
|
| 476 |
+
"max_steps": 11375,
|
| 477 |
+
"num_input_tokens_seen": 0,
|
| 478 |
+
"num_train_epochs": 25,
|
| 479 |
+
"save_steps": 500,
|
| 480 |
+
"stateful_callbacks": {
|
| 481 |
+
"TrainerControl": {
|
| 482 |
+
"args": {
|
| 483 |
+
"should_epoch_stop": false,
|
| 484 |
+
"should_evaluate": false,
|
| 485 |
+
"should_log": false,
|
| 486 |
+
"should_save": true,
|
| 487 |
+
"should_training_stop": true
|
| 488 |
+
},
|
| 489 |
+
"attributes": {}
|
| 490 |
+
}
|
| 491 |
+
},
|
| 492 |
+
"total_flos": 2472027466589526.0,
|
| 493 |
+
"train_batch_size": 8,
|
| 494 |
+
"trial_name": null,
|
| 495 |
+
"trial_params": null
|
| 496 |
+
}
|