End of training
Browse files- README.md +34 -1
- all_results.json +15 -15
- eval_results.json +9 -9
- train_results.json +6 -6
- trainer_state.json +39 -9
README.md
CHANGED
@@ -5,9 +5,36 @@ tags:
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- conll2003
|
|
|
|
|
|
|
|
|
|
|
8 |
model-index:
|
9 |
- name: test-ner-run
|
10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
|
13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -16,6 +43,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
# test-ner-run
|
17 |
|
18 |
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the conll2003 dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- conll2003
|
8 |
+
metrics:
|
9 |
+
- precision
|
10 |
+
- recall
|
11 |
+
- f1
|
12 |
+
- accuracy
|
13 |
model-index:
|
14 |
- name: test-ner-run
|
15 |
+
results:
|
16 |
+
- task:
|
17 |
+
name: Token Classification
|
18 |
+
type: token-classification
|
19 |
+
dataset:
|
20 |
+
name: conll2003
|
21 |
+
type: conll2003
|
22 |
+
config: conll2003
|
23 |
+
split: validation
|
24 |
+
args: conll2003
|
25 |
+
metrics:
|
26 |
+
- name: Precision
|
27 |
+
type: precision
|
28 |
+
value: 0.946991165194199
|
29 |
+
- name: Recall
|
30 |
+
type: recall
|
31 |
+
value: 0.9560753954897341
|
32 |
+
- name: F1
|
33 |
+
type: f1
|
34 |
+
value: 0.9515115986935768
|
35 |
+
- name: Accuracy
|
36 |
+
type: accuracy
|
37 |
+
value: 0.9903625248237997
|
38 |
---
|
39 |
|
40 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
43 |
# test-ner-run
|
44 |
|
45 |
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the conll2003 dataset.
|
46 |
+
It achieves the following results on the evaluation set:
|
47 |
+
- Loss: 0.0414
|
48 |
+
- Precision: 0.9470
|
49 |
+
- Recall: 0.9561
|
50 |
+
- F1: 0.9515
|
51 |
+
- Accuracy: 0.9904
|
52 |
|
53 |
## Model description
|
54 |
|
all_results.json
CHANGED
@@ -1,18 +1,18 @@
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_f1": 0.
|
5 |
-
"eval_loss": 0.
|
6 |
-
"eval_precision": 0.
|
7 |
-
"eval_recall": 0.
|
8 |
-
"eval_runtime":
|
9 |
-
"eval_samples":
|
10 |
-
"eval_samples_per_second":
|
11 |
-
"eval_steps_per_second":
|
12 |
-
"total_flos":
|
13 |
-
"train_loss": 0.
|
14 |
-
"train_runtime":
|
15 |
-
"train_samples":
|
16 |
-
"train_samples_per_second":
|
17 |
-
"train_steps_per_second":
|
18 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
+
"eval_accuracy": 0.9903625248237997,
|
4 |
+
"eval_f1": 0.9515115986935768,
|
5 |
+
"eval_loss": 0.041357845067977905,
|
6 |
+
"eval_precision": 0.946991165194199,
|
7 |
+
"eval_recall": 0.9560753954897341,
|
8 |
+
"eval_runtime": 11.0076,
|
9 |
+
"eval_samples": 3250,
|
10 |
+
"eval_samples_per_second": 295.251,
|
11 |
+
"eval_steps_per_second": 18.533,
|
12 |
+
"total_flos": 1022109083576478.0,
|
13 |
+
"train_loss": 0.04506039325269498,
|
14 |
+
"train_runtime": 475.8923,
|
15 |
+
"train_samples": 14041,
|
16 |
+
"train_samples_per_second": 88.514,
|
17 |
+
"train_steps_per_second": 5.535
|
18 |
}
|
eval_results.json
CHANGED
@@ -1,12 +1,12 @@
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_f1": 0.
|
5 |
-
"eval_loss": 0.
|
6 |
-
"eval_precision": 0.
|
7 |
-
"eval_recall": 0.
|
8 |
-
"eval_runtime":
|
9 |
-
"eval_samples":
|
10 |
-
"eval_samples_per_second":
|
11 |
-
"eval_steps_per_second":
|
12 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
+
"eval_accuracy": 0.9903625248237997,
|
4 |
+
"eval_f1": 0.9515115986935768,
|
5 |
+
"eval_loss": 0.041357845067977905,
|
6 |
+
"eval_precision": 0.946991165194199,
|
7 |
+
"eval_recall": 0.9560753954897341,
|
8 |
+
"eval_runtime": 11.0076,
|
9 |
+
"eval_samples": 3250,
|
10 |
+
"eval_samples_per_second": 295.251,
|
11 |
+
"eval_steps_per_second": 18.533
|
12 |
}
|
train_results.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
-
"total_flos":
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples":
|
7 |
-
"train_samples_per_second":
|
8 |
-
"train_steps_per_second":
|
9 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
+
"total_flos": 1022109083576478.0,
|
4 |
+
"train_loss": 0.04506039325269498,
|
5 |
+
"train_runtime": 475.8923,
|
6 |
+
"train_samples": 14041,
|
7 |
+
"train_samples_per_second": 88.514,
|
8 |
+
"train_steps_per_second": 5.535
|
9 |
}
|
trainer_state.json
CHANGED
@@ -3,26 +3,56 @@
|
|
3 |
"best_model_checkpoint": null,
|
4 |
"epoch": 3.0,
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
{
|
12 |
"epoch": 3.0,
|
13 |
-
"step":
|
14 |
-
"total_flos":
|
15 |
-
"train_loss": 0.
|
16 |
-
"train_runtime":
|
17 |
-
"train_samples_per_second":
|
18 |
-
"train_steps_per_second":
|
19 |
}
|
20 |
],
|
21 |
"logging_steps": 500,
|
22 |
-
"max_steps":
|
23 |
"num_train_epochs": 3,
|
24 |
"save_steps": 500,
|
25 |
-
"total_flos":
|
26 |
"trial_name": null,
|
27 |
"trial_params": null
|
28 |
}
|
|
|
3 |
"best_model_checkpoint": null,
|
4 |
"epoch": 3.0,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 2634,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.57,
|
13 |
+
"learning_rate": 4.050873196659074e-05,
|
14 |
+
"loss": 0.1325,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 1.14,
|
19 |
+
"learning_rate": 3.1017463933181475e-05,
|
20 |
+
"loss": 0.047,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 1.71,
|
25 |
+
"learning_rate": 2.152619589977221e-05,
|
26 |
+
"loss": 0.0251,
|
27 |
+
"step": 1500
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 2.28,
|
31 |
+
"learning_rate": 1.2034927866362947e-05,
|
32 |
+
"loss": 0.0193,
|
33 |
+
"step": 2000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 2.85,
|
37 |
+
"learning_rate": 2.5436598329536827e-06,
|
38 |
+
"loss": 0.0106,
|
39 |
+
"step": 2500
|
40 |
+
},
|
41 |
{
|
42 |
"epoch": 3.0,
|
43 |
+
"step": 2634,
|
44 |
+
"total_flos": 1022109083576478.0,
|
45 |
+
"train_loss": 0.04506039325269498,
|
46 |
+
"train_runtime": 475.8923,
|
47 |
+
"train_samples_per_second": 88.514,
|
48 |
+
"train_steps_per_second": 5.535
|
49 |
}
|
50 |
],
|
51 |
"logging_steps": 500,
|
52 |
+
"max_steps": 2634,
|
53 |
"num_train_epochs": 3,
|
54 |
"save_steps": 500,
|
55 |
+
"total_flos": 1022109083576478.0,
|
56 |
"trial_name": null,
|
57 |
"trial_params": null
|
58 |
}
|