wennycooper
commited on
Commit
•
fffdaf1
1
Parent(s):
1e60211
End of training
Browse files- README.md +34 -2
- all_results.json +17 -0
- eval_results.json +12 -0
- train_results.json +8 -0
- trainer_state.json +88 -0
README.md
CHANGED
@@ -3,9 +3,35 @@ license: apache-2.0
|
|
3 |
base_model: bert-base-uncased
|
4 |
tags:
|
5 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
model-index:
|
7 |
- name: token-classification-bert-base-uncased
|
8 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -13,7 +39,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# token-classification-bert-base-uncased
|
15 |
|
16 |
-
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
|
18 |
## Model description
|
19 |
|
|
|
3 |
base_model: bert-base-uncased
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- conll2003
|
8 |
+
metrics:
|
9 |
+
- precision
|
10 |
+
- recall
|
11 |
+
- f1
|
12 |
+
- accuracy
|
13 |
model-index:
|
14 |
- name: token-classification-bert-base-uncased
|
15 |
+
results:
|
16 |
+
- task:
|
17 |
+
name: Token Classification
|
18 |
+
type: token-classification
|
19 |
+
dataset:
|
20 |
+
name: conll2003
|
21 |
+
type: conll2003
|
22 |
+
metrics:
|
23 |
+
- name: Precision
|
24 |
+
type: precision
|
25 |
+
value: 0.9465865464863963
|
26 |
+
- name: Recall
|
27 |
+
type: recall
|
28 |
+
value: 0.9543924604510265
|
29 |
+
- name: F1
|
30 |
+
type: f1
|
31 |
+
value: 0.9504734769127628
|
32 |
+
- name: Accuracy
|
33 |
+
type: accuracy
|
34 |
+
value: 0.9898757836532845
|
35 |
---
|
36 |
|
37 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
39 |
|
40 |
# token-classification-bert-base-uncased
|
41 |
|
42 |
+
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the conll2003 dataset.
|
43 |
+
It achieves the following results on the evaluation set:
|
44 |
+
- Loss: 0.0480
|
45 |
+
- Precision: 0.9466
|
46 |
+
- Recall: 0.9544
|
47 |
+
- F1: 0.9505
|
48 |
+
- Accuracy: 0.9899
|
49 |
|
50 |
## Model description
|
51 |
|
all_results.json
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_accuracy": 0.9898757836532845,
|
4 |
+
"eval_f1": 0.9504734769127628,
|
5 |
+
"eval_loss": 0.04797038435935974,
|
6 |
+
"eval_precision": 0.9465865464863963,
|
7 |
+
"eval_recall": 0.9543924604510265,
|
8 |
+
"eval_runtime": 10.3106,
|
9 |
+
"eval_samples": 3250,
|
10 |
+
"eval_samples_per_second": 315.208,
|
11 |
+
"eval_steps_per_second": 39.474,
|
12 |
+
"train_loss": 0.043311251413668296,
|
13 |
+
"train_runtime": 605.4597,
|
14 |
+
"train_samples": 14041,
|
15 |
+
"train_samples_per_second": 69.572,
|
16 |
+
"train_steps_per_second": 8.701
|
17 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_accuracy": 0.9898757836532845,
|
4 |
+
"eval_f1": 0.9504734769127628,
|
5 |
+
"eval_loss": 0.04797038435935974,
|
6 |
+
"eval_precision": 0.9465865464863963,
|
7 |
+
"eval_recall": 0.9543924604510265,
|
8 |
+
"eval_runtime": 10.3106,
|
9 |
+
"eval_samples": 3250,
|
10 |
+
"eval_samples_per_second": 315.208,
|
11 |
+
"eval_steps_per_second": 39.474
|
12 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"train_loss": 0.043311251413668296,
|
4 |
+
"train_runtime": 605.4597,
|
5 |
+
"train_samples": 14041,
|
6 |
+
"train_samples_per_second": 69.572,
|
7 |
+
"train_steps_per_second": 8.701
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 5268,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.28,
|
13 |
+
"learning_rate": 4.525436598329537e-05,
|
14 |
+
"loss": 0.154,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.57,
|
19 |
+
"learning_rate": 4.050873196659074e-05,
|
20 |
+
"loss": 0.0783,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.85,
|
25 |
+
"learning_rate": 3.5763097949886106e-05,
|
26 |
+
"loss": 0.0625,
|
27 |
+
"step": 1500
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 1.14,
|
31 |
+
"learning_rate": 3.1017463933181475e-05,
|
32 |
+
"loss": 0.0427,
|
33 |
+
"step": 2000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 1.42,
|
37 |
+
"learning_rate": 2.6271829916476843e-05,
|
38 |
+
"loss": 0.0245,
|
39 |
+
"step": 2500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 1.71,
|
43 |
+
"learning_rate": 2.152619589977221e-05,
|
44 |
+
"loss": 0.0296,
|
45 |
+
"step": 3000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 1.99,
|
49 |
+
"learning_rate": 1.678056188306758e-05,
|
50 |
+
"loss": 0.0264,
|
51 |
+
"step": 3500
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 2.28,
|
55 |
+
"learning_rate": 1.2034927866362947e-05,
|
56 |
+
"loss": 0.0115,
|
57 |
+
"step": 4000
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 2.56,
|
61 |
+
"learning_rate": 7.289293849658315e-06,
|
62 |
+
"loss": 0.0108,
|
63 |
+
"step": 4500
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 2.85,
|
67 |
+
"learning_rate": 2.5436598329536827e-06,
|
68 |
+
"loss": 0.0103,
|
69 |
+
"step": 5000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 3.0,
|
73 |
+
"step": 5268,
|
74 |
+
"total_flos": 895156246626660.0,
|
75 |
+
"train_loss": 0.043311251413668296,
|
76 |
+
"train_runtime": 605.4597,
|
77 |
+
"train_samples_per_second": 69.572,
|
78 |
+
"train_steps_per_second": 8.701
|
79 |
+
}
|
80 |
+
],
|
81 |
+
"logging_steps": 500,
|
82 |
+
"max_steps": 5268,
|
83 |
+
"num_train_epochs": 3,
|
84 |
+
"save_steps": 500,
|
85 |
+
"total_flos": 895156246626660.0,
|
86 |
+
"trial_name": null,
|
87 |
+
"trial_params": null
|
88 |
+
}
|