hiendang7613
commited on
Commit
•
44fab51
1
Parent(s):
c22c04d
commit_1
Browse files
README.md
CHANGED
@@ -3,11 +3,6 @@ license: mit
|
|
3 |
base_model: xlm-roberta-base
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
-
metrics:
|
7 |
-
- precision
|
8 |
-
- recall
|
9 |
-
- f1
|
10 |
-
- accuracy
|
11 |
model-index:
|
12 |
- name: xlmr-lstm-crf-resume-ner
|
13 |
results: []
|
@@ -19,12 +14,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
19 |
# xlmr-lstm-crf-resume-ner
|
20 |
|
21 |
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset.
|
22 |
-
It achieves the following results on the evaluation set:
|
23 |
-
- Loss: 26.2892
|
24 |
-
- Precision: 0.8714
|
25 |
-
- Recall: 0.8971
|
26 |
-
- F1: 0.8841
|
27 |
-
- Accuracy: 0.9398
|
28 |
|
29 |
## Model description
|
30 |
|
@@ -50,35 +39,7 @@ The following hyperparameters were used during training:
|
|
50 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
51 |
- lr_scheduler_type: linear
|
52 |
- num_epochs: 100
|
53 |
-
|
54 |
-
### Training results
|
55 |
-
|
56 |
-
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
57 |
-
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
58 |
-
| 82.4955 | 0.11 | 100 | 57.1952 | 0.0 | 0.0 | 0.0 | 0.7590 |
|
59 |
-
| 58.2029 | 0.22 | 200 | 47.6007 | 0.9412 | 0.4706 | 0.6275 | 0.8253 |
|
60 |
-
| 44.4656 | 0.33 | 300 | 41.2301 | 0.7674 | 0.4853 | 0.5946 | 0.8283 |
|
61 |
-
| 38.3105 | 0.44 | 400 | 34.7889 | 0.6140 | 0.5147 | 0.5600 | 0.8464 |
|
62 |
-
| 33.1407 | 0.55 | 500 | 32.4219 | 0.5 | 0.5294 | 0.5143 | 0.8735 |
|
63 |
-
| 29.2323 | 0.66 | 600 | 29.6968 | 0.6780 | 0.5882 | 0.6299 | 0.8825 |
|
64 |
-
| 27.7611 | 0.77 | 700 | 29.1259 | 0.65 | 0.5735 | 0.6094 | 0.8780 |
|
65 |
-
| 25.626 | 0.88 | 800 | 27.6424 | 0.6027 | 0.6471 | 0.6241 | 0.8901 |
|
66 |
-
| 22.9444 | 0.99 | 900 | 26.4424 | 0.7424 | 0.7206 | 0.7313 | 0.9066 |
|
67 |
-
| 23.0567 | 1.1 | 1000 | 26.6253 | 0.8033 | 0.7206 | 0.7597 | 0.9021 |
|
68 |
-
| 18.3865 | 1.21 | 1100 | 26.5443 | 0.8060 | 0.7941 | 0.8 | 0.9307 |
|
69 |
-
| 20.5945 | 1.32 | 1200 | 25.1565 | 0.9 | 0.7941 | 0.8438 | 0.9247 |
|
70 |
-
| 20.0981 | 1.43 | 1300 | 25.7204 | 0.7917 | 0.8382 | 0.8143 | 0.9413 |
|
71 |
-
| 17.4745 | 1.54 | 1400 | 25.5618 | 0.7568 | 0.8235 | 0.7887 | 0.9473 |
|
72 |
-
| 18.2305 | 1.64 | 1500 | 24.7287 | 0.7397 | 0.7941 | 0.7660 | 0.9398 |
|
73 |
-
| 17.8145 | 1.75 | 1600 | 25.8965 | 0.8056 | 0.8529 | 0.8286 | 0.9488 |
|
74 |
-
| 16.4689 | 1.86 | 1700 | 25.3026 | 0.8824 | 0.8824 | 0.8824 | 0.9729 |
|
75 |
-
| 15.3847 | 1.97 | 1800 | 24.7269 | 0.8507 | 0.8382 | 0.8444 | 0.9578 |
|
76 |
-
| 13.6318 | 2.08 | 1900 | 24.8136 | 0.8592 | 0.8971 | 0.8777 | 0.9473 |
|
77 |
-
| 13.4369 | 2.19 | 2000 | 26.2700 | 0.8841 | 0.8971 | 0.8905 | 0.9608 |
|
78 |
-
| 13.7298 | 2.3 | 2100 | 24.8570 | 0.8472 | 0.8971 | 0.8714 | 0.9518 |
|
79 |
-
| 13.0787 | 2.41 | 2200 | 25.9105 | 0.8696 | 0.8824 | 0.8759 | 0.9428 |
|
80 |
-
| 13.1534 | 2.52 | 2300 | 26.2892 | 0.8714 | 0.8971 | 0.8841 | 0.9398 |
|
81 |
-
|
82 |
|
83 |
### Framework versions
|
84 |
|
|
|
3 |
base_model: xlm-roberta-base
|
4 |
tags:
|
5 |
- generated_from_trainer
|
|
|
|
|
|
|
|
|
|
|
6 |
model-index:
|
7 |
- name: xlmr-lstm-crf-resume-ner
|
8 |
results: []
|
|
|
14 |
# xlmr-lstm-crf-resume-ner
|
15 |
|
16 |
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
|
18 |
## Model description
|
19 |
|
|
|
39 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
40 |
- lr_scheduler_type: linear
|
41 |
- num_epochs: 100
|
42 |
+
- mixed_precision_training: Native AMP
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
|
44 |
### Framework versions
|
45 |
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1114678192
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fa8d3b4efa666f5e516dd1088d74919d9c38f04af1dd3be815596f63871bf9c6
|
3 |
size 1114678192
|
runs/Nov19_17-25-32_d1be6ed5b3f3/events.out.tfevents.1700422054.d1be6ed5b3f3.6890.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8bbfbeed081d4ba84e62476ec57ad578427ceab01958ad8c8bb14346f43ebef5
|
3 |
+
size 7774
|
runs/Nov19_19-34-17_d1be6ed5b3f3/events.out.tfevents.1700422460.d1be6ed5b3f3.6890.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3ae5fbcaf9b7904ac1d2ccad8cb21bbaec560fd62036c11066a2aacbaa30ca0a
|
3 |
+
size 4184
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b9160c322ce8df28b42cb603469e5b22e84067c9f8fd70b9f2b3fa7c65fb5860
|
3 |
+
size 4600
|