Update README.md
Browse files
README.md
CHANGED
@@ -9,6 +9,11 @@ metrics:
|
|
9 |
model-index:
|
10 |
- name: pos-polish-gpt2-large
|
11 |
results: []
|
|
|
|
|
|
|
|
|
|
|
12 |
---
|
13 |
|
14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -16,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
# pos-polish-gpt2-large
|
18 |
|
19 |
-
This model was trained from
|
20 |
It achieves the following results on the evaluation set:
|
21 |
- Loss: 0.2290
|
22 |
- Precision: 0.8910
|
@@ -26,18 +31,29 @@ It achieves the following results on the evaluation set:
|
|
26 |
|
27 |
## Model description
|
28 |
|
29 |
-
|
30 |
|
31 |
## Intended uses & limitations
|
32 |
|
33 |
-
|
|
|
34 |
|
35 |
## Training and evaluation data
|
36 |
|
37 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
|
39 |
## Training procedure
|
40 |
|
|
|
|
|
|
|
|
|
41 |
### Training hyperparameters
|
42 |
|
43 |
The following hyperparameters were used during training:
|
@@ -56,6 +72,7 @@ The following hyperparameters were used during training:
|
|
56 |
|
57 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
58 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
|
|
59 |
| 0.1952 | 1.0 | 2444 | 0.1942 | 0.8865 | 0.9304 | 0.9079 | 0.9426 |
|
60 |
| 0.1287 | 2.0 | 4889 | 0.1984 | 0.8903 | 0.9322 | 0.9108 | 0.9449 |
|
61 |
| 0.0832 | 3.0 | 7332 | 0.2290 | 0.8910 | 0.9328 | 0.9114 | 0.9450 |
|
@@ -66,4 +83,4 @@ The following hyperparameters were used during training:
|
|
66 |
- Transformers 4.36.2
|
67 |
- Pytorch 2.1.2+cu121
|
68 |
- Datasets 2.16.1
|
69 |
-
- Tokenizers 0.15.0
|
|
|
9 |
model-index:
|
10 |
- name: pos-polish-gpt2-large
|
11 |
results: []
|
12 |
+
license: mit
|
13 |
+
datasets:
|
14 |
+
- clarin-pl/nkjp-pos
|
15 |
+
language:
|
16 |
+
- pl
|
17 |
---
|
18 |
|
19 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
21 |
|
22 |
# pos-polish-gpt2-large
|
23 |
|
24 |
+
This model was trained from [polish-gpt2-large](https://huggingface.co/sdadas/polish-gpt2-large) on [clarin-pl/nkjp-pos](https://huggingface.co/datasets/clarin-pl/nkjp-pos) dataset.
|
25 |
It achieves the following results on the evaluation set:
|
26 |
- Loss: 0.2290
|
27 |
- Precision: 0.8910
|
|
|
31 |
|
32 |
## Model description
|
33 |
|
34 |
+
Trained from [polish-gpt2-large](https://huggingface.co/sdadas/polish-gpt2-large)
|
35 |
|
36 |
## Intended uses & limitations
|
37 |
|
38 |
+
Part-of-speech tagging for Polish language.
|
39 |
+
Tags description at the bottom of http://nkjp.pl/poliqarp/help/plse2.html
|
40 |
|
41 |
## Training and evaluation data
|
42 |
|
43 |
+
Dataset: [clarin-pl/nkjp-pos](https://huggingface.co/datasets/clarin-pl/nkjp-pos)
|
44 |
+
|
45 |
+
Datacollator:
|
46 |
+
```py
|
47 |
+
from transformers import DataCollatorForTokenClassification
|
48 |
+
data_collator = DataCollatorForTokenClassification(tokenizer=tokenizer)
|
49 |
+
```
|
50 |
|
51 |
## Training procedure
|
52 |
|
53 |
+
GPU: RTX 3090
|
54 |
+
|
55 |
+
Training time: 01:15:31
|
56 |
+
|
57 |
### Training hyperparameters
|
58 |
|
59 |
The following hyperparameters were used during training:
|
|
|
72 |
|
73 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
74 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
75 |
+
| | 0.0 | 0 | 3.8487 | 3.8487 | 3.8487 | 3.8487 | 3.8487 |
|
76 |
| 0.1952 | 1.0 | 2444 | 0.1942 | 0.8865 | 0.9304 | 0.9079 | 0.9426 |
|
77 |
| 0.1287 | 2.0 | 4889 | 0.1984 | 0.8903 | 0.9322 | 0.9108 | 0.9449 |
|
78 |
| 0.0832 | 3.0 | 7332 | 0.2290 | 0.8910 | 0.9328 | 0.9114 | 0.9450 |
|
|
|
83 |
- Transformers 4.36.2
|
84 |
- Pytorch 2.1.2+cu121
|
85 |
- Datasets 2.16.1
|
86 |
+
- Tokenizers 0.15.0
|