anowakowski
commited on
Commit
•
6833ebf
1
Parent(s):
4935ed2
upload slavlemma-small
Browse files- .gitattributes +1 -0
- README.md +61 -0
- config.json +31 -0
- generation_config.json +7 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +5 -0
- spiece.model +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +12 -0
.gitattributes
CHANGED
@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
32 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
33 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
34 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
32 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
33 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
34 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
35 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
@@ -1,3 +1,64 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- pl
|
4 |
+
- cs
|
5 |
+
- ru
|
6 |
+
tags:
|
7 |
+
- mT5
|
8 |
+
- lemmatization
|
9 |
license: apache-2.0
|
10 |
---
|
11 |
+
|
12 |
+
|
13 |
+
# SlavLemma Small
|
14 |
+
|
15 |
+
SlavLemma models are intended for lemmatization of named entities and multi-word expressions in Polish, Czech and Russian languages.
|
16 |
+
|
17 |
+
They were fine-tuned from the google/mT5 models, e.g.: [google/mt5-small](https://huggingface.co/google/mt5-small).
|
18 |
+
|
19 |
+
## Usage
|
20 |
+
|
21 |
+
When using the model, prepend one of the language tokens (`>>pl<<`, `>>cs<<`, `>>ru<<`) to the input, based on the language of the phrase you want to lemmatize.
|
22 |
+
|
23 |
+
Sample usage:
|
24 |
+
|
25 |
+
```
|
26 |
+
from transformers import pipeline
|
27 |
+
|
28 |
+
pipe = pipeline(task="text2text-generation", model="amu-cai/slavlemma-small", tokenizer="amu-cai/slavlemma-small")
|
29 |
+
hyp = [res['generated_text'] for res in pipe([">>pl<< federalnego urzędu statystycznego"], clean_up_tokenization_spaces=True, num_beams=5)][0]
|
30 |
+
```
|
31 |
+
|
32 |
+
|
33 |
+
## Evaluation results
|
34 |
+
|
35 |
+
Lemmatization Exact Match was computed on the SlavNER 2021 test sets (COVID-19 and USA 2020 Elections).
|
36 |
+
|
37 |
+
|
38 |
+
COVID-19:
|
39 |
+
| Model | pl | cs | ru |
|
40 |
+
| :------ | ------: | ------: | ------: |
|
41 |
+
| [slavlemma-large](https://huggingface.co/amu-cai/slavlemma-large) | 93.76 | 89.80 | 77.30
|
42 |
+
| [slavlemma-base](https://huggingface.co/amu-cai/slavlemma-base) | 91.00 |86.29| 76.10
|
43 |
+
| [slavlemma-small](https://huggingface.co/amu-cai/slavlemma-small)| 86.80 |80.98| 73.83
|
44 |
+
|
45 |
+
USA 2020 Elections:
|
46 |
+
| Model | pl | cs | ru |
|
47 |
+
| :------ | ------: | ------: | ------: |
|
48 |
+
| [slavlemma-large](https://huggingface.co/amu-cai/slavlemma-large) | 89.12 | 87.27| 82.50
|
49 |
+
| [slavlemma-base](https://huggingface.co/amu-cai/slavlemma-base) | 84.19 |81.97| 80.27
|
50 |
+
| [slavlemma-small](https://huggingface.co/amu-cai/slavlemma-small)| 78.85 |75.86| 76.18
|
51 |
+
|
52 |
+
|
53 |
+
## Citation
|
54 |
+
|
55 |
+
If you use the model, please cite the following paper:
|
56 |
+
|
57 |
+
TBD
|
58 |
+
|
59 |
+
### Framework versions
|
60 |
+
|
61 |
+
- Transformers 4.26.0
|
62 |
+
- Pytorch 1.13.1.post200
|
63 |
+
- Datasets 2.9.0
|
64 |
+
- Tokenizers 0.13.2
|
config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/mt5-small",
|
3 |
+
"architectures": [
|
4 |
+
"MT5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"d_ff": 1024,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 512,
|
9 |
+
"decoder_start_token_id": 0,
|
10 |
+
"dense_act_fn": "gelu_new",
|
11 |
+
"dropout_rate": 0.1,
|
12 |
+
"eos_token_id": 1,
|
13 |
+
"feed_forward_proj": "gated-gelu",
|
14 |
+
"initializer_factor": 1.0,
|
15 |
+
"is_encoder_decoder": true,
|
16 |
+
"is_gated_act": true,
|
17 |
+
"layer_norm_epsilon": 1e-06,
|
18 |
+
"model_type": "mt5",
|
19 |
+
"num_decoder_layers": 8,
|
20 |
+
"num_heads": 6,
|
21 |
+
"num_layers": 8,
|
22 |
+
"pad_token_id": 0,
|
23 |
+
"relative_attention_max_distance": 128,
|
24 |
+
"relative_attention_num_buckets": 32,
|
25 |
+
"tie_word_embeddings": false,
|
26 |
+
"tokenizer_class": "T5Tokenizer",
|
27 |
+
"torch_dtype": "float32",
|
28 |
+
"transformers_version": "4.26.0",
|
29 |
+
"use_cache": true,
|
30 |
+
"vocab_size": 250112
|
31 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"decoder_start_token_id": 0,
|
4 |
+
"eos_token_id": 1,
|
5 |
+
"pad_token_id": 0,
|
6 |
+
"transformers_version": "4.26.0"
|
7 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:16c72aba61d4a15a445c67e4aef92be8403121824490873c441bd4d50c2db163
|
3 |
+
size 1200772485
|
special_tokens_map.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eos_token": "</s>",
|
3 |
+
"pad_token": "<pad>",
|
4 |
+
"unk_token": "<unk>"
|
5 |
+
}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
|
3 |
+
size 4309802
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:93c3578052e1605d8332eb961bc08d72e246071974e4cc54aa6991826b802aa5
|
3 |
+
size 16330369
|
tokenizer_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": null,
|
3 |
+
"eos_token": "</s>",
|
4 |
+
"extra_ids": 0,
|
5 |
+
"model_max_length": 1000000000000000019884624838656,
|
6 |
+
"name_or_path": "google/mt5-small",
|
7 |
+
"pad_token": "<pad>",
|
8 |
+
"sp_model_kwargs": {},
|
9 |
+
"special_tokens_map_file": "/home/wmi/.cache/huggingface/hub/models--google--mt5-small/snapshots/38f23af8ec210eb6c376d40e9c56bd25a80f195d/special_tokens_map.json",
|
10 |
+
"tokenizer_class": "T5Tokenizer",
|
11 |
+
"unk_token": "<unk>"
|
12 |
+
}
|