anowakowski commited on
Commit
3455472
1 Parent(s): e4f585c

Upload polemma-small

Browse files
README.md CHANGED
@@ -1,3 +1,50 @@
1
  ---
 
 
 
 
2
  license: apache-2.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language: pl
3
+ tags:
4
+ - T5
5
+ - lemmatization
6
  license: apache-2.0
7
  ---
8
+
9
+
10
+ # PoLemma Small
11
+
12
+ PoLemma models are intended for lemmatization of named entities and multi-word expressions in the Polish language.
13
+
14
+ They were fine-tuned from the allegro/plT5 models, e.g.: [allegro/plt5-small](https://huggingface.co/allegro/plt5-small).
15
+
16
+ ## Usage
17
+
18
+ Sample usage:
19
+
20
+ ```
21
+ from transformers import pipeline
22
+
23
+ pipe = pipeline(task="text2text-generation", model="amu-cai/polemma-base", tokenizer="amu-cai/polemma-base")
24
+ hyp = [res['generated_text'] for res in pipe(["federalnego urzędu statystycznego"], clean_up_tokenization_spaces=True, num_beams=5)][0]
25
+ ```
26
+
27
+
28
+ ## Evaluation results
29
+
30
+ Lemmatization Exact Match was computed on the SlavNER 2021 test set.
31
+
32
+ | Model | Exact Match ||
33
+ | :------ | ------: | ------: |
34
+ | [polemma-large]() | 92.61 |
35
+ | [polemma-base]() | 91.34 |
36
+ | [polemma-small]()| 88.46 |
37
+
38
+
39
+ ## Citation
40
+
41
+ If you use the model, please cite the following paper:
42
+
43
+ TBD
44
+
45
+ ### Framework versions
46
+
47
+ - Transformers 4.26.0
48
+ - Pytorch 1.13.1.post200
49
+ - Datasets 2.9.0
50
+ - Tokenizers 0.13.2
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allegro/plt5-small",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
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
+ "gradient_checkpointing": false,
15
+ "initializer_factor": 1,
16
+ "is_encoder_decoder": true,
17
+ "is_gated_act": true,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "model_type": "t5",
20
+ "num_decoder_layers": 8,
21
+ "num_heads": 6,
22
+ "num_layers": 8,
23
+ "pad_token_id": 0,
24
+ "relative_attention_max_distance": 128,
25
+ "relative_attention_num_buckets": 32,
26
+ "tie_word_embeddings": false,
27
+ "tokenizer_class": "T5Tokenizer",
28
+ "torch_dtype": "float32",
29
+ "transformers_version": "4.26.0",
30
+ "use_cache": true,
31
+ "vocab_size": 50048
32
+ }
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:e2e407de5282ce8f6119a7aa433ef20e35a9fd2f8b76b23614263a5419dfdc56
3
+ size 381310469
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:a200b4b9d7106b32ea2340e5cd27033a5c03ae5ece9241ab6e1f685b938800b9
3
+ size 1124950
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
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": "allegro/plt5-small",
7
+ "pad_token": "<pad>",
8
+ "sp_model_kwargs": {},
9
+ "special_tokens_map_file": "/home/wmi/.cache/huggingface/hub/models--allegro--plt5-small/snapshots/6ab71258c53f77f075fdda380992c0e691703f6b/special_tokens_map.json",
10
+ "tokenizer_class": "T5Tokenizer",
11
+ "unk_token": "<unk>"
12
+ }