ZJaume commited on
Commit
99975ad
1 Parent(s): 4202110

Multilingual models Bicleaner AI v3.0.0

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ - ar
5
+ - az
6
+ - bg
7
+ - bs
8
+ - ca
9
+ - cnr
10
+ - cs
11
+ - cy
12
+ - da
13
+ - de
14
+ - el
15
+ - es
16
+ - et
17
+ - eu
18
+ - fi
19
+ - fr
20
+ - ga
21
+ - gd
22
+ - gl
23
+ - he
24
+ - hi
25
+ - hr
26
+ - hu
27
+ - is
28
+ - it
29
+ - ja
30
+ - lt
31
+ - lv
32
+ - mk
33
+ - mt
34
+ - nb
35
+ - nl
36
+ - nn
37
+ - pl
38
+ - pt
39
+ - ro
40
+ - ru
41
+ - sk
42
+ - sl
43
+ - sq
44
+ - sr
45
+ - sv
46
+ - sw
47
+ - tr
48
+ - uk
49
+ - vi
50
+ - zh
51
+ - multilingual
52
+ license: cc-by-sa-4.0
53
+ tags:
54
+ - bicleaner-ai
55
+ tasks:
56
+ - text-classification
57
+ ---
58
+
59
+ # Bicleaner AI multilingual full model for en-xx
60
+ Bicleaner AI is a tool that aims at detecting noisy sentence pairs in a parallel corpus. It
61
+ indicates the likelihood of a pair of sentences being mutual translations (with a value near to 1) or not (with a value near to 0).
62
+ Sentence pairs considered very noisy are scored with 0.
63
+
64
+ Find out at our repository for further instructions on how to use it: https://github.com/bitextor/bicleaner-ai
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bitextor/bicleaner-ai-full-en-xx",
3
+ "architectures": [
4
+ "XLMRBicleanerAI"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "head_activation": "relu",
11
+ "head_dropout": 0.1,
12
+ "head_hidden_size": 2048,
13
+ "hidden_act": "gelu",
14
+ "hidden_dropout_prob": 0.1,
15
+ "hidden_size": 768,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "layer_norm_eps": 1e-05,
19
+ "max_position_embeddings": 514,
20
+ "model_type": "xlm-roberta",
21
+ "num_attention_heads": 12,
22
+ "num_hidden_layers": 12,
23
+ "output_past": true,
24
+ "pad_token_id": 1,
25
+ "position_embedding_type": "absolute",
26
+ "transformers_version": "4.30.2",
27
+ "type_vocab_size": 1,
28
+ "use_cache": true,
29
+ "vocab_size": 250002
30
+ }
metadata.yaml ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bicleaner_ai_version: 3.0.0
2
+ precision_score: 0.881
3
+ recall_score: 0.902
4
+ f1_score: 0.891
5
+ matthews_corr_coef: 0.879
6
+ source_lang: en
7
+ target_lang: xx
8
+ classifier_type: xlmr
9
+ classifier_settings:
10
+ activation: relu
11
+ base_model: jplu/tf-xlm-roberta-base
12
+ batch_size: 128
13
+ calibration_params: [1.80103183, -0.55578989]
14
+ epochs: 10
15
+ loss: binary_crossentropy
16
+ maxlen: 150
17
+ model_name: bitextor/bicleaner-ai-full-en-xx
18
+ n_classes: 2
19
+ n_hidden: 2048
20
+ patience: 3
21
+ steps_per_epoch: 20000
22
+ warmup_steps: 1000
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64df947c6d047fd0140d982589510d12768142bbb9ca88531c6470ceacdd1266
3
+ size 1116401472
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:672b5602c725cafd47011e4e49d4b211fd746d0de2e8e187549f0324685987d0
3
+ size 17098332
tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "cls_token": "<s>",
5
+ "eos_token": "</s>",
6
+ "mask_token": {
7
+ "__type": "AddedToken",
8
+ "content": "<mask>",
9
+ "lstrip": true,
10
+ "normalized": true,
11
+ "rstrip": false,
12
+ "single_word": false
13
+ },
14
+ "model_max_length": 1000000000000000019884624838656,
15
+ "pad_token": "<pad>",
16
+ "sep_token": "</s>",
17
+ "sp_model_kwargs": {},
18
+ "tokenizer_class": "XLMRobertaTokenizer",
19
+ "unk_token": "<unk>"
20
+ }