ZJaume commited on
Commit
b2d118f
1 Parent(s): 6d68196

Full models v2.0

Browse files
.gitattributes CHANGED
@@ -32,3 +32,5 @@ 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
+ lm.* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ - hu
5
+ - multilingual
6
+ license: cc-by-sa-4.0
7
+ tags:
8
+ - bicleaner-ai
9
+ tasks:
10
+ - text-classification
11
+ ---
12
+
13
+ # Bicleaner AI full model for en-hu
14
+ Bicleaner AI is a tool that aims at detecting noisy sentence pairs in a parallel corpus. It
15
+ 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).
16
+ Sentence pairs considered very noisy are scored with 0.
17
+
18
+ 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-hu",
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": "4.26.0",
27
+ "type_vocab_size": 1,
28
+ "use_cache": true,
29
+ "vocab_size": 250002
30
+ }
lm.en ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b95661023cc4cf545d4044343e39c052fbda3a45cfda6097ae3baaf8db7c1091
3
+ size 142219749
lm.hu ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8eda64008b5b627fdec43a598e337e31dd8ecd81b8e79ca5283225953e3d093
3
+ size 201085554
metadata.yaml ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bicleaner_ai_version: 2.1.0
2
+ precision_score: 0.871
3
+ recall_score: 0.939
4
+ f1_score: 0.904
5
+ matthews_corr_coef: 0.893
6
+ source_lang: en
7
+ target_lang: hu
8
+ porn_removal_file: porn_removal.ftz
9
+ porn_removal_side: sl
10
+ source_lm: lm.en
11
+ target_lm: lm.hu
12
+ lm_type: CHARACTER
13
+ clean_mean_perp: -1.0598056354291676
14
+ clean_stddev_perp: 0.1914313482325003
15
+ noisy_mean_perp: -4.163701672983073
16
+ noisy_stddev_perp: 0.19304848303193686
17
+ classifier_type: xlmr
18
+ classifier_settings:
19
+ activation: relu
20
+ base_model: jplu/tf-xlm-roberta-base
21
+ batch_size: 96
22
+ calibration_params: [8.60229492, -3.50441098]
23
+ epochs: 15
24
+ loss: binary_crossentropy
25
+ maxlen: 150
26
+ model_name: bitextor/bicleaner-ai-full-en-hu
27
+ n_classes: 2
28
+ n_hidden: 2048
29
+ patience: 3
30
+ steps_per_epoch: 3000
31
+ warmup_steps: 1000
porn_removal.ftz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6c6c0baff6fae466af596d37c66f7d8395e80db44be49001ca20a2f212f62c0
3
+ size 1213624
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:bdc2e46569ce0343f22848632d1c56027cad58f599e8d19ddedff33cc443c265
3
+ size 1116401536
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 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "__type": "AddedToken",
7
+ "content": "<mask>",
8
+ "lstrip": true,
9
+ "normalized": true,
10
+ "rstrip": false,
11
+ "single_word": false
12
+ },
13
+ "model_max_length": 1000000000000000019884624838656,
14
+ "name_or_path": "jplu/tf-xlm-roberta-base",
15
+ "pad_token": "<pad>",
16
+ "sep_token": "</s>",
17
+ "sp_model_kwargs": {},
18
+ "special_tokens_map_file": null,
19
+ "tokenizer_class": "XLMRobertaTokenizer",
20
+ "unk_token": "<unk>"
21
+ }