Upload folder using huggingface_hub
Browse files- README.md +26 -0
- best-checkpoint.ckpt +3 -0
- config.json +31 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +22 -0
README.md
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
language:
|
| 4 |
+
- ru
|
| 5 |
+
license: mit
|
| 6 |
+
pipeline_tag: text-classification
|
| 7 |
+
tags:
|
| 8 |
+
- bert
|
| 9 |
+
- toxicity
|
| 10 |
+
- russian
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Toxicity Classifier
|
| 14 |
+
|
| 15 |
+
Модель для определения токсичности сообщений службы поддержки.
|
| 16 |
+
|
| 17 |
+
## Метрики
|
| 18 |
+
|
| 19 |
+
Пример результатов на тестовой выборке:
|
| 20 |
+
|
| 21 |
+
accuracy: 0.99
|
| 22 |
+
precision: 0.99
|
| 23 |
+
recall: 0.99
|
| 24 |
+
f1: 0.99
|
| 25 |
+
mcc: 0.99
|
| 26 |
+
roc_auc: 0.99
|
best-checkpoint.ckpt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:317d7b8dd64b40adf0230d78bcc6a0c2e28624901a9886205fc609891ee21c46
|
| 3 |
+
size 1627626449
|
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.2,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"classifier_dropout": 0.2,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.2,
|
| 14 |
+
"hidden_size": 1024,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 4096,
|
| 17 |
+
"is_decoder": false,
|
| 18 |
+
"layer_norm_eps": 1e-05,
|
| 19 |
+
"max_position_embeddings": 514,
|
| 20 |
+
"model_type": "roberta",
|
| 21 |
+
"num_attention_heads": 16,
|
| 22 |
+
"num_hidden_layers": 24,
|
| 23 |
+
"pad_token_id": 1,
|
| 24 |
+
"position_embedding_type": "absolute",
|
| 25 |
+
"problem_type": "single_label_classification",
|
| 26 |
+
"tie_word_embeddings": true,
|
| 27 |
+
"transformers_version": "5.12.1",
|
| 28 |
+
"type_vocab_size": 1,
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"vocab_size": 98505
|
| 31 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2cbeaa212fa20c345d3fc1deac6515006c790373adf0aaaecc125580d64abc53
|
| 3 |
+
size 1619086576
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<s>",
|
| 5 |
+
"clean_up_tokenization_spaces": true,
|
| 6 |
+
"cls_token": "<s>",
|
| 7 |
+
"eos_token": "</s>",
|
| 8 |
+
"errors": "replace",
|
| 9 |
+
"is_local": false,
|
| 10 |
+
"local_files_only": false,
|
| 11 |
+
"mask_token": "<mask>",
|
| 12 |
+
"max_length": 512,
|
| 13 |
+
"model_max_length": 512,
|
| 14 |
+
"pad_token": "<pad>",
|
| 15 |
+
"sep_token": "</s>",
|
| 16 |
+
"stride": 0,
|
| 17 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 18 |
+
"trim_offsets": true,
|
| 19 |
+
"truncation_side": "right",
|
| 20 |
+
"truncation_strategy": "longest_first",
|
| 21 |
+
"unk_token": "<unk>"
|
| 22 |
+
}
|