Fine-tuned BERT for churn detection
Browse files- README.md +28 -0
- config.json +36 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: pt
|
| 3 |
+
tags:
|
| 4 |
+
- bert
|
| 5 |
+
- portuguese
|
| 6 |
+
- text-classification
|
| 7 |
+
- b2b
|
| 8 |
+
license: mit
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# bert-churn-detector-pt
|
| 12 |
+
|
| 13 |
+
BERT fine-tuned para deteccao de risco de cancelamento em reunioes B2B
|
| 14 |
+
|
| 15 |
+
## Modelo
|
| 16 |
+
- **Base**: neuralmind/bert-base-portuguese-cased
|
| 17 |
+
- **Idioma**: Portugues
|
| 18 |
+
- **Dataset**: 77,447 reunioes B2B
|
| 19 |
+
|
| 20 |
+
## Resultados
|
| 21 |
+
|
| 22 |
+
| Metrica | Score |
|
| 23 |
+
|-----------|-------|
|
| 24 |
+
| Accuracy | 0.9995 |
|
| 25 |
+
| Precision | 0.9792 |
|
| 26 |
+
| Recall | 0.9038 |
|
| 27 |
+
| F1 Score | 0.9400 |
|
| 28 |
+
| ROC-AUC | 0.9927 |
|
config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": null,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"directionality": "bidi",
|
| 10 |
+
"dtype": "float32",
|
| 11 |
+
"eos_token_id": null,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.1,
|
| 14 |
+
"hidden_size": 768,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 3072,
|
| 17 |
+
"is_decoder": false,
|
| 18 |
+
"layer_norm_eps": 1e-12,
|
| 19 |
+
"max_position_embeddings": 512,
|
| 20 |
+
"model_type": "bert",
|
| 21 |
+
"num_attention_heads": 12,
|
| 22 |
+
"num_hidden_layers": 12,
|
| 23 |
+
"output_past": true,
|
| 24 |
+
"pad_token_id": 0,
|
| 25 |
+
"pooler_fc_size": 768,
|
| 26 |
+
"pooler_num_attention_heads": 12,
|
| 27 |
+
"pooler_num_fc_layers": 3,
|
| 28 |
+
"pooler_size_per_head": 128,
|
| 29 |
+
"pooler_type": "first_token_transform",
|
| 30 |
+
"problem_type": "single_label_classification",
|
| 31 |
+
"tie_word_embeddings": true,
|
| 32 |
+
"transformers_version": "5.0.0",
|
| 33 |
+
"type_vocab_size": 2,
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 29794
|
| 36 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca059978fb44a75faca776e632cf64d71ece009a641f05f004271b989ccdbe8e
|
| 3 |
+
size 435722200
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_lower_case": false,
|
| 5 |
+
"is_local": false,
|
| 6 |
+
"mask_token": "[MASK]",
|
| 7 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 8 |
+
"pad_token": "[PAD]",
|
| 9 |
+
"sep_token": "[SEP]",
|
| 10 |
+
"strip_accents": null,
|
| 11 |
+
"tokenize_chinese_chars": true,
|
| 12 |
+
"tokenizer_class": "BertTokenizer",
|
| 13 |
+
"unk_token": "[UNK]"
|
| 14 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80af2f5c919e5d3f63f8438d7298a4298084fc989cad8be0df01a5c02b7c09cf
|
| 3 |
+
size 5265
|