metadata
library_name: transformers
language:
- en
base_model:
- microsoft/deberta-v3-base
pipeline_tag: text-classification
Model Performance
| Model | Epoch | Learning Rate | Grad Norm (Mean) | Training Loss | Validation Loss | Accuracy | F1 Score (Weighted) | F1 Score (Macro) | Precision (Weighted) | Precision (Macro) | Recall (Weighted) | Recall (Macro) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DeBERTaV3 (Text Only) | 10 | 0.0000050 | 3.755 | 0.102 | 0.309 | 0.913 | 0.914 | 0.858 | 0.918 | 0.855 | 0.913 | 0.868 |
How to Use
from transformers import AutoTokenizer, AutoConfig, pipeline, \
DebertaV2ForSequenceClassification
config = AutoConfig.from_pretrained('wanadzhar913/debertav3-finetuned-banking-transaction-classification-text-only')
model = DebertaV2ForSequenceClassification.from_pretrained('wanadzhar913/debertav3-finetuned-banking-transaction-classification-text-only', config = config)
tokenizer = AutoTokenizer.from_pretrained('wanadzhar913/debertav3-finetuned-banking-transaction-classification-text-only')
pipe = pipeline(
"text-classification",
tokenizer = tokenizer,
model=model,
padding=True,
device=0,
)
pipe([
"Online Banking transfer from CHK 6479 Confirmation# 1425 ",
"DEPOSIT", # Supposed to be 'Payroll'
"SELF LENDER AUSTIN TX 23267 Debit Card Purchase 09/23 10:20a #6410",
"SECU Foundation",
"RECURRING PAYMENT AUTHORIZED ON 06/02 GEICO *AUTO 1036 DC S583153489705993 111",
])
>>>[{'label': 'Internal Account Transfer', 'score': 0.9998998641967773},
>>> {'label': 'Transfer Deposit', 'score': 0.35954612493515015},
>>> {'label': 'Uncategorized', 'score': 0.9998960494995117},
>>> {'label': 'Restaurants', 'score': 0.6260305047035217},
>>> {'label': 'Insurance', 'score': 0.9998502731323242}]