Downloads · 30 days
9
21% of all-time downloads
zerodte/tradepulse-finbert-sentiment
tradepulse-finbert-sentiment is a text classification model from zerodte. Use it when you need a label for a piece of text. The card lists the license as apache-2.0.
Fine-tuned FinBERT model for financial sentiment analysis in TradePulse.
Downloads · 30 days
9
21% of all-time downloads
All-time downloads
42
Public
Parameters
110M
24.8 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors881 MB · 100%
From the Hugging Face model README
Fine-tuned FinBERT model for financial sentiment analysis in TradePulse.
Task: Sentiment Classification
Target Column: label
Labels: ['negative', 'neutral', 'positive']
Last training: 2026-04-20 17:22
Dataset: base_reference.csv (1797 samples)
| Metric | Value |
|---|---|
| Loss | 0.0000 |
| Accuracy | 1.0000 |
| F1 Score | 1.0000 |
| F1 Macro | 1.0000 |
| Precision | 1.0000 | | Recall | 1.0000 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch
tokenizer = AutoTokenizer.from_pretrained("Bencode92/tradepulse-finbert-sentiment")
model = AutoModelForSequenceClassification.from_pretrained("Bencode92/tradepulse-finbert-sentiment")
# Example prediction
text = "Apple reported strong quarterly earnings beating expectations"
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
outputs = model(**inputs)
predictions = outputs.logits.softmax(dim=-1)