Instructions to use mrm8488/bert-tiny-finetuned-sms-spam-detection with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mrm8488/bert-tiny-finetuned-sms-spam-detection with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="mrm8488/bert-tiny-finetuned-sms-spam-detection")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("mrm8488/bert-tiny-finetuned-sms-spam-detection") model = AutoModelForSequenceClassification.from_pretrained("mrm8488/bert-tiny-finetuned-sms-spam-detection") - Inference
- Notebooks
- Google Colab
- Kaggle
Bias-ness of Model
#3
by RAWx18 - opened
The Model is Skewed Towards "Important" Messages
or i would rather say reverse the labels and it will start working above 0.8 it is important (ham) and below 0.8 it is spam works for some reason.
Spam Probability : .93.
Your package has been shipped. No action is required on your part.
Spam Probability : .86
Hi , When you get chance call our call centers on the number from our official website.
There you can discuss you can discuss your concerns.
Note we NEVER ask for your personal details like bank account , credit card number etc over phone.
Spam Probability : .89
Hi , When you get chance call our call centers on the number from our official website.
There you can discuss you can discuss your concerns.
Note we always ask for your personal details like bank account , credit card number etc over phone to scam you.
Am I missing something, model seems strange