| This model is used to determine if a domain name is malicious or not. | |
| It should be used only as a suggestion as it does have some false positives, especialy on domains | |
| from other countries. | |
| --- | |
| tags: | |
| - autotrain | |
| - text-classification | |
| language: | |
| - en | |
| widget: | |
| - text: "I love AutoTrain 🤗" | |
| datasets: | |
| - Cubicz/autotrain-data-domain-filer-2 | |
| co2_eq_emissions: | |
| emissions: 2.854698438703003 | |
| --- | |
| # Model Trained Using AutoTrain | |
| - Problem type: Binary Classification | |
| - Model ID: 3436193686 | |
| - CO2 Emissions (in grams): 2.8547 | |
| ## Validation Metrics | |
| - Loss: 0.124 | |
| - Accuracy: 0.967 | |
| - Precision: 0.963 | |
| - Recall: 0.963 | |
| - AUC: 0.991 | |
| - F1: 0.963 | |
| ## Usage | |
| You can use cURL to access this model: | |
| ``` | |
| $ curl -X POST -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"inputs": "I love AutoTrain"}' https://api-inference.huggingface.co/models/Cubicz/autotrain-domain-filer-2-3436193686 | |
| ``` | |
| Or Python API: | |
| ``` | |
| from transformers import AutoModelForSequenceClassification, AutoTokenizer | |
| model = AutoModelForSequenceClassification.from_pretrained("Cubicz/autotrain-domain-filer-2-3436193686", use_auth_token=True) | |
| tokenizer = AutoTokenizer.from_pretrained("Cubicz/autotrain-domain-filer-2-3436193686", use_auth_token=True) | |
| inputs = tokenizer("I love AutoTrain", return_tensors="pt") | |
| outputs = model(**inputs) | |
| ``` |