README.md

metadata

tags: autotrain
language: tr
widget:
  - text: Bu ürün gerçekten güzel çıktı
datasets:
  - emre/autotrain-data-turkish-sentiment-analysis
co2_eq_emissions: 120.82460124309924

Model Trained Using AutoTrain

• Problem type: Multi-class Classification
• Model ID: 870727732
• CO2 Emissions (in grams): 120.82460124309924

Validation Metrics

• Loss: 0.1098366305232048
• Accuracy: 0.9697853317600073
• Macro F1: 0.9482820974460786
• Micro F1: 0.9697853317600073
• Weighted F1: 0.9695237873890088
• Macro Precision: 0.9540948884759232
• Micro Precision: 0.9697853317600073
• Weighted Precision: 0.9694186941924757
• Macro Recall: 0.9428467518468838
• Micro Recall: 0.9697853317600073
• Weighted Recall: 0.9697853317600073

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": "Bu ürün gerçekten güzel çıktı"}' https://api-inference.huggingface.co/models/emre/turkish-sentiment-analysis

Or Python API:

from transformers import AutoModelForSequenceClassification, AutoTokenizer

model = AutoModelForSequenceClassification.from_pretrained("emre/turkish-sentiment-analysis", use_auth_token=True)

tokenizer = AutoTokenizer.from_pretrained("emre/turkish-sentiment-analysis", use_auth_token=True)

inputs = tokenizer("Bu ürün gerçekten güzel çıktı", return_tensors="pt")

outputs = model(**inputs)