Glossary entry
Inference
इन्फरेन्स
Definition (English)
Running a trained model to generate outputs or predictions on new inputs.
अर्थ (नेपालीमा)
नयाँ इनपुटमा नतिजा वा अनुमान निकाल्न तालिम पाएको मोडेल चलाउने काम।
Related terms
Other terms related to Inference.
-
Batch Inference
ब्याच इन्फरेन्स
Running predictions on many inputs together rather than one at a time.
-
Inference Endpoint
इन्फरेन्स एन्डपोइन्ट
A network address where applications send inputs to get model predictions.
-
Large Language Model (LLM)
ठूलो भाषा मोडेल
A very large neural network trained on massive text to understand and generate language.
-
Model Serving
मोडेल सर्भिङ
Hosting a trained model so applications can send it requests and get predictions.
-
Supervised Learning
पर्यवेक्षित सिकाइ
Training a model on labeled examples so it learns to map inputs to known outputs.
-
Accuracy
शुद्धता
The proportion of predictions a model gets correct out of all predictions.
-
Adversarial Attack
प्रतिकूल आक्रमण
Deliberately crafted inputs designed to fool an AI model into making mistakes.
-
Batch Normalization
ब्याच नर्मलाइजेसन
A technique that stabilizes and speeds up training by normalizing layer inputs.