deberta-v3-base-multi-intent-model

This model is a fine-tuned version of microsoft/deberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.2104
• Accuracy: 0.9028

Model description

The deberta-v3-large-multi-intent-model is a fine-tuned version of Microsoft's DeBERTa-v3-large transformer model, specifically adapted for the task of multi-intent detection. DeBERTa-v3-large leverages advanced architectural features such as disentangled attention and enhanced mask decoding to deliver superior language understanding and representation. This model is designed to accurately identify and classify two distinct intents within a single user utterance, making it highly suitable for applications that require nuanced natural language understanding.

By focusing on dual-intent classification, the model enhances the capability of conversational agents to handle more complex and multifaceted user requests. Whether deployed in virtual assistants, customer support chatbots, or dialog systems, the deberta-v3-large-multi-intent-model ensures that multiple user intentions are recognized and addressed simultaneously, leading to more efficient and effective interactions.

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
5.0161	1.0	1801	3.3852	0.5545
2.3903	2.0	3602	2.4106	0.8619
2.056	3.0	5403	2.2104	0.9028

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.20.3