pump_intent_test

This model is a fine-tuned version of roberta-base on an unknown dataset.

Model description

Custom data generated labeling text according to these three categories. These three categories are the subcategories of Pump - essentially when a user asks a question and expects an answer in response

• Value: a slot value or a calculation
• Clarification: Asking for further information on a previous answer
• Testing: Testing for knowledge of facts and definitions

Takes a user input of string text and classifies it according to one of three categories.

Intended uses & limitations

from transformers import pipeline classifier = pipeline("text-classification",model="mp6kv/pump_intent_test")

output = classifier("What is the value of the length of the blue object?")

score = output[0]['score']

label = output[0]['label']

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Framework versions

• Transformers 4.17.0
• Pytorch 1.10.0+cu111
• Datasets 1.18.3
• Tokenizers 0.11.6