Model Card for Model ID

Intent classification is the act of classifying customer's in to different pre defined categories. Sometimes intent classification is referred to as topic classification. By fine tuning a T5 model with prompts containing sythetic data that resembles customer's requests this model is able to classify intents in a dynamic way by adding all of the categories to the prompt

Model Details

Fine tuned Flan-T5-Base

Model Description

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

Developed by: Serj Smorodinsky
Model type: Flan-T5-Base
Language(s) (NLP): [More Information Needed]
License: [More Information Needed]
Finetuned from model [optional]: Flan-T5-Base

Model Sources [optional]

Repository: https://github.com/SerjSmor/intent_classification

How to Get Started with the Model

[More Information Needed]

Training Details

Training Data

https://github.com/SerjSmor/intent_classification HF dataset will be added in the future.

[More Information Needed]

Training Procedure

https://github.com/SerjSmor/intent_classification/blob/main/t5_generator_trainer.py

Using HF trainer

training_args = TrainingArguments(
    output_dir='./results',
    num_train_epochs=epochs,
    per_device_train_batch_size=batch_size,
    per_device_eval_batch_size=batch_size,
    warmup_steps=500,
    weight_decay=0.01,
    logging_dir='./logs',
    logging_steps=10,
    evaluation_strategy="epoch"
)

trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=train_dataset,
    eval_dataset=val_dataset,
    tokenizer=tokenizer,
    # compute_metrics=compute_metrics
)

Serj
/

intent-classifier