|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- generated_from_keras_callback |
|
|
model-index: |
|
|
- name: Intent-Classification-Bert-Base-Cased |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- This model card has been generated automatically according to the information Keras had access to. You should |
|
|
probably proofread and complete it, then remove this comment. --> |
|
|
|
|
|
# Intent-Classification-Bert-Base-Cased |
|
|
|
|
|
This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on an [Intent-Classification-Commands](https://huggingface.co/datasets/dipesh/Intent-Classification-Commands) dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Train Loss: 0.6110 |
|
|
- Train Sparse Categorical Accuracy: 0.9836 |
|
|
- Validation Loss: 0.4073 |
|
|
- Validation Sparse Categorical Accuracy: 0.9583 |
|
|
- Epoch: 3 |
|
|
|
|
|
## Model description |
|
|
|
|
|
Base model: 'bert-base-cased' can be used for intent classification. It trained on the [Intent-Classification-Commands](https://huggingface.co/datasets/dipesh/Intent-Classification-Commands) dataset. With the following classes- |
|
|
|
|
|
``` |
|
|
{ |
|
|
"0": "asking date", |
|
|
"1": "asking time", |
|
|
"2": "asking weather", |
|
|
"3": "check internet speed", |
|
|
"4": "click photo", |
|
|
"5": "covid cases", |
|
|
"6": "download youtube video", |
|
|
"7": "goodbye", |
|
|
"8": "greet", |
|
|
"9": "open website", |
|
|
"10": "play games", |
|
|
"11": "play on youtube", |
|
|
"12": "send email", |
|
|
"13": "send whatsapp message", |
|
|
"14": "take screenshot", |
|
|
"15": "tell me about", |
|
|
"16": "tell me joke", |
|
|
"17": "tell me news" |
|
|
} |
|
|
``` |
|
|
|
|
|
## Intended uses & limitations |
|
|
|
|
|
Intent Classifications for Chatbot or Virtual Assistant. |
|
|
Only supports the English language. It can't work in outside classes. But you can fine-tune it for your own use. |
|
|
|
|
|
## Training and evaluation data |
|
|
|
|
|
Dataset Used: [Intent-Classification-Commands](https://huggingface.co/datasets/dipesh/Intent-Classification-Commands) |
|
|
|
|
|
## Training procedure |
|
|
|
|
|
https://colab.research.google.com/drive/1KHg14glvhdV_ziOcY0pHm66PBYoBZMS0?usp=sharing |
|
|
|
|
|
### Training hyperparameters |
|
|
|
|
|
|
|
|
The following hyperparameters were used during training: |
|
|
- optimizer: {'name': 'Adam', 'learning_rate': 5e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False} |
|
|
- training_precision: float32 |
|
|
|
|
|
### Training results |
|
|
|
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.19.2 |
|
|
- TensorFlow 2.8.0 |
|
|
- Datasets 2.2.2 |
|
|
- Tokenizers 0.12.1 |
|
|
|
|
|
|
|
|
## Connect me on- |
|
|
|
|
|
* Subscribe to me on: https://youtube.com/techportofficial |
|
|
|
|
|
* DM me on (for quick response): https://instagram.com/dipesh_pal17 |
|
|
|
