distilbert-base-multilingual-cased-language_detection_tweets

This model is a fine-tuned version of distilbert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0052
Accuracy: 0.9992
F1
- Weighted: 0.9992
- Micro: 0.9992
- Macro: 0.9992
Recall
- Weighted: 0.9992
- Micro: 0.9992
- Macro: 0.9992
Precision
- Weighted: 0.9992
- Micro: 0.9992
- Macro: 0.9992

Model description

This is a classification model of tweets in 3 different languages.

For more information on how it was created, check out the following link: https://github.com/DunnBC22/NLP_Projects/blob/main/Language%20Detection/Language%20Detection%20of%20Tweets/Language%20Detection%20of%20Tweets.ipynb

Intended uses & limitations

This model is intended to demonstrate my ability to solve a complex problem using technology.

Training and evaluation data

Dataset Source: https://www.kaggle.com/datasets/lailaboullous/language-detection-dataset

Input Word Length:

Length of Input Text (in Words)

Input Word Length By Class:

Length of Input Text (in Words) By Class

Class Distribution:

Class Distribution

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Weighted F1	Micro F1	Macro F1	Weighted Recall	Micro Recall	Macro Recall	Weighted Precision	Micro Precision	Macro Precision
0.1713	1.0	164	0.0162	0.9954	0.9954	0.9954	0.9954	0.9954	0.9954	0.9955	0.9954	0.9954	0.9954
0.0088	2.0	328	0.0123	0.9977	0.9977	0.9977	0.9977	0.9977	0.9977	0.9977	0.9977	0.9977	0.9977
0.0025	3.0	492	0.0052	0.9992	0.9992	0.9992	0.9992	0.9992	0.9992	0.9992	0.9992	0.9992	0.9992

Framework versions

Transformers 4.26.1
Pytorch 1.12.1
Datasets 2.9.0
Tokenizers 0.12.1