ijelid-indobertweet

fathan/ijelid-indobertweet

ijelid-indobertweet is a token classification model from fathan. Use it when you need labels on individual words, such as names. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

3% of all-time downloads

All-time downloads

549

Public

Parameters

110M

9.2 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin440 MB · 50%

Parameter types

How the weights are stored.

F32110M · 100%

Try a prompt

Model card

From the Hugging Face model README

ijelid-indobertweet

This model is a fine-tuned version of indolem/indobertweet-base-uncased on the Indonesian-Javanese-English code-mixed Twitter dataset.

Label ID and its corresponding name:

Label ID	Label Name
LABEL_0	English (EN)
LABEL_1	Indonesian (ID)
LABEL_2	Javanese (JV)
LABEL_3	Mixed Indonesian-English (MIX-ID-EN)
LABEL_4	Mixed Indonesian-Javanese (MIX-ID-JV)
LABEL_5	Mixed Javanese-English (MIX-JV-EN)
LABEL_6	Other (O)

It achieves the following results on the evaluation set:

Loss: 0.2804
Precision: 0.9323
Recall: 0.9394
F1: 0.9356
Accuracy: 0.9587

It achieves the following results on the test set:

Overall Precision: 0.9326
Overall Recall: 0.9421
Overall F1: 0.9371
Overall Accuracy: 0.9643

Model description

More information needed

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: 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	386	0.1666	0.8968	0.9014	0.8982	0.9465
0.257	2.0	772	0.1522	0.9062	0.9368	0.9206	0.9517
0.1092	3.0	1158	0.1462	0.9233	0.9335	0.9280	0.9556
0.0739	4.0	1544	0.1563	0.9312	0.9361	0.9336	0.9568
0.0739	5.0	1930	0.1671	0.9224	0.9413	0.9312	0.9573
0.0474	6.0	2316	0.1719	0.9303	0.9394	0.9346	0.9578
0.0339	7.0	2702	0.1841	0.9249	0.9389	0.9314	0.9576
0.0237	8.0	3088	0.2030	0.9224	0.9380	0.9297	0.9570
0.0237	9.0	3474	0.2106	0.9289	0.9377	0.9331	0.9576
0.0185	10.0	3860	0.2264	0.9277	0.9389	0.9330	0.9571
0.0132	11.0	4246	0.2331	0.9336	0.9344	0.9339	0.9574
0.0101	12.0	4632	0.2403	0.9353	0.9375	0.9363	0.9586
0.0082	13.0	5018	0.2509	0.9311	0.9373	0.9340	0.9582
0.0082	14.0	5404	0.2548	0.9344	0.9351	0.9346	0.9578
0.0062	15.0	5790	0.2608	0.9359	0.9372	0.9365	0.9588
0.005	16.0	6176	0.2667	0.9298	0.9407	0.9350	0.9587
0.0045	17.0	6562	0.2741	0.9337	0.9408	0.9371	0.9592
0.0045	18.0	6948	0.2793	0.9342	0.9371	0.9355	0.9589
0.0035	19.0	7334	0.2806	0.9299	0.9391	0.9342	0.9588
0.0034	20.0	7720	0.2804	0.9323	0.9394	0.9356	0.9587

Framework versions

Transformers 4.21.2
Pytorch 1.7.1
Datasets 2.5.1
Tokenizers 0.12.1

Task: Token Classification
Library: transformers
Type: bert
Created: Oct 4, 2022
Updated: Mar 27, 2023