roberta-large_ALL_BCE_translations_multihead_19_shuffled_special_tokens

SotirisLegkas/roberta-large_ALL_BCE_translations_multihead_19_shuffled_special_tokens

roberta-large_ALL_BCE_translations_multihead_19_shuffled_special_tokens is a machine learning model from SotirisLegkas. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as mit.

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

Downloads · 30 days

8% of all-time downloads

All-time downloads

Public

Parameters

705M

2.8 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors2.8 GB · 100%

At a glance

Library: transformers
License: mit
Model type: roberta
Access: Public
Created: May 6, 2024
Updated: May 6, 2024
SHA: 95649575

Base models

Model card

From the Hugging Face model README

roberta-large_ALL_BCE_translations_multihead_19_shuffled_special_tokens

This model is a fine-tuned version of roberta-large on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.8569
F1 Macro 0.1: 0.1078
F1 Macro 0.15: 0.1410
F1 Macro 0.2: 0.1700
F1 Macro 0.25: 0.1957
F1 Macro 0.3: 0.2179
F1 Macro 0.35: 0.2383
F1 Macro 0.4: 0.2580
F1 Macro 0.45: 0.2757
F1 Macro 0.5: 0.2935
F1 Macro 0.55: 0.3110
F1 Macro 0.6: 0.3275
F1 Macro 0.65: 0.3425
F1 Macro 0.7: 0.3592
F1 Macro 0.75: 0.3717
F1 Macro 0.8: 0.3829
F1 Macro 0.85: 0.3903
F1 Macro 0.9: 0.3847
F1 Macro 0.95: 0.3225
Threshold 0: 0.85
Threshold 1: 0.8
Threshold 2: 0.9
Threshold 3: 0.9
Threshold 4: 0.8
Threshold 5: 0.8
Threshold 6: 0.8
Threshold 7: 0.9
Threshold 8: 0.85
Threshold 9: 0.8
Threshold 10: 0.9
Threshold 11: 0.85
Threshold 12: 0.9
Threshold 13: 0.85
Threshold 14: 0.85
Threshold 15: 0.9
Threshold 16: 0.85
Threshold 17: 0.9
Threshold 18: 0.9
0: 0.1654
1: 0.3112
2: 0.3764
3: 0.3436
4: 0.4800
5: 0.4880
6: 0.4593
7: 0.3694
8: 0.3882
9: 0.5533
10: 0.5439
11: 0.5492
12: 0.2443
13: 0.2278
14: 0.4014
15: 0.3373
16: 0.4511
17: 0.6215
18: 0.2339
Max F1: 0.3903
Mean F1: 0.3971

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: 5e-06
train_batch_size: 8
eval_batch_size: 8
seed: 2024
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 3
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Macro 0.1	F1 Macro 0.15	F1 Macro 0.2	F1 Macro 0.25	F1 Macro 0.3	F1 Macro 0.35	F1 Macro 0.4	F1 Macro 0.45	F1 Macro 0.5	F1 Macro 0.55	F1 Macro 0.6	F1 Macro 0.65	F1 Macro 0.7	F1 Macro 0.75	F1 Macro 0.8	F1 Macro 0.85	F1 Macro 0.9	F1 Macro 0.95	Threshold 0	Threshold 1	Threshold 2	Threshold 3	Threshold 4	Threshold 5	Threshold 6	Threshold 7	Threshold 8	Threshold 9	Threshold 10	Threshold 11	Threshold 12	Threshold 13	Threshold 14	Threshold 15	Threshold 16	Threshold 17	Threshold 18	0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	Max F1	Mean F1
1.1311	1.0	5595	0.8741	0.0694	0.0845	0.1012	0.1184	0.1361	0.1543	0.1721	0.1896	0.2072	0.2254	0.2452	0.2638	0.2836	0.3001	0.3161	0.3204	0.3019	0.2198	0.75	0.8	0.85	0.9	0.7	0.8	0.85	0.85	0.8	0.8	0.95	0.8	0.85	0.9	0.9	0.9	0.85	0.95	0.9	0.0977	0.2012	0.3069	0.2180	0.3982	0.4146	0.4235	0.3110	0.3433	0.5029	0.5039	0.5275	0.2241	0.1802	0.3434	0.2343	0.3988	0.6105	0.2014	0.3204	0.3390
0.7682	2.0	11190	0.8513	0.0938	0.1227	0.1492	0.1724	0.1944	0.2135	0.2336	0.2515	0.2706	0.2880	0.3058	0.3210	0.3374	0.3576	0.3733	0.3780	0.3697	0.3019	0.8	0.85	0.85	0.9	0.8	0.9	0.8	0.9	0.9	0.8	0.9	0.85	0.9	0.8	0.85	0.9	0.85	0.9	0.9	0.1535	0.3002	0.3611	0.3365	0.4672	0.4768	0.4414	0.3609	0.3684	0.5407	0.5423	0.5455	0.2423	0.1915	0.3768	0.3296	0.4296	0.6282	0.2284	0.3780	0.3853
0.606	3.0	16785	0.8569	0.1078	0.1410	0.1700	0.1957	0.2179	0.2383	0.2580	0.2757	0.2935	0.3110	0.3275	0.3425	0.3592	0.3717	0.3829	0.3903	0.3847	0.3225	0.85	0.8	0.9	0.9	0.8	0.8	0.8	0.9	0.85	0.8	0.9	0.85	0.9	0.85	0.85	0.9	0.85	0.9	0.9	0.1654	0.3112	0.3764	0.3436	0.4800	0.4880	0.4593	0.3694	0.3882	0.5533	0.5439	0.5492	0.2443	0.2278	0.4014	0.3373	0.4511	0.6215	0.2339	0.3903	0.3971

Framework versions

Transformers 4.39.3
Pytorch 2.2.2+cu121
Datasets 2.18.0
Tokenizers 0.15.2

Library: transformers
Type: roberta
License: mit
Created: May 6, 2024
Updated: May 6, 2024