FNST_trad_2i

mrovejaxd/FNST_trad_2i

FNST_trad_2i is a machine learning model from mrovejaxd. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

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

Downloads · 30 days

9% of all-time downloads

All-time downloads

Public

Parameters

110M

21.5 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors439 MB · 100%

At a glance

Model type: bert
Access: Public
Created: Sep 30, 2024
Updated: Oct 1, 2024
SHA: 8dbcf881

Base models

dccuchile/bert-base-spanish-wwm-cased

Model card

From the Hugging Face model README

FNST_trad_2i

This model is a fine-tuned version of dccuchile/bert-base-spanish-wwm-cased on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.2303
Accuracy: 0.4261
F1: 0.2542

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: 1e-07
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 16
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.3933	0.32	500	1.3964	0.2558	0.1034
1.3919	0.64	1000	1.3958	0.2556	0.1033
1.3926	0.96	1500	1.3945	0.2556	0.1033
1.3924	1.28	2000	1.3922	0.2556	0.1033
1.3893	1.6	2500	1.3892	0.2554	0.1036
1.3838	1.92	3000	1.3853	0.2558	0.1045
1.3801	2.24	3500	1.3808	0.2554	0.1050
1.3775	2.56	4000	1.3757	0.2558	0.1060
1.3739	2.88	4500	1.3702	0.2558	0.1073
1.3666	3.2	5000	1.3643	0.2549	0.1084
1.359	3.52	5500	1.3579	0.2556	0.1116
1.356	3.84	6000	1.3513	0.2580	0.1176
1.3502	4.16	6500	1.3445	0.2601	0.1248
1.3423	4.48	7000	1.3378	0.2688	0.1429
1.3344	4.8	7500	1.3309	0.2769	0.1629
1.3308	5.12	8000	1.3245	0.2896	0.1892
1.3249	5.44	8500	1.3183	0.3035	0.2153
1.3188	5.76	9000	1.3124	0.3177	0.2375
1.3112	6.08	9500	1.3068	0.3314	0.2516
1.3111	6.4	10000	1.3017	0.3428	0.2576
1.2993	6.72	10500	1.2968	0.3537	0.2560
1.2998	7.04	11000	1.2925	0.3640	0.2501
1.296	7.36	11500	1.2888	0.3620	0.2349
1.2954	7.68	12000	1.2854	0.3676	0.2236
1.2882	8.0	12500	1.2824	0.3707	0.2156
1.2898	8.32	13000	1.2798	0.3726	0.2074
1.2848	8.64	13500	1.2776	0.3717	0.2007
1.2825	8.96	14000	1.2755	0.3739	0.1966
1.2838	9.28	14500	1.2738	0.3746	0.1955
1.2756	9.6	15000	1.2718	0.3766	0.1888
1.2784	9.92	15500	1.2700	0.3755	0.1821
1.2803	10.24	16000	1.2684	0.3784	0.1814
1.2765	10.56	16500	1.2668	0.3782	0.1799
1.2716	10.88	17000	1.2653	0.3798	0.1806
1.271	11.2	17500	1.2636	0.3807	0.1802
1.2702	11.52	18000	1.2620	0.3813	0.1809
1.2666	11.84	18500	1.2603	0.3825	0.1810
1.2666	12.16	19000	1.2585	0.3842	0.1834
1.2656	12.48	19500	1.2567	0.3865	0.1863
1.2676	12.8	20000	1.2546	0.3883	0.1887
1.26	13.12	20500	1.2526	0.3928	0.1976
1.2595	13.44	21000	1.2505	0.3969	0.2044
1.2574	13.76	21500	1.2482	0.4	0.2093
1.2541	14.08	22000	1.2456	0.4018	0.2112
1.2505	14.4	22500	1.2429	0.4063	0.2186
1.2516	14.72	23000	1.2401	0.4106	0.2263
1.2502	15.04	23500	1.2370	0.4180	0.2396
1.2408	15.36	24000	1.2336	0.4200	0.2423
1.2437	15.68	24500	1.2303	0.4261	0.2542

Framework versions

Transformers 4.37.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.1

Type: bert
Created: Sep 30, 2024
Updated: Oct 1, 2024