model_v1_complete_training_wt_init_48_mini

gokuls/model_v1_complete_training_wt_init_48_mini

model_v1_complete_training_wt_init_48_mini is a fill-mask model from gokuls. Use it when you need the model to fill a missing word. It is set up for transformers.

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

Downloads · 30 days

32% of all-time downloads

All-time downloads

Public

Repo size

2.9 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin46.4 MB · 96%

At a glance

Task: Fill-Mask
Library: transformers
Model type: hybridbert
Access: Public
Created: Jun 21, 2023
Updated: Jun 23, 2023
SHA: 4b38e335

Try a prompt

Model card

From the Hugging Face model README

model_v1_complete_training_wt_init_48_mini

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

Loss: 2.7920
Accuracy: 0.4992

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-05
train_batch_size: 48
eval_batch_size: 48
seed: 10
distributed_type: multi-GPU
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 10000
num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
5.9411	0.25	30000	5.8833	0.1518
5.6408	0.49	60000	5.5265	0.1908
4.5385	0.74	90000	4.3133	0.3138
4.1015	0.98	120000	3.8996	0.3583
3.9119	1.23	150000	3.7199	0.3783
3.7832	1.47	180000	3.6039	0.3920
3.6686	1.72	210000	3.5057	0.4033
3.5793	1.97	240000	3.4227	0.4137
3.5128	2.21	270000	3.3645	0.4209
3.4597	2.46	300000	3.3219	0.4261
3.4263	2.7	330000	3.2841	0.4312
3.3909	2.95	360000	3.2547	0.4348
3.3635	3.2	390000	3.2284	0.4379
3.3488	3.44	420000	3.2060	0.4409
3.3239	3.69	450000	3.1872	0.4436
3.3062	3.93	480000	3.1660	0.4462
3.2841	4.18	510000	3.1493	0.4485
3.2663	4.42	540000	3.1355	0.4503
3.259	4.67	570000	3.1229	0.4519
3.2429	4.92	600000	3.1096	0.4535
3.2234	5.16	630000	3.0947	0.4554
3.2115	5.41	660000	3.0818	0.4573
3.2011	5.65	690000	3.0685	0.4590
3.1898	5.9	720000	3.0464	0.4619
3.1651	6.14	750000	3.0226	0.4658
3.1477	6.39	780000	3.0025	0.4689
3.1276	6.64	810000	2.9838	0.4718
3.1102	6.88	840000	2.9690	0.4740
3.1046	7.13	870000	2.9563	0.4757
3.0817	7.37	900000	2.9477	0.4771
3.0813	7.62	930000	2.9397	0.4785
3.0709	7.87	960000	2.9259	0.4804
3.0528	8.11	990000	2.9208	0.4812
3.0541	8.36	1020000	2.9089	0.4829
3.0469	8.6	1050000	2.9015	0.4839
3.0377	8.85	1080000	2.8960	0.4848
3.0284	9.09	1110000	2.8859	0.4861
3.0224	9.34	1140000	2.8819	0.4867
3.019	9.59	1170000	2.8731	0.4878
3.0094	9.83	1200000	2.8687	0.4885
3.0065	10.08	1230000	2.8635	0.4893
2.9983	10.32	1260000	2.8561	0.4900
2.9834	10.57	1290000	2.8524	0.4907
2.9873	10.81	1320000	2.8484	0.4911
2.978	11.06	1350000	2.8414	0.4924
2.9709	11.31	1380000	2.8375	0.4927
2.9695	11.55	1410000	2.8353	0.4932
2.9607	11.8	1440000	2.8290	0.4941
2.9636	12.04	1470000	2.8267	0.4944
2.9584	12.29	1500000	2.8247	0.4946
2.9546	12.54	1530000	2.8196	0.4951
2.9544	12.78	1560000	2.8146	0.4959
2.9486	13.03	1590000	2.8132	0.4964
2.9413	13.27	1620000	2.8099	0.4967
2.9381	13.52	1650000	2.8081	0.4968
2.9389	13.76	1680000	2.8057	0.4973
2.9374	14.01	1710000	2.8028	0.4977
2.9341	14.26	1740000	2.8000	0.4978
2.9275	14.5	1770000	2.7978	0.4984
2.9319	14.75	1800000	2.7947	0.4989
2.9304	14.99	1830000	2.7920	0.4992

Framework versions

Transformers 4.30.2
Pytorch 1.14.0a0+410ce96
Datasets 2.13.0
Tokenizers 0.13.3

Task: Fill-Mask
Library: transformers
Type: hybridbert
Created: Jun 21, 2023
Updated: Jun 23, 2023