Swin-Bert_Mimic

ChayanM/Swin-Bert_Mimic

Swin-Bert_Mimic is a image-text-to-text model from ChayanM. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for transformers.

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

Downloads · 30 days

2% of all-time downloads

All-time downloads

240

Public

Parameters

226M

17.2 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors906 MB · 100%

Parameter types

How the weights are stored.

F32226M · 100%

Try a prompt

Model card

From the Hugging Face model README

Swin-Bert_Mimic

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.1025
Rouge1: 35.8104
Rouge2: 22.5915
Rougel: 34.3056
Rougelsum: 35.1416
Gen Len: 21.289

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-05
train_batch_size: 8
eval_batch_size: 8
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	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
0.0677	1.0	7500	0.0742	34.0952	25.4639	34.0546	34.0407	14.412
0.0621	2.0	15000	0.0686	37.767	26.9356	37.0596	37.4647	18.921
0.0595	3.0	22500	0.0670	38.07	26.9203	37.1384	37.7633	22.422
0.0536	4.0	30000	0.0655	38.064	27.0799	37.3483	37.7981	18.476
0.0484	5.0	37500	0.0655	38.8419	27.551	37.992	38.573	19.552
0.0436	6.0	45000	0.0672	39.2556	27.3445	38.1583	38.9199	19.699
0.0394	7.0	52500	0.0680	38.6881	27.1077	37.6518	38.3678	19.322
0.0355	8.0	60000	0.0697	39.2775	27.1638	38.1169	38.786	20.125
0.0318	9.0	67500	0.0719	38.8973	27.0819	37.8138	38.4725	20.237
0.0265	10.0	75000	0.0746	38.2854	26.3015	37.0627	37.8955	20.799
0.0241	11.0	82500	0.0769	37.7814	25.9821	36.6626	37.3682	20.437
0.0204	12.0	90000	0.0810	37.7945	26.012	36.5089	37.3188	20.945
0.0172	13.0	97500	0.0846	37.5296	25.3082	36.2752	36.9433	20.397
0.0147	14.0	105000	0.0876	36.6675	24.5001	35.264	36.034	22.044
0.012	15.0	112500	0.0907	35.8928	23.4706	34.3812	35.2234	21.344
0.0103	16.0	120000	0.0947	35.6648	22.8131	34.1013	35.0637	22.095
0.0084	17.0	127500	0.0971	35.7702	22.9984	34.2882	35.1362	21.501
0.0068	18.0	135000	0.0996	35.4212	22.3513	33.9646	34.8255	22.152
0.0058	19.0	142500	0.1019	35.9704	23.1195	34.4672	35.3553	21.404
0.0048	20.0	150000	0.1025	35.8104	22.5915	34.3056	35.1416	21.289

Framework versions

Transformers 4.37.1
Pytorch 1.13.1+cu117
Datasets 2.15.0
Tokenizers 0.15.1

Task: Image-Text-to-Text
Library: transformers
Type: vision-encoder-decoder
Created: Apr 13, 2024
Updated: Apr 15, 2024