mt5-small-finetuned-xsum

steveabecassis/mt5-small-finetuned-xsum

mt5-small-finetuned-xsum is a machine learning model from steveabecassis. 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 apache-2.0.

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

Downloads · 30 days

17% of all-time downloads

All-time downloads

Public

Repo size

2.4 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.json16.3 MB · 79%

At a glance

Library: transformers
License: apache-2.0
Model type: mt5
Access: Public
Created: Jan 4, 2023
Updated: May 4, 2023
SHA: e4fc8422

Model card

From the Hugging Face model README

mt5-small-finetuned-xsum

This model is a fine-tuned version of google/mt5-small on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 2.5196
Rouge1: 0.3378
Rouge2: 0.275
Rougel: 0.3372
Rougelsum: 0.3367

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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
No log	1.0	21	11.8500	0.0	0.0	0.0	0.0
No log	2.0	42	11.1279	0.0	0.0	0.0	0.0
No log	3.0	63	10.0382	0.0	0.0	0.0	0.0
No log	4.0	84	9.1579	0.0	0.0	0.0	0.0
No log	5.0	105	8.6827	0.0	0.0	0.0	0.0
No log	6.0	126	7.3651	0.0028	0.0016	0.0028	0.0028
No log	7.0	147	6.4400	0.019	0.0129	0.0191	0.0197
No log	8.0	168	5.2631	0.0272	0.0229	0.0288	0.0288
No log	9.0	189	4.5832	0.1095	0.0688	0.1053	0.1051
No log	10.0	210	4.2350	0.1263	0.0824	0.1216	0.1235
No log	11.0	231	3.9249	0.1541	0.1051	0.1513	0.1532
No log	12.0	252	3.5469	0.1701	0.1156	0.1665	0.1683
No log	13.0	273	3.3689	0.2672	0.2095	0.2667	0.2659
No log	14.0	294	3.1733	0.3102	0.2483	0.3103	0.3104
No log	15.0	315	3.0810	0.3073	0.2457	0.3074	0.3071
No log	16.0	336	3.0005	0.3071	0.2451	0.3075	0.3069
No log	17.0	357	2.9663	0.3015	0.2364	0.3022	0.3018
No log	18.0	378	2.8718	0.3195	0.2583	0.3197	0.3187
No log	19.0	399	2.8061	0.3159	0.2554	0.316	0.3143
No log	20.0	420	2.7009	0.3351	0.273	0.3338	0.3341
No log	21.0	441	2.6307	0.3384	0.2763	0.3382	0.3381
No log	22.0	462	2.6006	0.3364	0.2743	0.3362	0.3357
No log	23.0	483	2.5819	0.3334	0.2712	0.3331	0.3333
13.1102	24.0	504	2.5606	0.3309	0.269	0.3302	0.3305
13.1102	25.0	525	2.5458	0.338	0.2744	0.3369	0.3373
13.1102	26.0	546	2.5366	0.3361	0.2715	0.3352	0.3352
13.1102	27.0	567	2.5301	0.3413	0.2787	0.3408	0.3406
13.1102	28.0	588	2.5236	0.341	0.2783	0.3402	0.3401
13.1102	29.0	609	2.5206	0.3405	0.2779	0.3399	0.3397
13.1102	30.0	630	2.5196	0.3378	0.275	0.3372	0.3367

Framework versions

Transformers 4.26.0.dev0
Pytorch 1.13.0
Datasets 2.8.0
Tokenizers 0.13.2

Library: transformers
Type: mt5
License: apache-2.0
Created: Jan 4, 2023
Updated: May 4, 2023