ptt5-wikilingua-30epochs

arthurmluz/ptt5-wikilingua-30epochs

ptt5-wikilingua-30epochs is a machine learning model from arthurmluz. 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

45% of all-time downloads

All-time downloads

Public

Repo size

1.8 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin892 MB · 100%

At a glance

Library: transformers
License: mit
Model type: t5
Access: Public
Created: Oct 19, 2023
Updated: Nov 27, 2024
SHA: 8a7b7daa

Base models

unicamp-dl/ptt5-base-portuguese-vocab

Model card

From the Hugging Face model README

ptt5-wikilingua-30epochs

This model is a fine-tuned version of unicamp-dl/ptt5-base-portuguese-vocab on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.9063
Rouge1: 0.2604
Rouge2: 0.1127
Rougel: 0.2222
Rougelsum: 0.2541
Gen Len: 18.4528

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: 2
eval_batch_size: 2
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	Gen Len
2.1668	1.0	28580	2.0384	0.2366	0.0935	0.2034	0.2311	18.2195
2.0348	2.0	57160	1.9725	0.2448	0.0998	0.2098	0.2391	18.3898
2.0152	3.0	85740	1.9346	0.2469	0.1024	0.2122	0.2414	18.2427
1.9769	4.0	114320	1.9096	0.2503	0.1047	0.2147	0.2446	18.2773
1.8471	5.0	142900	1.8957	0.253	0.1076	0.2169	0.2473	18.2612
1.8504	6.0	171480	1.8840	0.2541	0.1084	0.2179	0.2483	18.3317
1.7456	7.0	200060	1.8768	0.2547	0.1084	0.2183	0.2488	18.3634
1.7254	8.0	228640	1.8747	0.2563	0.1099	0.2196	0.2505	18.3577
1.7742	9.0	257220	1.8739	0.2562	0.11	0.2194	0.2504	18.3904
1.7211	10.0	285800	1.8667	0.2572	0.1109	0.2205	0.2513	18.3616
1.696	11.0	314380	1.8677	0.2568	0.1112	0.2204	0.251	18.349
1.6762	12.0	342960	1.8695	0.2571	0.1108	0.2202	0.2513	18.3528
1.6404	13.0	371540	1.8738	0.2582	0.1115	0.2208	0.2523	18.3909
1.6523	14.0	400120	1.8727	0.259	0.1118	0.2215	0.253	18.4077
1.626	15.0	428700	1.8736	0.2596	0.1124	0.2222	0.2537	18.4245
1.5922	16.0	457280	1.8750	0.259	0.1123	0.2215	0.253	18.4125
1.5345	17.0	485860	1.8783	0.2591	0.112	0.2214	0.2529	18.4013
1.5785	18.0	514440	1.8797	0.2588	0.112	0.2212	0.2527	18.3965
1.5097	19.0	543020	1.8868	0.2592	0.1115	0.221	0.2531	18.4567
1.5091	20.0	571600	1.8851	0.2593	0.1124	0.2216	0.2533	18.397
1.5116	21.0	600180	1.8895	0.2599	0.1124	0.2219	0.2537	18.4505
1.5351	22.0	628760	1.8901	0.2606	0.113	0.2225	0.2544	18.4369
1.5125	23.0	657340	1.8953	0.2598	0.1125	0.2218	0.2535	18.4273
1.5246	24.0	685920	1.8980	0.2609	0.1129	0.2226	0.2544	18.4464
1.5113	25.0	714500	1.8990	0.2604	0.1127	0.2221	0.2542	18.4562
1.4814	26.0	743080	1.9029	0.261	0.1133	0.223	0.2547	18.4634
1.5212	27.0	771660	1.9014	0.2606	0.1129	0.2226	0.2544	18.4458
1.4469	28.0	800240	1.9032	0.2609	0.1129	0.2226	0.2546	18.4577
1.4844	29.0	828820	1.9050	0.2602	0.1125	0.2221	0.2539	18.4553
1.4561	30.0	857400	1.9063	0.2604	0.1127	0.2222	0.2541	18.4528

Framework versions

Transformers 4.33.2
Pytorch 2.0.1+cu117
Datasets 2.14.5
Tokenizers 0.13.3

Library: transformers
Type: t5
License: mit
Languages: pt
Created: Oct 19, 2023
Updated: Nov 27, 2024