checkpoints_29_9_microsoft_deberta_V1

VuongQuoc/checkpoints_29_9_microsoft_deberta_V1

checkpoints_29_9_microsoft_deberta_V1 is a multiple choice model from VuongQuoc. Use it for the multiple choice 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

0% of all-time downloads

All-time downloads

3.2K

Public

Repo size

167 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin1.7 GB · 99%

At a glance

Task: multiple choice
Library: transformers
License: mit
Model type: deberta-v2
Access: Public
Created: Sep 29, 2023
Updated: Sep 30, 2023
SHA: 15f93b00

Base models

microsoft/deberta-v3-large

Model card

From the Hugging Face model README

checkpoints_29_9_microsoft_deberta_V1

This model is a fine-tuned version of microsoft/deberta-v3-large on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.7815
Map@3: 0.8290
Accuracy: 0.7333

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-06
train_batch_size: 1
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Map@3	Accuracy
1.6045	0.05	200	1.6095	0.4593	0.3030
1.3669	0.11	400	1.3360	0.7215	0.5980
0.9993	0.16	600	1.0403	0.7737	0.6727
0.9608	0.21	800	0.9539	0.7966	0.6990
0.9017	0.27	1000	0.9125	0.7997	0.6970
0.885	0.32	1200	0.8719	0.8172	0.7192
0.8222	0.37	1400	0.8462	0.8125	0.7030
0.769	0.43	1600	0.8376	0.8158	0.7131
0.7676	0.48	1800	0.8109	0.8178	0.7152
0.8413	0.53	2000	0.8279	0.8212	0.7212
0.809	0.59	2200	0.8012	0.8212	0.7212
0.8809	0.64	2400	0.8037	0.8290	0.7333
0.8028	0.69	2600	0.7949	0.8249	0.7293
0.8259	0.75	2800	0.7938	0.8283	0.7354
0.7548	0.8	3000	0.7818	0.8300	0.7354
0.7422	0.85	3200	0.7797	0.8316	0.7374
0.801	0.91	3400	0.7811	0.8303	0.7354
0.7	0.96	3600	0.7815	0.8290	0.7333

Framework versions

Transformers 4.32.1
Pytorch 2.0.0
Datasets 2.9.0
Tokenizers 0.13.3

Task: multiple choice
Library: transformers
Type: deberta-v2
License: mit
Created: Sep 29, 2023
Updated: Sep 30, 2023