clip-roberta-finetuned

sharkMeow/clip-roberta-finetuned

clip-roberta-finetuned is a feature extraction model from sharkMeow. Use it when you need embeddings to search or compare text. It is set up for transformers. The card lists the license as gpl-3.0.

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

Downloads · 30 days

34% of all-time downloads

All-time downloads

Public

Parameters

189M

24.2 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors756 MB · 100%

At a glance

Task: Feature Extraction
Library: transformers
License: gpl-3.0
Model type: vision-text-dual-encoder
Access: Public
Created: Jun 27, 2024
Updated: Jul 30, 2024
SHA: 5f8b6946

Base models

ckiplab/bert-base-chinese

Model card

From the Hugging Face model README

clip-roberta-finetuned

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

Loss: 0.3715

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: 64
eval_batch_size: 100
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 100.0

Training results

Training Loss	Epoch	Step	Validation Loss
3.3102	10.0	390	2.7681
1.6079	20.0	780	1.5404
0.7749	30.0	1170	0.9966
0.4468	40.0	1560	0.7465
0.2965	50.0	1950	0.5970
0.2199	60.0	2340	0.5014
0.1751	70.0	2730	0.4469
0.1487	80.0	3120	0.4024
0.1317	90.0	3510	0.3746
0.1234	100.0	3900	0.3715

Framework versions

Transformers 4.42.3
Pytorch 2.3.1+cu121
Datasets 2.20.0
Tokenizers 0.19.1

Task: Feature Extraction
Library: transformers
Type: vision-text-dual-encoder
License: gpl-3.0
Created: Jun 27, 2024
Updated: Jul 30, 2024