mdeberta-v3-base-finetuned-climate-support-new

mljn/mdeberta-v3-base-finetuned-climate-support-new

mdeberta-v3-base-finetuned-climate-support-new is a text classification model from mljn. Use it when you need a label for a piece of text. 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

32% of all-time downloads

All-time downloads

Public

Parameters

279M

26.8 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors1.1 GB · 98%

At a glance

Task: Text Classification
Library: transformers
License: mit
Model type: deberta-v2
Access: Public
Created: Oct 9, 2025
Updated: Oct 10, 2025
SHA: a131adaa

Try a prompt

Base models

microsoft/mdeberta-v3-base

Model card

From the Hugging Face model README

mdeberta-v3-base-finetuned-climate-support-new

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

Loss: 0.5523
Accuracy: 0.8990
Accuracy Balanced: 0.8461
Precision Macro: 0.8550
Recall Macro: 0.8461
F1 Macro: 0.8504
Precision Micro: 0.8990
Recall Micro: 0.8990
F1 Micro: 0.8990
Precision Weighted: 0.8978
Recall Weighted: 0.8990
F1 Weighted: 0.8983
Precision Class 0: 0.7788
Recall Class 0: 0.7521
F1 Class 0: 0.7652
Support Class 0: 468
Precision Class 1: 0.9312
Recall Class 1: 0.9401
F1 Class 1: 0.9356
Support Class 1: 1670

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: 8
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.06
num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Accuracy Balanced	Precision Macro	Recall Macro	F1 Macro	Precision Micro	Recall Micro	F1 Micro	Precision Weighted	Recall Weighted	F1 Weighted	Precision Class 0	Recall Class 0	F1 Class 0	Support Class 0	Precision Class 1	Recall Class 1	F1 Class 1	Support Class 1
0.4698	0.6046	500	0.4537	0.8770	0.7744	0.8435	0.7744	0.8010	0.8770	0.8770	0.8770	0.8714	0.8770	0.8701	0.7937	0.5919	0.6781	468	0.8932	0.9569	0.9240	1670
0.3089	1.2092	1000	0.5018	0.8873	0.7825	0.8692	0.7825	0.8145	0.8873	0.8873	0.8873	0.8839	0.8873	0.8798	0.8429	0.5962	0.6984	468	0.8954	0.9689	0.9307	1670
0.2397	1.8138	1500	0.4276	0.8896	0.8178	0.8480	0.8178	0.8314	0.8896	0.8896	0.8896	0.8862	0.8896	0.8871	0.7802	0.6902	0.7324	468	0.9159	0.9455	0.9305	1670
0.1779	2.4184	2000	0.5046	0.8648	0.8543	0.7991	0.8543	0.8200	0.8648	0.8648	0.8648	0.8839	0.8648	0.8705	0.6484	0.8355	0.7302	468	0.9498	0.8731	0.9098	1670
0.1417	3.0230	2500	0.5791	0.8784	0.8506	0.8168	0.8506	0.8315	0.8784	0.8784	0.8784	0.8870	0.8784	0.8815	0.6919	0.8013	0.7426	468	0.9417	0.9	0.9204	1670
0.091	3.6276	3000	0.5523	0.8990	0.8461	0.8550	0.8461	0.8504	0.8990	0.8990	0.8990	0.8978	0.8990	0.8983	0.7788	0.7521	0.7652	468	0.9312	0.9401	0.9356	1670

Framework versions

Transformers 4.56.2
Pytorch 2.8.0+cu126
Datasets 4.0.0
Tokenizers 0.22.1

Task: Text Classification
Library: transformers
Type: deberta-v2
License: mit
Created: Oct 9, 2025
Updated: Oct 10, 2025