deberta-v3-large-survey-rater-sample-1

domenicrosati/deberta-v3-large-survey-rater-sample-1

deberta-v3-large-survey-rater-sample-1 is a text classification model from domenicrosati. 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

8% of all-time downloads

All-time downloads

Public

Repo size

48.7 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin1.7 GB · 100%

At a glance

Task: Text Classification
Library: transformers
License: mit
Model type: deberta-v2
Access: Public
Created: Nov 9, 2023
Updated: Nov 10, 2023
SHA: 2a92a443

Try a prompt

Model card

From the Hugging Face model README

deberta-v3-large-survey-rater-sample-1

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: 0.0866
F1: 0.9786
Precision: 0.9786
Recall: 0.9786
Krippendorff: 0.9798
Krippendorff Swap: -0.0006
Accuracy: 0.9786
Absolute Agreement: 0.9786
Agreement Within One: 0.9975

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: 6e-06
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 4
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1000
num_epochs: 20
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Precision	Recall	Krippendorff	Krippendorff Swap	Accuracy	Absolute Agreement	Agreement Within One
No log	1.0	199	2.0697	0.1667	0.1667	0.1667	-0.6139	0.6225	0.1667	0.1667	1.0
No log	2.0	398	2.2223	0.1667	0.1667	0.1667	-0.4277	0.3187	0.1667	0.1667	0.9167
1.467	3.0	597	2.3201	0.1806	0.1806	0.1806	-0.4999	0.5023	0.1806	0.1806	0.9722
1.467	4.0	796	2.4445	0.2778	0.2778	0.2778	-0.2117	-0.0048	0.2778	0.2778	0.6806
1.467	5.0	995	2.7116	0.2778	0.2778	0.2778	0.0729	0.0473	0.2778	0.2778	0.875
1.0771	6.0	1194	2.7475	0.4444	0.4444	0.4444	0.4812	-0.0044	0.4444	0.4444	0.875
1.0771	7.0	1393	3.0423	0.4306	0.4306	0.4306	0.4082	-0.0048	0.4306	0.4306	0.8194
0.8943	8.0	1592	2.8918	0.375	0.375	0.375	0.5259	-0.0056	0.375	0.375	0.8611
0.8943	9.0	1791	3.3449	0.4583	0.4583	0.4583	0.5552	-0.0049	0.4583	0.4583	0.8889
0.8943	10.0	1990	3.4565	0.375	0.375	0.375	0.4672	-0.0028	0.375	0.375	0.8889
0.7342	11.0	2189	3.6890	0.4028	0.4028	0.4028	0.5451	-0.0064	0.4028	0.4028	0.8889
0.7342	12.0	2388	4.0203	0.3889	0.3889	0.3889	0.5472	-0.0063	0.3889	0.3889	0.8611
0.4512	13.0	2587	4.1194	0.4028	0.4028	0.4028	0.5616	-0.0070	0.4028	0.4028	0.8889
0.4512	14.0	2786	4.6649	0.3889	0.3889	0.3889	0.5616	-0.0070	0.3889	0.3889	0.8889
0.4512	15.0	2985	4.6183	0.4306	0.4306	0.4306	0.5465	-0.0054	0.4306	0.4306	0.8889
0.2842	16.0	3184	4.5873	0.4444	0.4444	0.4444	0.5504	-0.0058	0.4444	0.4444	0.8889
0.2842	17.0	3383	4.7853	0.4306	0.4306	0.4306	0.5259	-0.0068	0.4306	0.4306	0.8611
0.1526	18.0	3582	4.7438	0.4306	0.4306	0.4306	0.5456	-0.0061	0.4306	0.4306	0.8889
0.1526	19.0	3781	4.7335	0.4306	0.4306	0.4306	0.5456	-0.0061	0.4306	0.4306	0.8889
0.1526	20.0	3980	4.7695	0.4444	0.4444	0.4444	0.5495	-0.0064	0.4444	0.4444	0.8889

Framework versions

Transformers 4.18.0
Pytorch 1.11.0
Datasets 2.1.0
Tokenizers 0.12.1

Task: Text Classification
Library: transformers
Type: deberta-v2
License: mit
Created: Nov 9, 2023
Updated: Nov 10, 2023