Open-source model robbert-v2-dutch-base-finetuned-emotion-arousal - Accurately predict emotional arousal in Dutch

Home

Robbert V2 Dutch Base Finetuned Emotion Arousal

Developed by antalvdb

Emotional arousal prediction model fine-tuned based on the RobBERT-v2 Dutch base model

Large Language Model

Transformers

Open Source License:MIT #Dutch Sentiment Analysis #Emotional Arousal Prediction #High-Precision Regression Model

Downloads 57

Release Time : 7/29/2024

Model Overview

This model is a fine-tuned version of the RobBERT-v2 Dutch base model, specifically designed for predicting the emotional arousal of text. It achieved an RMSE of 0.1517 on the evaluation set.

Model Features

Dutch-Specific

Emotion analysis model optimized specifically for Dutch text

High-Precision Prediction

Achieved an RMSE of 0.1517 in emotional arousal prediction tasks

Transfer Learning

Fine-tuned based on the powerful RobBERT-v2 base model

Model Capabilities

Dutch Text Analysis

Emotional Arousal Prediction

Continuous Emotion Scoring

Use Cases

Sentiment Analysis

Social Media Emotion Monitoring

Analyze the emotional intensity of Dutch social media content

Quantifies user emotional responses

Customer Feedback Analysis

Assess the emotional arousal level of Dutch customer reviews

Helps identify high-emotion feedback

🚀 robbert-v2-dutch-base-finetuned-emotion-arousal

This model is a fine - tuned version of pdelobelle/robbert-v2-dutch-base on the None dataset. It can achieve specific evaluation results, which is valuable for relevant natural language processing tasks.

📚 Documentation

Model Information

Property	Details
Model Type	robbert-v2-dutch-base-finetuned-emotion-arousal
Base Model	pdelobelle/robbert-v2-dutch-base
License	MIT

Evaluation Results

This model achieves the following results on the evaluation set:

Loss: 0.0230
Rmse: 0.1517

Training and Evaluation

Training Hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e - 05
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e - 08
lr_scheduler_type: linear
num_epochs: 50

Training Results

Training Loss	Epoch	Step	Validation Loss	Rmse
0.0581	1.0	25	0.0349	0.1868
0.0355	2.0	50	0.0340	0.1845
0.0295	3.0	75	0.0271	0.1645
0.0288	4.0	100	0.0279	0.1670
0.0277	5.0	125	0.0316	0.1777
0.0254	6.0	150	0.0263	0.1620
0.0193	7.0	175	0.0247	0.1571
0.0173	8.0	200	0.0304	0.1745
0.0179	9.0	225	0.0239	0.1547
0.0149	10.0	250	0.0244	0.1563
0.0134	11.0	275	0.0248	0.1576
0.0113	12.0	300	0.0256	0.1601
0.0112	13.0	325	0.0265	0.1627
0.0114	14.0	350	0.0299	0.1730
0.0111	15.0	375	0.0268	0.1638
0.0098	16.0	400	0.0256	0.1599
0.009	17.0	425	0.0252	0.1588
0.0078	18.0	450	0.0256	0.1601
0.0093	19.0	475	0.0235	0.1532
0.009	20.0	500	0.0246	0.1568
0.0084	21.0	525	0.0238	0.1543
0.0083	22.0	550	0.0255	0.1598
0.0074	23.0	575	0.0250	0.1582
0.0079	24.0	600	0.0248	0.1574
0.0077	25.0	625	0.0261	0.1616
0.0073	26.0	650	0.0261	0.1615
0.0071	27.0	675	0.0247	0.1571
0.0068	28.0	700	0.0254	0.1593
0.0062	29.0	725	0.0250	0.1581
0.006	30.0	750	0.0255	0.1597
0.0066	31.0	775	0.0241	0.1553
0.0064	32.0	800	0.0242	0.1555
0.006	33.0	825	0.0240	0.1549
0.0055	34.0	850	0.0244	0.1561
0.0055	35.0	875	0.0235	0.1533
0.0053	36.0	900	0.0241	0.1551
0.0056	37.0	925	0.0238	0.1542
0.0052	38.0	950	0.0248	0.1576
0.0055	39.0	975	0.0247	0.1570
0.0054	40.0	1000	0.0233	0.1526
0.0052	41.0	1025	0.0233	0.1525
0.0048	42.0	1050	0.0231	0.1519
0.0051	43.0	1075	0.0237	0.1538
0.0051	44.0	1100	0.0231	0.1520
0.0053	45.0	1125	0.0234	0.1531
0.0046	46.0	1150	0.0230	0.1517
0.0049	47.0	1175	0.0230	0.1518
0.005	48.0	1200	0.0230	0.1518
0.0047	49.0	1225	0.0237	0.1540
0.0047	50.0	1250	0.0230	0.1517

Framework Versions

Transformers 4.42.4
Pytorch 2.3.1+cu121
Datasets 2.20.0
Tokenizers 0.19.1

📄 License

This project is licensed under the MIT license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご