data2vec-cat

74.3% passed tests (457 passed / 158 failed).

Tests overview

Topic

Passed Tests

Correctness classification

51.0%

Correctness distribution

70.0%

Correctness speaker average

58.3%

Correctness speaker ranking

37.5%

Fairness accent

100.0%

Fairness language

83.3%

Fairness linguistic sentiment

86.5%

Fairness pitch

96.3%

Fairness sex

100.0%

Robustness background noise

38.3%

Robustness low quality phone

90.0%

Robustness recording condition

50.0%

Robustness simulated recording condition

33.3%

Robustness small changes

56.0%

Robustness spectral tilt

80.0%

Model parameters

Entry

Value

model

{‘architecture’: ‘data2vec’, ‘basemodel’: ‘facebook/data2vec-audio-large’, ‘n_layers’: 24}

data

[‘msppodcast’]

sampling_rate

16000

quantization

False

Model information

Entry

Value

Author

aderington

Date

2024-10-02

Name

onnx

Subgroup

ser.basic-4.data2vec

Version

1.0.0

Data

{‘train’: {‘msppodcast’: {‘version’: ‘2.6.1’, ‘tables’: [‘emotion.categories.train.gold_standard’]}}, ‘dev’: {‘msppodcast’: {‘version’: ‘2.6.1’, ‘tables’: [‘emotion.categories.dev.gold_standard’]}}, ‘config’: {‘datasets’: {‘msppodcast-train-dev’: {‘name’: ‘msppodcast’, ‘version’: ‘2.6.1’, ‘tables’: {‘train’: ‘emotion.categories.train.gold_standard’, ‘dev’: ‘emotion.categories.dev.gold_standard’}}, ‘msppodcast-test-1’: {‘name’: ‘msppodcast’, ‘version’: ‘2.6.1’, ‘tables’: {‘test’: ‘emotion.categories.test-1.gold_standard’}}, ‘msppodcast-test-2’: {‘name’: ‘msppodcast’, ‘version’: ‘2.6.1’, ‘tables’: {‘test’: ‘emotion.categories.test-2.gold_standard’}}}}}

Model

{‘basemodel’: ‘facebook/data2vec-audio-large’, ‘n_layers’: 24, ‘gradient_accum_steps’: 2, ‘use_fp16’: True, ‘per_device_train_batch_size’: 4, ‘per_device_eval_batch_size’: 4, ‘learning_rate’: 5e-05, ‘n_epochs’: 5, ‘eval_steps’: 100, ‘save_steps’: 100, ‘min_length_train’: 0.5, ‘max_length_train’: 8, ‘metric_for_eval’: ‘combined’, ‘main_metrics’: {‘emotion’: ‘UAR’}, ‘sampling_rate’: 16000, ‘seed’: 1, ‘tasks’: [‘emotion’], ‘labels’: {‘emotion’: [‘anger’, ‘happiness’, ‘neutral’, ‘sadness’]}, ‘task2problemtype’: {‘emotion’: ‘classification’}, ‘task2heads’: {‘emotion’: 0}, ‘task2postprocessing’: {‘emotion’: None}, ‘task2lossweight’: {‘emotion’: 1}}