Datasets:
File size: 1,102 Bytes
b64c2a9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 |
{
"dataset": "emotion_classification",
"task_type": "dialog_emotion_recognition_multimodal",
"total_target_emotions": 7,
"valid_target_emotions": [
"positive_joy",
"positive_surprise",
"negative_anger",
"negative_surprise",
"negative_disgust",
"negative_fear",
"negative_sadness"
],
"questions_per_emotion": 100,
"total_questions_per_combination": 700,
"emotion_stats": {
"positive_joy": 100,
"positive_surprise": 100,
"negative_anger": 100,
"negative_surprise": 100,
"negative_disgust": 100,
"negative_fear": 100,
"negative_sadness": 100
},
"combinations": [
"audio_vision",
"vision_audio",
"vision_text",
"text_vision",
"text_audio",
"audio_text"
],
"sampling_strategy": "1_correct + 1_neutral + 2_opposite_category",
"face_media_type": "mp4_videos",
"audio_media_type": "wav_dialog_files",
"task_description": "Cross-modal emotion recognition from dialog and conversation data"
} |