MultiModalSA/Hyperparameters.json at master · Ighina/MultiModalSA · GitHub

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
38
39
40
41
42
43
44
{
	"--data_dir": "data",
	"--exclude_audio": "FALSE",
	"--exclude_video":	"FALSE",
	"--AttentionModel": "TRUE",
	"--Transformer": "FALSE",
	"--Convolution": "FALSE",
	"--Raw": "FALSE",
	"--ModifiedAttention": "FALSE",
	"--EarlyFusion": "FALSE",
	"--LateFusion": "FALSE",
	"--TensorFusion": "FALSE",
	"--Normalise": "TRUE",
	"--audio_shape": 74,
	"--video_shape": 35,
	"--text_shape": 768,
	"--output_dimension": 1,
	"--hidden_audio_size": 64,
	"--hidden_video_size": 64,
	"--hidden_text_size": 64,
	"--hidden_size": 32,
	"--num_layers": 1,
	"--learning_rate": 0.00005,
	"--batch_size": 32,
	"--num_epochs": 200,
	"--shuffle": "FALSE",
	"--bidirectional": "TRUE",
	"--early_stop": 10,
	"--dropout_in": 0.2,
	"--dropout_out": 0.0,
	"--dropout_encoder_in": 0.0,
	"--dropout_encoder_out": 0.0,
	"--optimizer": "Adam",
	"--criterion": "MSE",
	"--clip": 4.0,
	"--save_model": "TRUE",
	"--seed_value": 56,
	"--From_checkpoint": "FALSE",
	"--audio_file": "COAVAREP_aligned_MOSEI.pkl",
	"--video_file": "FACET_aligned_MOSEI.pkl",
	"--text_file": "BERT_MOSEI.pkl",
	"--training_file": "df_MOSEI.tsv",
	"--validation_file": "df_valid_MOSEI.tsv"
}