ift6266h14/experiments/mlp_acoustic/ac160_ph3_rl2_malespkr.yaml at master · jfsantos/ift6266h14 · 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
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
!obj:pylearn2.train.Train {
    dataset: &train !obj:research.code.pylearn2.datasets.timit.TIMIT {
        which_set: 'train',
        frame_length: &flen 160,
        frames_per_example: &fpe 1,
        samples_to_predict: &ylen 1,
        n_next_phones: 1,
        n_prev_phones: 1,
        filter_fn: 'male_speakers'
        #start: 0,
        #stop: 100,
    },
    model: !obj:mlp_with_source.MLPWithSource {
        batch_size: 512,
        layers: [
            !obj:mlp_with_source.CompositeLayerWithSource {
                layer_name: 'c',
                layers: [
                    !obj:pylearn2.models.mlp.RectifiedLinear {
                        layer_name: 'h1',
                        dim: 250,
                        irange: 0.05,
                    },
                    !obj:pylearn2.models.mlp.RectifiedLinear {
                        layer_name: 'h2',
                        dim: 250,
                        irange: 0.05,
                    },
                ],
            },
            !obj:pylearn2.models.mlp.RectifiedLinear {
                layer_name: 'h3',
                dim: 100,
                irange: 0.05,
            },
            !obj:pylearn2.models.mlp.Linear {
                layer_name: 'y',
                dim: *ylen,
                irange: 0.05,
            },
        ],
        input_space: !obj:pylearn2.space.CompositeSpace {
            components: [
                !obj:pylearn2.space.VectorSpace {
                    dim: *flen,
                },
                !obj:pylearn2.space.VectorSpace {
                    dim: 186,
                },
            ],
        },
        input_source: ['features', 'phones'],
    },
    algorithm: !obj:pylearn2.training_algorithms.sgd.SGD {
        learning_rate: .01,
        monitoring_dataset: {
            'train': *train,
            'valid': !obj:research.code.pylearn2.datasets.timit.TIMIT {
                which_set: 'valid',
                frame_length: *flen,
                frames_per_example: *fpe,
                samples_to_predict: *ylen,
                n_next_phones: 1,
                n_prev_phones: 1,
                filter_fn: 'male_speakers'
            },
            'test': !obj:research.code.pylearn2.datasets.timit.TIMIT {
                which_set: 'test',
                frame_length: *flen,
                frames_per_example: *fpe,
                samples_to_predict: *ylen,
                n_next_phones: 1,
                n_prev_phones: 1,
                filter_fn: 'male_speakers'
            },
        },
        cost: !obj:pylearn2.costs.mlp.Default {},
        termination_criterion: !obj:pylearn2.termination_criteria.And {
            criteria: [
                !obj:pylearn2.termination_criteria.MonitorBased {
                    channel_name: "valid_objective",
                    prop_decrease: 0.0000000001,
                    N: 10
                },
                !obj:pylearn2.termination_criteria.EpochCounter {
                    max_epochs: 200
                }
            ]
        }
    },
    extensions: [
        !obj:pylearn2.train_extensions.best_params.MonitorBasedSaveBest {
            channel_name: 'valid_objective',
            save_path: "ac160_ph3_rl2_malespkr.pkl"
        }
    ]
}