Spaces:
Sleeping
Sleeping
File size: 1,706 Bytes
2875fe6 |
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 |
model:
target: models.Tiffusion.tiffusion.Tiffusion
params:
seq_length: 192
feature_size: 50
n_layer_enc: 4
n_layer_dec: 4
d_model: 96 # 4 X 24
timesteps: 1000
sampling_timesteps: 1000
loss_type: 'l1'
beta_schedule: 'cosine'
n_heads: 4
mlp_hidden_times: 4
attn_pd: 0.0
resid_pd: 0.0
kernel_size: 5
padding_size: 2
solver:
base_lr: 1.0e-5
max_epochs: 15000
results_folder: ../../../data/Checkpoints_fmri
gradient_accumulate_every: 2
save_cycle: 1500 # max_epochs // 10
ema:
decay: 0.995
update_interval: 10
scheduler:
target: engine.lr_sch.ReduceLROnPlateauWithWarmup
params:
factor: 0.5
patience: 3000
min_lr: 1.0e-5
threshold: 1.0e-1
threshold_mode: rel
warmup_lr: 8.0e-4
warmup: 500
verbose: False
dataloader:
train_dataset:
target: utils.data_utils.real_datasets.fMRIDataset
params:
name: fMRI
proportion: 0.9 # Set to rate < 1 if training conditional generation
data_root: ./data/fMRI
window: 192 # seq_length
save2npy: True
neg_one_to_one: True
seed: 123
period: train
test_dataset:
target: utils.data_utils.real_datasets.fMRIDataset
params:
name: fMRI
proportion: 0.9 # rate
data_root: ./data/fMRI
window: 192 # seq_length
save2npy: True
neg_one_to_one: True
seed: 123
period: test
style: separate
distribution: geometric
coefficient: 1.0e-2
step_size: 5.0e-2
sampling_steps: 250
batch_size: 64
sample_size: 256
shuffle: True |