search_params.toml

[grid]
scheduler_type = ["huggingface_cosine_with_restarts", "huggingface_linear", "cosine", "onecycle"]
smoothing_probability = [0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0]
# mamba_ker_size = [3, 4]
# mamba_layer_number = [3, 4, 5]
# hidden_dim_gated     = [128]
# num_transformer_heads = [2, 4, 8]
# tr_layer_number       = [1, 2]
# out_features         = [128, 256]
# num_graph_heads      = [2, 4]
# dropout              = [0.0, 0.1, 0.2]
# positional_encoding  = [true, false]

[defaults]
hidden_dim = 128
# # hidden_dim_gated = 128
# num_transformer_heads = 2
# tr_layer_number = 1
# out_features = 128
# # num_graph_heads = 2
# dropout = 0
# positional_encoding = false