| [ |
| { |
| "trial": 2, |
| "name": "ssm_ss64_nl4_emb64_do0.1_lr0.001_bs32_ep30", |
| "params": { |
| "model.state_size": 64, |
| "model.num_layers": 4, |
| "model.input_size": 64, |
| "model.dropout": 0.1, |
| "training.learning_rate": 0.001, |
| "training.batch_size": 32, |
| "training.epochs": 30 |
| }, |
| "best_val_loss": 5.375143, |
| "best_val_acc": 0.188612, |
| "best_epoch": 27, |
| "epochs_run": 30, |
| "param_count": 1407632, |
| "model_path": "models/ssm_ss64_nl4_emb64_do0.1_lr0.001_bs32_ep30.pt" |
| }, |
| { |
| "trial": 4, |
| "name": "ssm_ss128_nl4_emb64_do0.1_lr0.001_bs32_ep30", |
| "params": { |
| "model.state_size": 128, |
| "model.num_layers": 4, |
| "model.input_size": 64, |
| "model.dropout": 0.1, |
| "training.learning_rate": 0.001, |
| "training.batch_size": 32, |
| "training.epochs": 30 |
| }, |
| "best_val_loss": 5.369389, |
| "best_val_acc": 0.188401, |
| "best_epoch": 29, |
| "epochs_run": 30, |
| "param_count": 1440656, |
| "model_path": "models/ssm_ss128_nl4_emb64_do0.1_lr0.001_bs32_ep30.pt" |
| }, |
| { |
| "trial": 6, |
| "name": "ssm_ss256_nl4_emb64_do0.1_lr0.001_bs32_ep30", |
| "params": { |
| "model.state_size": 256, |
| "model.num_layers": 4, |
| "model.input_size": 64, |
| "model.dropout": 0.1, |
| "training.learning_rate": 0.001, |
| "training.batch_size": 32, |
| "training.epochs": 30 |
| }, |
| "best_val_loss": 5.367797, |
| "best_val_acc": 0.188095, |
| "best_epoch": 29, |
| "epochs_run": 30, |
| "param_count": 1506704, |
| "model_path": "models/ssm_ss256_nl4_emb64_do0.1_lr0.001_bs32_ep30.pt" |
| }, |
| { |
| "trial": 3, |
| "name": "ssm_ss128_nl2_emb64_do0.1_lr0.001_bs32_ep30", |
| "params": { |
| "model.state_size": 128, |
| "model.num_layers": 2, |
| "model.input_size": 64, |
| "model.dropout": 0.1, |
| "training.learning_rate": 0.001, |
| "training.batch_size": 32, |
| "training.epochs": 30 |
| }, |
| "best_val_loss": 5.470606, |
| "best_val_acc": 0.184723, |
| "best_epoch": 23, |
| "epochs_run": 30, |
| "param_count": 1365392, |
| "model_path": "models/ssm_ss128_nl2_emb64_do0.1_lr0.001_bs32_ep30.pt" |
| }, |
| { |
| "trial": 5, |
| "name": "ssm_ss256_nl2_emb64_do0.1_lr0.001_bs32_ep30", |
| "params": { |
| "model.state_size": 256, |
| "model.num_layers": 2, |
| "model.input_size": 64, |
| "model.dropout": 0.1, |
| "training.learning_rate": 0.001, |
| "training.batch_size": 32, |
| "training.epochs": 30 |
| }, |
| "best_val_loss": 5.459764, |
| "best_val_acc": 0.184554, |
| "best_epoch": 18, |
| "epochs_run": 28, |
| "param_count": 1398416, |
| "model_path": "models/ssm_ss256_nl2_emb64_do0.1_lr0.001_bs32_ep30.pt" |
| }, |
| { |
| "trial": 1, |
| "name": "ssm_ss64_nl2_emb64_do0.1_lr0.001_bs32_ep30", |
| "params": { |
| "model.state_size": 64, |
| "model.num_layers": 2, |
| "model.input_size": 64, |
| "model.dropout": 0.1, |
| "training.learning_rate": 0.001, |
| "training.batch_size": 32, |
| "training.epochs": 30 |
| }, |
| "best_val_loss": 5.472807, |
| "best_val_acc": 0.1843, |
| "best_epoch": 21, |
| "epochs_run": 30, |
| "param_count": 1348880, |
| "model_path": "models/ssm_ss64_nl2_emb64_do0.1_lr0.001_bs32_ep30.pt" |
| } |
| ] |
