| {"module": null, "class_name": "Transformer", "config": {"name": "transformer_4", "trainable": true, "dtype": {"module": "keras", "class_name": "DTypePolicy", "config": {"name": "float32"}, "registered_name": null}, "num_layers": 6, "heads": 8, "d_model": 128, "units": 512, "input_vocab": 32000, "target_vocab": 32000, "drop_rate": 0.1}, "registered_name": "Custom>Transformer", "build_config": {"input_shape": {"input_ids": [null, null], "decoder_input_ids": [null, null]}}, "compile_config": {"optimizer": {"module": null, "class_name": "CustomOptimizer", "config": {"name": "adamw", "learning_rate": {"module": "keras.optimizers.schedules", "class_name": "CosineDecay", "config": {"initial_learning_rate": 0.0, "decay_steps": 36250, "alpha": 0.0, "name": "CosineDecay", "warmup_target": 0.0005, "warmup_steps": 4000}, "registered_name": null}, "weight_decay": 0.01, "clipnorm": 1.0, "global_clipnorm": null, "clipvalue": null, "use_ema": false, "ema_momentum": 0.99, "ema_overwrite_frequency": null, "loss_scale_factor": null, "gradient_accumulation_steps": null, "beta_1": 0.9, "beta_2": 0.999, "epsilon": 1e-07, "amsgrad": false}, "registered_name": "CustomOptimizer"}, "loss": {"module": null, "class_name": "MaskedCategoricalCrossEntropy", "config": {"name": "masked_categorical_crossentropy", "reduction": "sum_over_batch_size", "pad_token_id": 3}, "registered_name": "Custom>MaskedCategoricalCrossEntropy"}, "loss_weights": null, "metrics": [{"module": null, "class_name": "MaskedAccuaracy", "config": {"name": "masked_accuracy", "dtype": "float32", "pad_token_id": 3}, "registered_name": "Custom>MaskedAccuaracy"}], "weighted_metrics": null, "run_eagerly": false, "steps_per_execution": 1, "jit_compile": false}} |