| { |
| "configuration": { |
| "framework": "jax", |
| "weights": null, |
| "seed": 123, |
| "output_base_dir": "outputs", |
| "device": { |
| "gpu_ids": [ |
| 0 |
| ], |
| "memory_limit": 1, |
| "keras_backend": "jax", |
| "precision": "float32" |
| }, |
| "num_workers": 4, |
| "train": { |
| "batch_size": 64, |
| "num_epochs": 20, |
| "learning_rate": 5e-05, |
| "gradient_clip_val": 1.0, |
| "grad_accum_steps": 1, |
| "early_stopping": { |
| "patience": 5, |
| "min_improvement": 0.01 |
| } |
| }, |
| "eval": { |
| "fast_evaluation": true, |
| "batch_size": 256, |
| "save_predictions": true, |
| "predictions_dir": "predictions", |
| "run_initial_eval": false, |
| "run_test_after_training": true |
| }, |
| "logging": { |
| "project_name": "NewsReX", |
| "enable_wandb": true, |
| "experiment_name": "jax/MIND-small/CAUM", |
| "wandb_group": "jax/MIND-small/CAUM", |
| "progress_backend": "tqdm" |
| }, |
| "metrics": { |
| "params": {}, |
| "_target_": "src.core.metrics.functions.NewsRecommenderMetrics" |
| }, |
| "multi_seed": { |
| "enabled": true, |
| "seeds": [ |
| 42, |
| 123, |
| 456 |
| ] |
| }, |
| "cache": { |
| "root_dir": ".cache", |
| "clear_on_start": false |
| }, |
| "spec": { |
| "model": { |
| "name": "caum", |
| "architecture": { |
| "news_encoder": { |
| "type": "multi_head_self_attention", |
| "num_heads": 20, |
| "head_dim": 20, |
| "attention_hidden_dim": 200, |
| "entity_embedding_dim": 100, |
| "entity_num_heads": 4, |
| "entity_head_dim": 40, |
| "category_embedding_dim": 100 |
| }, |
| "user_encoder": { |
| "type": "candidate_aware", |
| "candi_selfatt": { |
| "num_heads": 20, |
| "head_dim": 20 |
| }, |
| "candi_cnn": { |
| "half_window": 1 |
| }, |
| "candi_att": { |
| "hidden_dim": 400, |
| "mid_dim": 256 |
| } |
| }, |
| "click_predictor": { |
| "type": "dot_product" |
| } |
| }, |
| "embedding": { |
| "size": 300, |
| "trainable": true |
| }, |
| "news_dim": 400, |
| "use_entity": true, |
| "use_category": true, |
| "dropout_rate": 0.2, |
| "seed": 42 |
| }, |
| "inputs": { |
| "title": { |
| "max_length": 30 |
| }, |
| "history": { |
| "max_length": 50 |
| }, |
| "impressions": { |
| "max_length": 5 |
| }, |
| "max_entities": 5, |
| "process_title": true, |
| "process_abstract": false, |
| "process_category": true, |
| "process_subcategory": false, |
| "process_user_id": false, |
| "process_entities": true |
| }, |
| "training": { |
| "loss": { |
| "name": "categorical_crossentropy", |
| "from_logits": true, |
| "reduction": "sum_over_batch_size", |
| "label_smoothing": 0.0 |
| }, |
| "optimizer": "adam", |
| "learning_rate": 5e-05, |
| "batch_size": 64, |
| "num_epochs": 20, |
| "gradient_clip_val": 1.0, |
| "grad_accum_steps": 1, |
| "early_stopping": { |
| "patience": 5, |
| "min_improvement": 0.01 |
| }, |
| "negative_sampling": { |
| "strategy": "random", |
| "candidates": 4 |
| } |
| }, |
| "evaluation": { |
| "mode": "fast", |
| "evaluator": "caum", |
| "metrics": [ |
| "auc", |
| "mrr", |
| "ndcg@5", |
| "ndcg@10" |
| ], |
| "batch_size": 256 |
| } |
| }, |
| "dataset": { |
| "seed": 42, |
| "popularity": { |
| "ctr_method": "wall_clock", |
| "bucket_hours": 2, |
| "max_buckets": 1500, |
| "ctr_smoothing": 0.01 |
| }, |
| "_target_": "src.core.data.datasets.mind.MINDDataset", |
| "name": "MIND-small", |
| "version": "small", |
| "embedding_type": "glove", |
| "embedding_size": 300, |
| "random_train_samples": false, |
| "validation_split_strategy": "random", |
| "validation_split_percentage": 0.05, |
| "validation_split_seed": 42, |
| "data_fraction_train": 1.0, |
| "data_fraction_val": 1.0, |
| "data_fraction_test": 1.0, |
| "mode": "train", |
| "word_threshold": 5, |
| "urls": { |
| "small": { |
| "train": "https://huggingface.co/datasets/yjw1029/MIND/resolve/main/MINDsmall_train.zip", |
| "valid": "https://huggingface.co/datasets/yjw1029/MIND/resolve/main/MINDsmall_dev.zip" |
| }, |
| "large": { |
| "train": "https://huggingface.co/datasets/yjw1029/MIND/resolve/main/MINDlarge_train.zip", |
| "valid": "https://huggingface.co/datasets/yjw1029/MIND/resolve/main/MINDlarge_dev.zip", |
| "test": "https://huggingface.co/datasets/yjw1029/MIND/resolve/main/MINDlarge_test.zip" |
| } |
| }, |
| "max_title_length": 30, |
| "max_abstract_length": 50, |
| "max_history_length": 50, |
| "max_impressions_length": 5, |
| "use_knowledge_graph": false, |
| "max_entities": 5, |
| "max_relations": 50, |
| "sampling": { |
| "max_impressions_length": 5, |
| "strategy": "random", |
| "random_seed": 42, |
| "random": { |
| "replace": false |
| }, |
| "topic_diverse": { |
| "min_categories": 3, |
| "category_weights": { |
| "entertainment": 1.0, |
| "sports": 1.0, |
| "technology": 1.0, |
| "business": 1.0, |
| "lifestyle": 1.0, |
| "news": 1.0 |
| } |
| }, |
| "temporal": { |
| "time_window": 24, |
| "recency_weight": 0.7 |
| }, |
| "popularity": { |
| "top_k_ratio": 0.8, |
| "popularity_metric": "clicks" |
| } |
| }, |
| "process_title": true, |
| "process_abstract": false, |
| "process_category": true, |
| "process_subcategory": false, |
| "process_user_id": false, |
| "process_entities": true |
| }, |
| "sampling": { |
| "max_impressions_length": 5, |
| "strategy": "random", |
| "random_seed": 42, |
| "random": { |
| "replace": false |
| }, |
| "topic_diverse": { |
| "min_categories": 3, |
| "category_weights": { |
| "entertainment": 1.0, |
| "sports": 1.0, |
| "technology": 1.0, |
| "business": 1.0, |
| "lifestyle": 1.0, |
| "news": 1.0 |
| } |
| }, |
| "temporal": { |
| "time_window": 24, |
| "recency_weight": 0.7 |
| }, |
| "popularity": { |
| "top_k_ratio": 0.8, |
| "popularity_metric": "clicks" |
| } |
| }, |
| "name": "mind_caum", |
| "model_name": "CAUM", |
| "_output_run_dir": "outputs/train/MIND-small/CAUM/jax/seed_123" |
| }, |
| "initial_validation_metrics": {}, |
| "best_validation_summary": { |
| "epoch_number": 10.0, |
| "train_loss": 1.2371471108698318, |
| "average_metric_value": 0.5206724943439838, |
| "val_loss": 4.4820805537700945, |
| "val_auc": 0.7497600715485296, |
| "val_mrr": 0.39857799672194205, |
| "val_ndcg@5": 0.4387125845284695, |
| "val_ndcg@10": 0.4956393245769941, |
| "val_num_impressions": 7824.0, |
| "timing": { |
| "epoch_training_times": [ |
| 152.7604115009308, |
| 128.3657763004303, |
| 128.71619987487793, |
| 126.83259797096252, |
| 126.7269389629364, |
| 127.33337998390198, |
| 126.36471319198608, |
| 127.29102325439453, |
| 126.6120913028717, |
| 126.97762560844421 |
| ], |
| "epoch_validation_times": [ |
| 156.19692087173462, |
| 156.4893136024475, |
| 156.90509462356567, |
| 155.93345546722412, |
| 156.1876072883606, |
| 154.95236468315125, |
| 157.0661187171936, |
| 156.74578547477722, |
| 156.08843064308167, |
| 156.18353486061096 |
| ], |
| "total_training_time": 2860.9791276454926 |
| } |
| }, |
| "final_test_metrics": { |
| "loss": 4.895912475360167, |
| "auc": 0.6746637875073116, |
| "mrr": 0.3214330456909908, |
| "ndcg@5": 0.35560859266126893, |
| "ndcg@10": 0.4193156331068628, |
| "num_impressions": 72903.0 |
| } |
| } |
