Trained classifier model on MIMIC-IV
Browse files
classification_log_2025-06-05_15-38-15.log
ADDED
|
@@ -0,0 +1,951 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-06-05 15:38:15,054 - INFO - ================================================================================ - [multilabel_classify.py:100:log_section]
|
| 2 |
+
2025-06-05 15:38:15,054 - INFO - = 📌 INITIALIZING TRAINING ENVIRONMENT = - [multilabel_classify.py:101:log_section]
|
| 3 |
+
2025-06-05 15:38:15,054 - INFO - ================================================================================ - [multilabel_classify.py:104:log_section]
|
| 4 |
+
2025-06-05 15:38:15,055 - INFO - 🚀 Setting up data paths and environment variables... - [multilabel_classify.py:3560:main]
|
| 5 |
+
2025-06-05 15:38:15,055 - INFO - 📂 Using output directory: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:3566:main]
|
| 6 |
+
2025-06-05 15:38:15,055 - INFO - 🛠️ Command-line Arguments: - [multilabel_classify.py:368:print_args]
|
| 7 |
+
2025-06-05 15:38:15,055 - INFO -
|
| 8 |
+
🔹 output_dir: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b
|
| 9 |
+
🔹 source_url: XURLs.MIMIC4_DEMO
|
| 10 |
+
🔹 data: mimic4_icd10_full
|
| 11 |
+
🔹 logfile: classification_log
|
| 12 |
+
🔹 base_dir: ../tmp/MIMIC4_DEMO
|
| 13 |
+
🔹 hub_model_id: deb101/mistral-7b-instruct-v0.3-mimic4-adapt
|
| 14 |
+
🔹 model_name: mistralai/Mistral-7B-Instruct-v0.3
|
| 15 |
+
🔹 max_length: 512
|
| 16 |
+
🔹 do_fresh_training: True
|
| 17 |
+
🔹 load_from_checkpoint: True
|
| 18 |
+
🔹 task: multilabel-classify
|
| 19 |
+
🔹 num_train_epochs: 7
|
| 20 |
+
🔹 per_device_train_batch_size: 8
|
| 21 |
+
🔹 per_device_eval_batch_size: 8
|
| 22 |
+
🔹 metric_for_best_model: precision_at_15
|
| 23 |
+
🔹 learning_rate: 0.0001
|
| 24 |
+
🔹 final_lr_scheduling: 1e-06
|
| 25 |
+
🔹 warmup_steps: 50
|
| 26 |
+
🔹 logfile_path: ../tmp/logs/classification_log_2025-06-05_15-38-15.log
|
| 27 |
+
🔹 source: /home/ubuntu/.xcube/data/mimic4_demo - [multilabel_classify.py:369:print_args]
|
| 28 |
+
2025-06-05 15:38:15,055 - INFO - ➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖ - [multilabel_classify.py:370:print_args]
|
| 29 |
+
2025-06-05 15:38:15,055 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 30 |
+
2025-06-05 15:38:15,055 - INFO - + ✨ LOADING DATASETS + - [multilabel_classify.py:101:log_section]
|
| 31 |
+
2025-06-05 15:38:15,055 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:104:log_section]
|
| 32 |
+
2025-06-05 15:38:15,055 - INFO - 📊 Loading main datasets.... - [multilabel_classify.py:3572:main]
|
| 33 |
+
2025-06-05 15:38:23,640 - INFO - 🔍 Total unique labels in dataset: 7942 - [multilabel_classify.py:3353:sample_df_with_full_label_coverage]
|
| 34 |
+
2025-06-05 15:38:23,653 - INFO - 🧪 Attempt 1: Sampled 122 rows covering 863 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 35 |
+
2025-06-05 15:38:23,663 - INFO - 🧪 Attempt 2: Sampled 122 rows covering 816 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 36 |
+
2025-06-05 15:38:23,672 - INFO - 🧪 Attempt 3: Sampled 122 rows covering 885 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 37 |
+
2025-06-05 15:38:23,682 - INFO - 🧪 Attempt 4: Sampled 122 rows covering 828 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 38 |
+
2025-06-05 15:38:23,690 - INFO - 🧪 Attempt 5: Sampled 122 rows covering 879 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 39 |
+
2025-06-05 15:38:23,699 - INFO - 🧪 Attempt 6: Sampled 122 rows covering 852 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 40 |
+
2025-06-05 15:38:23,708 - INFO - 🧪 Attempt 7: Sampled 122 rows covering 838 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 41 |
+
2025-06-05 15:38:23,716 - INFO - 🧪 Attempt 8: Sampled 122 rows covering 851 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 42 |
+
2025-06-05 15:38:23,725 - INFO - 🧪 Attempt 9: Sampled 122 rows covering 825 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 43 |
+
2025-06-05 15:38:23,733 - INFO - 🧪 Attempt 10: Sampled 122 rows covering 833 labels. - [multilabel_classify.py:3367:sample_df_with_full_label_coverage]
|
| 44 |
+
2025-06-05 15:38:23,733 - INFO - ⚠️ Skipping label coverage fix. 7109 labels are missing. - [multilabel_classify.py:3385:sample_df_with_full_label_coverage]
|
| 45 |
+
2025-06-05 15:38:23,734 - INFO - ✅ Final row count: 122 (Valid: 20, Not-valid: 102) - [multilabel_classify.py:3390:sample_df_with_full_label_coverage]
|
| 46 |
+
2025-06-05 15:38:23,752 - INFO - ******************************************************************************** - [multilabel_classify.py:100:log_section]
|
| 47 |
+
2025-06-05 15:38:23,752 - INFO - * 🌟 STARTING MULTI_LABEL CLASSIFICATION MODEL TRAINING * - [multilabel_classify.py:101:log_section]
|
| 48 |
+
2025-06-05 15:38:23,752 - INFO - ******************************************************************************** - [multilabel_classify.py:104:log_section]
|
| 49 |
+
2025-06-05 15:38:23,752 - INFO - 🔐 Loaded authentication token from environment - [multilabel_classify.py:3599:main]
|
| 50 |
+
2025-06-05 15:38:23,753 - INFO - 🏷️ Hub Model ID for this Classification task: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify - [multilabel_classify.py:3603:main]
|
| 51 |
+
2025-06-05 15:38:23,753 - INFO - -------------------------------------------------------------------------------- - [multilabel_classify.py:100:log_section]
|
| 52 |
+
2025-06-05 15:38:23,753 - INFO - - 📋 MODEL EXISTENCE CHECK - - [multilabel_classify.py:101:log_section]
|
| 53 |
+
2025-06-05 15:38:23,753 - INFO - -------------------------------------------------------------------------------- - [multilabel_classify.py:104:log_section]
|
| 54 |
+
2025-06-05 15:38:23,753 - INFO - 🔍 Checking model existence locally and on Hugging Face Hub... - [multilabel_classify.py:3468:check_model_existence]
|
| 55 |
+
2025-06-05 15:38:23,753 - INFO - ❌ Model not found locally at: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:3475:check_model_existence]
|
| 56 |
+
2025-06-05 15:38:23,805 - INFO - ✅ Model exists on Hugging Face Hub with ID: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify - [multilabel_classify.py:3487:check_model_existence]
|
| 57 |
+
2025-06-05 15:38:23,805 - INFO - 📁 Model exists either locally or on Hub - [multilabel_classify.py:3513:check_model_existence]
|
| 58 |
+
2025-06-05 15:38:23,805 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 59 |
+
2025-06-05 15:38:23,805 - INFO - + ✨ STARTING FRESH TRAINING + - [multilabel_classify.py:101:log_section]
|
| 60 |
+
2025-06-05 15:38:23,805 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:104:log_section]
|
| 61 |
+
2025-06-05 15:38:23,805 - INFO - 🔄 Starting fresh training (either forced or model not found)... - [multilabel_classify.py:3616:main]
|
| 62 |
+
2025-06-05 15:38:23,817 - WARNING - Note: Environment variable`HF_TOKEN` is set and is the current active token independently from the token you've just configured. - [_login.py:415:_login]
|
| 63 |
+
2025-06-05 15:38:23,817 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 64 |
+
2025-06-05 15:38:23,817 - INFO - + ✨ LOADING FROM CHECKPOINT + - [multilabel_classify.py:101:log_section]
|
| 65 |
+
2025-06-05 15:38:23,818 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:104:log_section]
|
| 66 |
+
2025-06-05 15:38:23,818 - INFO - 📥 Loading existing model from checkpoint... - [multilabel_classify.py:3623:main]
|
| 67 |
+
2025-06-05 15:38:23,818 - INFO - 🔍 Model exists locally: False - [multilabel_classify.py:2972:load_custom_task_model_and_tokenizer]
|
| 68 |
+
2025-06-05 15:38:23,835 - INFO - ✅ Validated Hugging Face model: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify - [multilabel_classify.py:3523:is_valid_hf_model]
|
| 69 |
+
2025-06-05 15:38:23,835 - INFO - 🌐 Loading model from Hub: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify - [multilabel_classify.py:2983:load_custom_task_model_and_tokenizer]
|
| 70 |
+
2025-06-05 15:38:23,899 - INFO - 🌐 Loaded config from Hub: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify - [multilabel_classify.py:3001:load_custom_task_model_and_tokenizer]
|
| 71 |
+
2025-06-05 15:38:23,899 - INFO - 🔢 Num labels: 833, Base model path: deb101/mistral-7b-instruct-v0.3-mimic4-adapt - [multilabel_classify.py:3013:load_custom_task_model_and_tokenizer]
|
| 72 |
+
2025-06-05 15:38:23,899 - INFO - 🚀 Starting model and tokenizer loading process... - [multilabel_classify.py:1243:load_base_model_and_tokenizer]
|
| 73 |
+
2025-06-05 15:38:23,900 - INFO - 📊 Quantization config: BitsAndBytesConfig {
|
| 74 |
+
"_load_in_4bit": true,
|
| 75 |
+
"_load_in_8bit": false,
|
| 76 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 77 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 78 |
+
"bnb_4bit_quant_type": "nf4",
|
| 79 |
+
"bnb_4bit_use_double_quant": true,
|
| 80 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 81 |
+
"llm_int8_has_fp16_weight": false,
|
| 82 |
+
"llm_int8_skip_modules": null,
|
| 83 |
+
"llm_int8_threshold": 6.0,
|
| 84 |
+
"load_in_4bit": true,
|
| 85 |
+
"load_in_8bit": false,
|
| 86 |
+
"quant_method": "bitsandbytes"
|
| 87 |
+
}
|
| 88 |
+
- [multilabel_classify.py:1252:load_base_model_and_tokenizer]
|
| 89 |
+
2025-06-05 15:38:23,900 - INFO - 🔤 Loading tokenizer for model: deb101/mistral-7b-instruct-v0.3-mimic4-adapt... - [multilabel_classify.py:1256:load_base_model_and_tokenizer]
|
| 90 |
+
2025-06-05 15:38:24,208 - INFO - 🔍 Checking if deb101/mistral-7b-instruct-v0.3-mimic4-adapt is a PEFT model... - [multilabel_classify.py:1268:load_base_model_and_tokenizer]
|
| 91 |
+
2025-06-05 15:38:24,232 - INFO - ✅ Detected PEFT model. Base model: mistralai/Mistral-7B-Instruct-v0.3 - [multilabel_classify.py:1272:load_base_model_and_tokenizer]
|
| 92 |
+
2025-06-05 15:38:24,232 - INFO - 🔍 Loading model configuration for mistralai/Mistral-7B-Instruct-v0.3... - [multilabel_classify.py:1282:load_base_model_and_tokenizer]
|
| 93 |
+
2025-06-05 15:38:24,253 - INFO - Model type: mistral, Architectures: ['MistralForCausalLM'] - [multilabel_classify.py:1288:load_base_model_and_tokenizer]
|
| 94 |
+
2025-06-05 15:38:24,253 - INFO - 🧠 Loading base model: mistralai/Mistral-7B-Instruct-v0.3... - [multilabel_classify.py:1351:load_base_model_and_tokenizer]
|
| 95 |
+
2025-06-05 15:38:24,771 - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk). - [modeling.py:991:get_balanced_memory]
|
| 96 |
+
2025-06-05 15:38:30,024 - INFO - 🧩 Loading PEFT adapters for deb101/mistral-7b-instruct-v0.3-mimic4-adapt... - [multilabel_classify.py:1366:load_base_model_and_tokenizer]
|
| 97 |
+
2025-06-05 15:38:30,887 - INFO - 🔧 Before enabling PEFT adapters for training - [multilabel_classify.py:1368:load_base_model_and_tokenizer]
|
| 98 |
+
2025-06-05 15:38:30,890 - INFO - 📊 trainable params: 0 || all params: 7,254,839,296 || trainable%: 0.0000 - [multilabel_classify.py:159:log_print_output]
|
| 99 |
+
2025-06-05 15:38:30,893 - INFO - 🔧 After Enabling PEFT adapters for training - [multilabel_classify.py:1375:load_base_model_and_tokenizer]
|
| 100 |
+
2025-06-05 15:38:30,895 - INFO - 📊 trainable params: 6,815,744 || all params: 7,254,839,296 || trainable%: 0.0939 - [multilabel_classify.py:159:log_print_output]
|
| 101 |
+
2025-06-05 15:38:30,896 - INFO - ✅ Model and tokenizer successfully loaded! - [multilabel_classify.py:1416:load_base_model_and_tokenizer]
|
| 102 |
+
2025-06-05 15:38:30,896 - INFO - 🤖 Loaded base model from: deb101/mistral-7b-instruct-v0.3-mimic4-adapt - [multilabel_classify.py:3017:load_custom_task_model_and_tokenizer]
|
| 103 |
+
2025-06-05 15:38:30,896 - INFO - Will now start to create Multilabel-Classification Model from the base model - [multilabel_classify.py:560:__init__]
|
| 104 |
+
2025-06-05 15:38:30,899 - INFO - 📊 trainable params: 6,815,744 || all params: 3,765,178,368 || trainable%: 0.1810 - [multilabel_classify.py:614:compute_trainable_params]
|
| 105 |
+
2025-06-05 15:38:31,236 - INFO - Creating the Multi-Label Classification Model from base model mistralai/Mistral-7B-Instruct-v0.3 completed!!! - [multilabel_classify.py:602:__init__]
|
| 106 |
+
2025-06-05 15:38:31,239 - INFO - 📊 trainable params: 84,177,025 || all params: 3,842,539,649 || trainable%: 2.1907 - [multilabel_classify.py:614:compute_trainable_params]
|
| 107 |
+
2025-06-05 15:38:31,239 - INFO - 🏗️ Initialized L2R model with 833 labels - [multilabel_classify.py:3025:load_custom_task_model_and_tokenizer]
|
| 108 |
+
2025-06-05 15:38:31,277 - INFO - 🖥️ Model moved to device: cuda - [multilabel_classify.py:3029:load_custom_task_model_and_tokenizer]
|
| 109 |
+
2025-06-05 15:39:32,232 - INFO - 💾 Loaded custom weights from Hub: /home/ubuntu/.cache/huggingface/hub/models--deb101--mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify/snapshots/fe1ea1ffc7211d1daaa3b0b6eb81b21bed5096a4/model.safetensors - [multilabel_classify.py:3052:load_custom_task_model_and_tokenizer]
|
| 110 |
+
2025-06-05 15:39:33,290 - INFO - ✅ Custom weights loaded into model - [multilabel_classify.py:3071:load_custom_task_model_and_tokenizer]
|
| 111 |
+
2025-06-05 15:39:33,298 - INFO - ✔️ Loaded weights match model's state_dict - [multilabel_classify.py:3082:load_custom_task_model_and_tokenizer]
|
| 112 |
+
2025-06-05 15:39:33,300 - INFO - 📋 Loaded files from Hub (directory: /home/ubuntu/.cache/huggingface/hub/models--deb101--mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify/snapshots/fe1ea1ffc7211d1daaa3b0b6eb81b21bed5096a4):
|
| 113 |
+
+---------+-------------------+------------+
|
| 114 |
+
| Index | Loaded File | Size |
|
| 115 |
+
+=========+===================+============+
|
| 116 |
+
| 1 | model.safetensors | 4267.74 MB |
|
| 117 |
+
+---------+-------------------+------------+
|
| 118 |
+
| 2 | config.json | 0.00 MB |
|
| 119 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:3122:load_custom_task_model_and_tokenizer]
|
| 120 |
+
2025-06-05 15:39:35,109 - INFO - 🖌️ Loaded tokenizer from: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify - [multilabel_classify.py:3137:load_custom_task_model_and_tokenizer]
|
| 121 |
+
2025-06-05 15:39:35,118 - INFO - 🖥️ Model moved to device: cuda - [multilabel_classify.py:3144:load_custom_task_model_and_tokenizer]
|
| 122 |
+
2025-06-05 15:39:35,125 - INFO - ✅ Successfully loaded from checkpoint, will now continue training for 7 epochs. - [multilabel_classify.py:3634:main]
|
| 123 |
+
2025-06-05 15:39:35,127 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 124 |
+
2025-06-05 15:39:35,127 - INFO - + ✨ PREPARING EVALUATION DATA + - [multilabel_classify.py:101:log_section]
|
| 125 |
+
2025-06-05 15:39:35,127 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:104:log_section]
|
| 126 |
+
2025-06-05 15:39:35,127 - INFO - 🔄 Loading and preprocessing evaluation data... - [multilabel_classify.py:3641:main]
|
| 127 |
+
2025-06-05 15:39:35,131 - INFO - Total number of labels: 833 - [multilabel_classify.py:1044:preprocess_data]
|
| 128 |
+
2025-06-05 15:39:35,131 - INFO - Rare labels (freq < 50): 832 - [multilabel_classify.py:1045:preprocess_data]
|
| 129 |
+
2025-06-05 15:39:35,131 - INFO - Not rare labels (freq >= 50): 1 - [multilabel_classify.py:1046:preprocess_data]
|
| 130 |
+
2025-06-05 15:39:35,131 - INFO - Label partitions and classes saved to ../tmp/MIMIC4_DEMO/labels_partition.json - [multilabel_classify.py:1047:preprocess_data]
|
| 131 |
+
2025-06-05 15:39:36,494 - INFO - The size of training set: 567 - [multilabel_classify.py:1143:preprocess_data]
|
| 132 |
+
2025-06-05 15:39:36,494 - INFO - The size of Evaluation set: 136 - [multilabel_classify.py:1144:preprocess_data]
|
| 133 |
+
2025-06-05 15:39:36,500 - INFO - 🏷️ Number of unique ICD-10 codes: 833 - [multilabel_classify.py:3645:main]
|
| 134 |
+
2025-06-05 15:39:36,500 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 135 |
+
2025-06-05 15:39:36,500 - INFO - + ✨ TRAINING PREPARATION + - [multilabel_classify.py:101:log_section]
|
| 136 |
+
2025-06-05 15:39:36,500 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:104:log_section]
|
| 137 |
+
2025-06-05 15:39:36,500 - INFO - ⚙️ Preparing training components and optimizers... - [multilabel_classify.py:3672:main]
|
| 138 |
+
2025-06-05 15:39:36,527 - INFO - 🖥️ Device: NVIDIA GH200 480GB - [multilabel_classify.py:891:log_training_configuration]
|
| 139 |
+
2025-06-05 15:39:36,528 - INFO - 🔋 CUDA Available: True - [multilabel_classify.py:894:log_training_configuration]
|
| 140 |
+
2025-06-05 15:39:36,528 - INFO - 💾 CUDA Device Count: 1 - [multilabel_classify.py:895:log_training_configuration]
|
| 141 |
+
2025-06-05 15:39:36,528 - INFO -
|
| 142 |
+
📋 Training Configuration 📋
|
| 143 |
+
+----------+-----------------------------+------------------------------------------------------------------+
|
| 144 |
+
| 🌟 Emoji | 🏷️ Parameter | 📊 Value |
|
| 145 |
+
+----------+-----------------------------+------------------------------------------------------------------+
|
| 146 |
+
| 📁 | Output Directory | ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b |
|
| 147 |
+
| 🔁 | Training Epochs | 7 |
|
| 148 |
+
| 🏋️ | Train Batch Size | 8 |
|
| 149 |
+
| 🔍 | Eval Batch Size | 8 |
|
| 150 |
+
| 📊 | Gradient Accumulation Steps | 4 |
|
| 151 |
+
| 🚀 | Learning Rate | 0.0001 |
|
| 152 |
+
| 🌅 | Warmup Steps | 50 |
|
| 153 |
+
| 💾 | Save Strategy | epoch |
|
| 154 |
+
| 💾 | Save Total Limit | 10 |
|
| 155 |
+
| 📊 | Evaluation Strategy | epoch |
|
| 156 |
+
| 🎯 | Best Model Metric | precision_at_15 |
|
| 157 |
+
| 📝 | Logging Strategy | steps (every 10 steps) |
|
| 158 |
+
| 🌐 | Push to Hub | True |
|
| 159 |
+
| 🌐 | Hub Model ID | deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify |
|
| 160 |
+
| 🔢 | Steps per Epoch | 17 |
|
| 161 |
+
| 🔢 | Total Training Steps | 119 |
|
| 162 |
+
| 🔢 | Evaluation Steps | 17 |
|
| 163 |
+
| 📊 | Training Dataset Size | 567 samples 🏋️ |
|
| 164 |
+
| 📊 | Evaluation Dataset Size | 136 samples 🔍 |
|
| 165 |
+
+----------+-----------------------------+------------------------------------------------------------------+ - [multilabel_classify.py:883:log_training_args]
|
| 166 |
+
2025-06-05 15:39:36,529 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 167 |
+
2025-06-05 15:39:36,529 - INFO - + ✨ MODEL TRAINING + - [multilabel_classify.py:101:log_section]
|
| 168 |
+
2025-06-05 15:39:36,529 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:104:log_section]
|
| 169 |
+
2025-06-05 15:39:36,529 - INFO - 🏋️ Starting model training process... - [multilabel_classify.py:3694:main]
|
| 170 |
+
2025-06-05 15:39:36,572 - INFO - We are registering the tokenizer deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify in Custom Trainer - [multilabel_classify.py:1988:__init__]
|
| 171 |
+
2025-06-05 15:39:36,834 - INFO - 🚀 Starting Training... - [multilabel_classify.py:1642:on_train_begin]
|
| 172 |
+
2025-06-05 15:39:57,096 - INFO -
|
| 173 |
+
[36m🚂 Training Metrics (Step 10) 🚂
|
| 174 |
+
+---------------+----------+
|
| 175 |
+
| Metric | Value |
|
| 176 |
+
+===============+==========+
|
| 177 |
+
| loss | 0.0989 |
|
| 178 |
+
+---------------+----------+
|
| 179 |
+
| grad_norm | 0.063781 |
|
| 180 |
+
+---------------+----------+
|
| 181 |
+
| learning_rate | 2e-05 |
|
| 182 |
+
+---------------+----------+
|
| 183 |
+
| epoch | 0.56338 |
|
| 184 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 185 |
+
2025-06-05 15:40:11,014 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:2000:evaluate]
|
| 186 |
+
2025-06-05 15:40:45,983 - INFO -
|
| 187 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 188 |
+
+-------------------------------+----------+
|
| 189 |
+
| Metric | Value |
|
| 190 |
+
+===============================+==========+
|
| 191 |
+
| eval_f1_micro | 0 |
|
| 192 |
+
+-------------------------------+----------+
|
| 193 |
+
| eval_f1_macro | 0 |
|
| 194 |
+
+-------------------------------+----------+
|
| 195 |
+
| eval_precision_at_5 | 0.272059 |
|
| 196 |
+
+-------------------------------+----------+
|
| 197 |
+
| eval_recall_at_5 | 0.108443 |
|
| 198 |
+
+-------------------------------+----------+
|
| 199 |
+
| eval_precision_at_8 | 0.195772 |
|
| 200 |
+
+-------------------------------+----------+
|
| 201 |
+
| eval_recall_at_8 | 0.124597 |
|
| 202 |
+
+-------------------------------+----------+
|
| 203 |
+
| eval_precision_at_15 | 0.126961 |
|
| 204 |
+
+-------------------------------+----------+
|
| 205 |
+
| eval_recall_at_15 | 0.146996 |
|
| 206 |
+
+-------------------------------+----------+
|
| 207 |
+
| eval_rare_f1_micro | 0 |
|
| 208 |
+
+-------------------------------+----------+
|
| 209 |
+
| eval_rare_f1_macro | 0 |
|
| 210 |
+
+-------------------------------+----------+
|
| 211 |
+
| eval_rare_precision | 0 |
|
| 212 |
+
+-------------------------------+----------+
|
| 213 |
+
| eval_rare_recall | 0 |
|
| 214 |
+
+-------------------------------+----------+
|
| 215 |
+
| eval_rare_precision_at_5 | 0.208824 |
|
| 216 |
+
+-------------------------------+----------+
|
| 217 |
+
| eval_rare_recall_at_5 | 0.088475 |
|
| 218 |
+
+-------------------------------+----------+
|
| 219 |
+
| eval_rare_precision_at_8 | 0.150735 |
|
| 220 |
+
+-------------------------------+----------+
|
| 221 |
+
| eval_rare_recall_at_8 | 0.101507 |
|
| 222 |
+
+-------------------------------+----------+
|
| 223 |
+
| eval_rare_precision_at_15 | 0.098529 |
|
| 224 |
+
+-------------------------------+----------+
|
| 225 |
+
| eval_rare_recall_at_15 | 0.120102 |
|
| 226 |
+
+-------------------------------+----------+
|
| 227 |
+
| eval_not_rare_f1_micro | 0.595588 |
|
| 228 |
+
+-------------------------------+----------+
|
| 229 |
+
| eval_not_rare_f1_macro | 0.373272 |
|
| 230 |
+
+-------------------------------+----------+
|
| 231 |
+
| eval_not_rare_precision | 0.595588 |
|
| 232 |
+
+-------------------------------+----------+
|
| 233 |
+
| eval_not_rare_recall | 0.595588 |
|
| 234 |
+
+-------------------------------+----------+
|
| 235 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 236 |
+
+-------------------------------+----------+
|
| 237 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 238 |
+
+-------------------------------+----------+
|
| 239 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 240 |
+
+-------------------------------+----------+
|
| 241 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 242 |
+
+-------------------------------+----------+
|
| 243 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 244 |
+
+-------------------------------+----------+
|
| 245 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 246 |
+
+-------------------------------+----------+
|
| 247 |
+
| eval_loss | 0.103884 |
|
| 248 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1855:on_evaluate]
|
| 249 |
+
2025-06-05 15:40:47,180 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-18 - [multilabel_classify.py:2093:_save]
|
| 250 |
+
2025-06-05 15:40:47,181 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-18 - [multilabel_classify.py:2098:_save]
|
| 251 |
+
2025-06-05 15:40:47,182 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-18:
|
| 252 |
+
+---------+-------------------+------------+
|
| 253 |
+
| Index | Saved File | Size |
|
| 254 |
+
+=========+===================+============+
|
| 255 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 256 |
+
+---------+-------------------+------------+
|
| 257 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 258 |
+
+---------+-------------------+------------+
|
| 259 |
+
| 3 | config.json | 0.00 MB |
|
| 260 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 261 |
+
2025-06-05 15:40:53,184 - INFO -
|
| 262 |
+
[36m🚂 Training Metrics (Step 20) 🚂
|
| 263 |
+
+---------------+----------+
|
| 264 |
+
| Metric | Value |
|
| 265 |
+
+===============+==========+
|
| 266 |
+
| loss | 0.1039 |
|
| 267 |
+
+---------------+----------+
|
| 268 |
+
| grad_norm | 0.113664 |
|
| 269 |
+
+---------------+----------+
|
| 270 |
+
| learning_rate | 4e-05 |
|
| 271 |
+
+---------------+----------+
|
| 272 |
+
| epoch | 1.11268 |
|
| 273 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 274 |
+
2025-06-05 15:41:11,244 - INFO -
|
| 275 |
+
[36m🚂 Training Metrics (Step 30) 🚂
|
| 276 |
+
+---------------+----------+
|
| 277 |
+
| Metric | Value |
|
| 278 |
+
+===============+==========+
|
| 279 |
+
| loss | 0.1013 |
|
| 280 |
+
+---------------+----------+
|
| 281 |
+
| grad_norm | 0.198499 |
|
| 282 |
+
+---------------+----------+
|
| 283 |
+
| learning_rate | 6e-05 |
|
| 284 |
+
+---------------+----------+
|
| 285 |
+
| epoch | 1.67606 |
|
| 286 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 287 |
+
2025-06-05 15:41:21,600 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:2000:evaluate]
|
| 288 |
+
2025-06-05 15:41:56,730 - INFO -
|
| 289 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 290 |
+
+-------------------------------+----------+
|
| 291 |
+
| Metric | Value |
|
| 292 |
+
+===============================+==========+
|
| 293 |
+
| eval_f1_micro | 0 |
|
| 294 |
+
+-------------------------------+----------+
|
| 295 |
+
| eval_f1_macro | 0 |
|
| 296 |
+
+-------------------------------+----------+
|
| 297 |
+
| eval_precision_at_5 | 0.270588 |
|
| 298 |
+
+-------------------------------+----------+
|
| 299 |
+
| eval_recall_at_5 | 0.111741 |
|
| 300 |
+
+-------------------------------+----------+
|
| 301 |
+
| eval_precision_at_8 | 0.230699 |
|
| 302 |
+
+-------------------------------+----------+
|
| 303 |
+
| eval_recall_at_8 | 0.153039 |
|
| 304 |
+
+-------------------------------+----------+
|
| 305 |
+
| eval_precision_at_15 | 0.148529 |
|
| 306 |
+
+-------------------------------+----------+
|
| 307 |
+
| eval_recall_at_15 | 0.172471 |
|
| 308 |
+
+-------------------------------+----------+
|
| 309 |
+
| eval_rare_f1_micro | 0 |
|
| 310 |
+
+-------------------------------+----------+
|
| 311 |
+
| eval_rare_f1_macro | 0 |
|
| 312 |
+
+-------------------------------+----------+
|
| 313 |
+
| eval_rare_precision | 0 |
|
| 314 |
+
+-------------------------------+----------+
|
| 315 |
+
| eval_rare_recall | 0 |
|
| 316 |
+
+-------------------------------+----------+
|
| 317 |
+
| eval_rare_precision_at_5 | 0.227941 |
|
| 318 |
+
+-------------------------------+----------+
|
| 319 |
+
| eval_rare_recall_at_5 | 0.091875 |
|
| 320 |
+
+-------------------------------+----------+
|
| 321 |
+
| eval_rare_precision_at_8 | 0.191176 |
|
| 322 |
+
+-------------------------------+----------+
|
| 323 |
+
| eval_rare_recall_at_8 | 0.134186 |
|
| 324 |
+
+-------------------------------+----------+
|
| 325 |
+
| eval_rare_precision_at_15 | 0.126961 |
|
| 326 |
+
+-------------------------------+----------+
|
| 327 |
+
| eval_rare_recall_at_15 | 0.150628 |
|
| 328 |
+
+-------------------------------+----------+
|
| 329 |
+
| eval_not_rare_f1_micro | 0.595588 |
|
| 330 |
+
+-------------------------------+----------+
|
| 331 |
+
| eval_not_rare_f1_macro | 0.373272 |
|
| 332 |
+
+-------------------------------+----------+
|
| 333 |
+
| eval_not_rare_precision | 0.595588 |
|
| 334 |
+
+-------------------------------+----------+
|
| 335 |
+
| eval_not_rare_recall | 0.595588 |
|
| 336 |
+
+-------------------------------+----------+
|
| 337 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 338 |
+
+-------------------------------+----------+
|
| 339 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 340 |
+
+-------------------------------+----------+
|
| 341 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 342 |
+
+-------------------------------+----------+
|
| 343 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 344 |
+
+-------------------------------+----------+
|
| 345 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 346 |
+
+-------------------------------+----------+
|
| 347 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 348 |
+
+-------------------------------+----------+
|
| 349 |
+
| eval_loss | 0.105386 |
|
| 350 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1855:on_evaluate]
|
| 351 |
+
2025-06-05 15:41:57,912 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-36 - [multilabel_classify.py:2093:_save]
|
| 352 |
+
2025-06-05 15:41:57,913 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-36 - [multilabel_classify.py:2098:_save]
|
| 353 |
+
2025-06-05 15:41:57,914 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-36:
|
| 354 |
+
+---------+-------------------+------------+
|
| 355 |
+
| Index | Saved File | Size |
|
| 356 |
+
+=========+===================+============+
|
| 357 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 358 |
+
+---------+-------------------+------------+
|
| 359 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 360 |
+
+---------+-------------------+------------+
|
| 361 |
+
| 3 | config.json | 0.00 MB |
|
| 362 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 363 |
+
2025-06-05 15:42:07,554 - INFO -
|
| 364 |
+
[36m🚂 Training Metrics (Step 40) 🚂
|
| 365 |
+
+---------------+----------+
|
| 366 |
+
| Metric | Value |
|
| 367 |
+
+===============+==========+
|
| 368 |
+
| loss | 0.0971 |
|
| 369 |
+
+---------------+----------+
|
| 370 |
+
| grad_norm | 0.073463 |
|
| 371 |
+
+---------------+----------+
|
| 372 |
+
| learning_rate | 8e-05 |
|
| 373 |
+
+---------------+----------+
|
| 374 |
+
| epoch | 2.22535 |
|
| 375 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 376 |
+
2025-06-05 15:42:25,650 - INFO -
|
| 377 |
+
[36m🚂 Training Metrics (Step 50) 🚂
|
| 378 |
+
+---------------+----------+
|
| 379 |
+
| Metric | Value |
|
| 380 |
+
+===============+==========+
|
| 381 |
+
| loss | 0.1012 |
|
| 382 |
+
+---------------+----------+
|
| 383 |
+
| grad_norm | 0.058595 |
|
| 384 |
+
+---------------+----------+
|
| 385 |
+
| learning_rate | 0.0001 |
|
| 386 |
+
+---------------+----------+
|
| 387 |
+
| epoch | 2.78873 |
|
| 388 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 389 |
+
2025-06-05 15:42:32,400 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:2000:evaluate]
|
| 390 |
+
2025-06-05 15:43:07,380 - INFO -
|
| 391 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 392 |
+
+-------------------------------+----------+
|
| 393 |
+
| Metric | Value |
|
| 394 |
+
+===============================+==========+
|
| 395 |
+
| eval_f1_micro | 0 |
|
| 396 |
+
+-------------------------------+----------+
|
| 397 |
+
| eval_f1_macro | 0 |
|
| 398 |
+
+-------------------------------+----------+
|
| 399 |
+
| eval_precision_at_5 | 0.276471 |
|
| 400 |
+
+-------------------------------+----------+
|
| 401 |
+
| eval_recall_at_5 | 0.111579 |
|
| 402 |
+
+-------------------------------+----------+
|
| 403 |
+
| eval_precision_at_8 | 0.237132 |
|
| 404 |
+
+-------------------------------+----------+
|
| 405 |
+
| eval_recall_at_8 | 0.15122 |
|
| 406 |
+
+-------------------------------+----------+
|
| 407 |
+
| eval_precision_at_15 | 0.148039 |
|
| 408 |
+
+-------------------------------+----------+
|
| 409 |
+
| eval_recall_at_15 | 0.171528 |
|
| 410 |
+
+-------------------------------+----------+
|
| 411 |
+
| eval_rare_f1_micro | 0 |
|
| 412 |
+
+-------------------------------+----------+
|
| 413 |
+
| eval_rare_f1_macro | 0 |
|
| 414 |
+
+-------------------------------+----------+
|
| 415 |
+
| eval_rare_precision | 0 |
|
| 416 |
+
+-------------------------------+----------+
|
| 417 |
+
| eval_rare_recall | 0 |
|
| 418 |
+
+-------------------------------+----------+
|
| 419 |
+
| eval_rare_precision_at_5 | 0.247059 |
|
| 420 |
+
+-------------------------------+----------+
|
| 421 |
+
| eval_rare_recall_at_5 | 0.105122 |
|
| 422 |
+
+-------------------------------+----------+
|
| 423 |
+
| eval_rare_precision_at_8 | 0.193015 |
|
| 424 |
+
+-------------------------------+----------+
|
| 425 |
+
| eval_rare_recall_at_8 | 0.135358 |
|
| 426 |
+
+-------------------------------+----------+
|
| 427 |
+
| eval_rare_precision_at_15 | 0.125 |
|
| 428 |
+
+-------------------------------+----------+
|
| 429 |
+
| eval_rare_recall_at_15 | 0.149296 |
|
| 430 |
+
+-------------------------------+----------+
|
| 431 |
+
| eval_not_rare_f1_micro | 0.595588 |
|
| 432 |
+
+-------------------------------+----------+
|
| 433 |
+
| eval_not_rare_f1_macro | 0.373272 |
|
| 434 |
+
+-------------------------------+----------+
|
| 435 |
+
| eval_not_rare_precision | 0.595588 |
|
| 436 |
+
+-------------------------------+----------+
|
| 437 |
+
| eval_not_rare_recall | 0.595588 |
|
| 438 |
+
+-------------------------------+----------+
|
| 439 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 440 |
+
+-------------------------------+----------+
|
| 441 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 442 |
+
+-------------------------------+----------+
|
| 443 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 444 |
+
+-------------------------------+----------+
|
| 445 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 446 |
+
+-------------------------------+----------+
|
| 447 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 448 |
+
+-------------------------------+----------+
|
| 449 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 450 |
+
+-------------------------------+----------+
|
| 451 |
+
| eval_loss | 0.103298 |
|
| 452 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1855:on_evaluate]
|
| 453 |
+
2025-06-05 15:43:08,572 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-54 - [multilabel_classify.py:2093:_save]
|
| 454 |
+
2025-06-05 15:43:08,574 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-54 - [multilabel_classify.py:2098:_save]
|
| 455 |
+
2025-06-05 15:43:08,575 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-54:
|
| 456 |
+
+---------+-------------------+------------+
|
| 457 |
+
| Index | Saved File | Size |
|
| 458 |
+
+=========+===================+============+
|
| 459 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 460 |
+
+---------+-------------------+------------+
|
| 461 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 462 |
+
+---------+-------------------+------------+
|
| 463 |
+
| 3 | config.json | 0.00 MB |
|
| 464 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 465 |
+
2025-06-05 15:43:21,824 - INFO -
|
| 466 |
+
[36m🚂 Training Metrics (Step 60) 🚂
|
| 467 |
+
+---------------+----------+
|
| 468 |
+
| Metric | Value |
|
| 469 |
+
+===============+==========+
|
| 470 |
+
| loss | 0.0995 |
|
| 471 |
+
+---------------+----------+
|
| 472 |
+
| grad_norm | 0.103198 |
|
| 473 |
+
+---------------+----------+
|
| 474 |
+
| learning_rate | 9.5e-05 |
|
| 475 |
+
+---------------+----------+
|
| 476 |
+
| epoch | 3.33803 |
|
| 477 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 478 |
+
2025-06-05 15:43:39,928 - INFO -
|
| 479 |
+
[36m🚂 Training Metrics (Step 70) 🚂
|
| 480 |
+
+---------------+----------+
|
| 481 |
+
| Metric | Value |
|
| 482 |
+
+===============+==========+
|
| 483 |
+
| loss | 0.1003 |
|
| 484 |
+
+---------------+----------+
|
| 485 |
+
| grad_norm | 0.184965 |
|
| 486 |
+
+---------------+----------+
|
| 487 |
+
| learning_rate | 8.1e-05 |
|
| 488 |
+
+---------------+----------+
|
| 489 |
+
| epoch | 3.90141 |
|
| 490 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 491 |
+
2025-06-05 15:43:43,080 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:2000:evaluate]
|
| 492 |
+
2025-06-05 15:44:18,190 - INFO -
|
| 493 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 494 |
+
+-------------------------------+----------+
|
| 495 |
+
| Metric | Value |
|
| 496 |
+
+===============================+==========+
|
| 497 |
+
| eval_f1_micro | 0 |
|
| 498 |
+
+-------------------------------+----------+
|
| 499 |
+
| eval_f1_macro | 0 |
|
| 500 |
+
+-------------------------------+----------+
|
| 501 |
+
| eval_precision_at_5 | 0.276471 |
|
| 502 |
+
+-------------------------------+----------+
|
| 503 |
+
| eval_recall_at_5 | 0.116576 |
|
| 504 |
+
+-------------------------------+----------+
|
| 505 |
+
| eval_precision_at_8 | 0.235294 |
|
| 506 |
+
+-------------------------------+----------+
|
| 507 |
+
| eval_recall_at_8 | 0.144086 |
|
| 508 |
+
+-------------------------------+----------+
|
| 509 |
+
| eval_precision_at_15 | 0.14951 |
|
| 510 |
+
+-------------------------------+----------+
|
| 511 |
+
| eval_recall_at_15 | 0.172205 |
|
| 512 |
+
+-------------------------------+----------+
|
| 513 |
+
| eval_rare_f1_micro | 0 |
|
| 514 |
+
+-------------------------------+----------+
|
| 515 |
+
| eval_rare_f1_macro | 0 |
|
| 516 |
+
+-------------------------------+----------+
|
| 517 |
+
| eval_rare_precision | 0 |
|
| 518 |
+
+-------------------------------+----------+
|
| 519 |
+
| eval_rare_recall | 0 |
|
| 520 |
+
+-------------------------------+----------+
|
| 521 |
+
| eval_rare_precision_at_5 | 0.245588 |
|
| 522 |
+
+-------------------------------+----------+
|
| 523 |
+
| eval_rare_recall_at_5 | 0.107489 |
|
| 524 |
+
+-------------------------------+----------+
|
| 525 |
+
| eval_rare_precision_at_8 | 0.193934 |
|
| 526 |
+
+-------------------------------+----------+
|
| 527 |
+
| eval_rare_recall_at_8 | 0.136584 |
|
| 528 |
+
+-------------------------------+----------+
|
| 529 |
+
| eval_rare_precision_at_15 | 0.125 |
|
| 530 |
+
+-------------------------------+----------+
|
| 531 |
+
| eval_rare_recall_at_15 | 0.149146 |
|
| 532 |
+
+-------------------------------+----------+
|
| 533 |
+
| eval_not_rare_f1_micro | 0.595588 |
|
| 534 |
+
+-------------------------------+----------+
|
| 535 |
+
| eval_not_rare_f1_macro | 0.373272 |
|
| 536 |
+
+-------------------------------+----------+
|
| 537 |
+
| eval_not_rare_precision | 0.595588 |
|
| 538 |
+
+-------------------------------+----------+
|
| 539 |
+
| eval_not_rare_recall | 0.595588 |
|
| 540 |
+
+-------------------------------+----------+
|
| 541 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 542 |
+
+-------------------------------+----------+
|
| 543 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 544 |
+
+-------------------------------+----------+
|
| 545 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 546 |
+
+-------------------------------+----------+
|
| 547 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 548 |
+
+-------------------------------+----------+
|
| 549 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 550 |
+
+-------------------------------+----------+
|
| 551 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 552 |
+
+-------------------------------+----------+
|
| 553 |
+
| eval_loss | 0.101713 |
|
| 554 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1855:on_evaluate]
|
| 555 |
+
2025-06-05 15:44:19,386 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-72 - [multilabel_classify.py:2093:_save]
|
| 556 |
+
2025-06-05 15:44:19,387 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-72 - [multilabel_classify.py:2098:_save]
|
| 557 |
+
2025-06-05 15:44:19,388 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-72:
|
| 558 |
+
+---------+-------------------+------------+
|
| 559 |
+
| Index | Saved File | Size |
|
| 560 |
+
+=========+===================+============+
|
| 561 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 562 |
+
+---------+-------------------+------------+
|
| 563 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 564 |
+
+---------+-------------------+------------+
|
| 565 |
+
| 3 | config.json | 0.00 MB |
|
| 566 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 567 |
+
2025-06-05 15:44:36,268 - INFO -
|
| 568 |
+
[36m🚂 Training Metrics (Step 80) 🚂
|
| 569 |
+
+---------------+---------+
|
| 570 |
+
| Metric | Value |
|
| 571 |
+
+===============+=========+
|
| 572 |
+
| loss | 0.0941 |
|
| 573 |
+
+---------------+---------+
|
| 574 |
+
| grad_norm | 0.19297 |
|
| 575 |
+
+---------------+---------+
|
| 576 |
+
| learning_rate | 6.1e-05 |
|
| 577 |
+
+---------------+---------+
|
| 578 |
+
| epoch | 4.4507 |
|
| 579 |
+
+---------------+---------+[0m - [multilabel_classify.py:1836:on_log]
|
| 580 |
+
2025-06-05 15:44:53,903 - INFO -
|
| 581 |
+
[36m🚂 Training Metrics (Step 90) 🚂
|
| 582 |
+
+---------------+----------+
|
| 583 |
+
| Metric | Value |
|
| 584 |
+
+===============+==========+
|
| 585 |
+
| loss | 0.0872 |
|
| 586 |
+
+---------------+----------+
|
| 587 |
+
| grad_norm | 0.214059 |
|
| 588 |
+
+---------------+----------+
|
| 589 |
+
| learning_rate | 3.8e-05 |
|
| 590 |
+
+---------------+----------+
|
| 591 |
+
| epoch | 5 |
|
| 592 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 593 |
+
2025-06-05 15:44:53,903 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:2000:evaluate]
|
| 594 |
+
2025-06-05 15:45:28,817 - INFO -
|
| 595 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 596 |
+
+-------------------------------+----------+
|
| 597 |
+
| Metric | Value |
|
| 598 |
+
+===============================+==========+
|
| 599 |
+
| eval_f1_micro | 0.0238 |
|
| 600 |
+
+-------------------------------+----------+
|
| 601 |
+
| eval_f1_macro | 0.00104 |
|
| 602 |
+
+-------------------------------+----------+
|
| 603 |
+
| eval_precision_at_5 | 0.276471 |
|
| 604 |
+
+-------------------------------+----------+
|
| 605 |
+
| eval_recall_at_5 | 0.116576 |
|
| 606 |
+
+-------------------------------+----------+
|
| 607 |
+
| eval_precision_at_8 | 0.235294 |
|
| 608 |
+
+-------------------------------+----------+
|
| 609 |
+
| eval_recall_at_8 | 0.144086 |
|
| 610 |
+
+-------------------------------+----------+
|
| 611 |
+
| eval_precision_at_15 | 0.153431 |
|
| 612 |
+
+-------------------------------+----------+
|
| 613 |
+
| eval_recall_at_15 | 0.174813 |
|
| 614 |
+
+-------------------------------+----------+
|
| 615 |
+
| eval_rare_f1_micro | 0.003331 |
|
| 616 |
+
+-------------------------------+----------+
|
| 617 |
+
| eval_rare_f1_macro | 0.000516 |
|
| 618 |
+
+-------------------------------+----------+
|
| 619 |
+
| eval_rare_precision | 0.4 |
|
| 620 |
+
+-------------------------------+----------+
|
| 621 |
+
| eval_rare_recall | 0.001672 |
|
| 622 |
+
+-------------------------------+----------+
|
| 623 |
+
| eval_rare_precision_at_5 | 0.245588 |
|
| 624 |
+
+-------------------------------+----------+
|
| 625 |
+
| eval_rare_recall_at_5 | 0.107489 |
|
| 626 |
+
+-------------------------------+----------+
|
| 627 |
+
| eval_rare_precision_at_8 | 0.200368 |
|
| 628 |
+
+-------------------------------+----------+
|
| 629 |
+
| eval_rare_recall_at_8 | 0.12225 |
|
| 630 |
+
+-------------------------------+----------+
|
| 631 |
+
| eval_rare_precision_at_15 | 0.126471 |
|
| 632 |
+
+-------------------------------+----------+
|
| 633 |
+
| eval_rare_recall_at_15 | 0.149956 |
|
| 634 |
+
+-------------------------------+----------+
|
| 635 |
+
| eval_not_rare_f1_micro | 0.507353 |
|
| 636 |
+
+-------------------------------+----------+
|
| 637 |
+
| eval_not_rare_f1_macro | 0.499533 |
|
| 638 |
+
+-------------------------------+----------+
|
| 639 |
+
| eval_not_rare_precision | 0.507353 |
|
| 640 |
+
+-------------------------------+----------+
|
| 641 |
+
| eval_not_rare_recall | 0.507353 |
|
| 642 |
+
+-------------------------------+----------+
|
| 643 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 644 |
+
+-------------------------------+----------+
|
| 645 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 646 |
+
+-------------------------------+----------+
|
| 647 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 648 |
+
+-------------------------------+----------+
|
| 649 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 650 |
+
+-------------------------------+----------+
|
| 651 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 652 |
+
+-------------------------------+----------+
|
| 653 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 654 |
+
+-------------------------------+----------+
|
| 655 |
+
| eval_loss | 0.104214 |
|
| 656 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1855:on_evaluate]
|
| 657 |
+
2025-06-05 15:45:30,011 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-90 - [multilabel_classify.py:2093:_save]
|
| 658 |
+
2025-06-05 15:45:30,012 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-90 - [multilabel_classify.py:2098:_save]
|
| 659 |
+
2025-06-05 15:45:30,013 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-90:
|
| 660 |
+
+---------+-------------------+------------+
|
| 661 |
+
| Index | Saved File | Size |
|
| 662 |
+
+=========+===================+============+
|
| 663 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 664 |
+
+---------+-------------------+------------+
|
| 665 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 666 |
+
+---------+-------------------+------------+
|
| 667 |
+
| 3 | config.json | 0.00 MB |
|
| 668 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 669 |
+
2025-06-05 15:45:50,487 - INFO -
|
| 670 |
+
[36m🚂 Training Metrics (Step 100) 🚂
|
| 671 |
+
+---------------+----------+
|
| 672 |
+
| Metric | Value |
|
| 673 |
+
+===============+==========+
|
| 674 |
+
| loss | 0.0891 |
|
| 675 |
+
+---------------+----------+
|
| 676 |
+
| grad_norm | 0.162411 |
|
| 677 |
+
+---------------+----------+
|
| 678 |
+
| learning_rate | 1.8e-05 |
|
| 679 |
+
+---------------+----------+
|
| 680 |
+
| epoch | 5.56338 |
|
| 681 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 682 |
+
2025-06-05 15:46:04,470 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:2000:evaluate]
|
| 683 |
+
2025-06-05 15:46:39,381 - INFO -
|
| 684 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 685 |
+
+-------------------------------+----------+
|
| 686 |
+
| Metric | Value |
|
| 687 |
+
+===============================+==========+
|
| 688 |
+
| eval_f1_micro | 0.041747 |
|
| 689 |
+
+-------------------------------+----------+
|
| 690 |
+
| eval_f1_macro | 0.001271 |
|
| 691 |
+
+-------------------------------+----------+
|
| 692 |
+
| eval_precision_at_5 | 0.276471 |
|
| 693 |
+
+-------------------------------+----------+
|
| 694 |
+
| eval_recall_at_5 | 0.116576 |
|
| 695 |
+
+-------------------------------+----------+
|
| 696 |
+
| eval_precision_at_8 | 0.235294 |
|
| 697 |
+
+-------------------------------+----------+
|
| 698 |
+
| eval_recall_at_8 | 0.144086 |
|
| 699 |
+
+-------------------------------+----------+
|
| 700 |
+
| eval_precision_at_15 | 0.153431 |
|
| 701 |
+
+-------------------------------+----------+
|
| 702 |
+
| eval_recall_at_15 | 0.174813 |
|
| 703 |
+
+-------------------------------+----------+
|
| 704 |
+
| eval_rare_f1_micro | 0.013093 |
|
| 705 |
+
+-------------------------------+----------+
|
| 706 |
+
| eval_rare_f1_macro | 0.000634 |
|
| 707 |
+
+-------------------------------+----------+
|
| 708 |
+
| eval_rare_precision | 0.307692 |
|
| 709 |
+
+-------------------------------+----------+
|
| 710 |
+
| eval_rare_recall | 0.006689 |
|
| 711 |
+
+-------------------------------+----------+
|
| 712 |
+
| eval_rare_precision_at_5 | 0.236765 |
|
| 713 |
+
+-------------------------------+----------+
|
| 714 |
+
| eval_rare_recall_at_5 | 0.101155 |
|
| 715 |
+
+-------------------------------+----------+
|
| 716 |
+
| eval_rare_precision_at_8 | 0.200368 |
|
| 717 |
+
+-------------------------------+----------+
|
| 718 |
+
| eval_rare_recall_at_8 | 0.12225 |
|
| 719 |
+
+-------------------------------+----------+
|
| 720 |
+
| eval_rare_precision_at_15 | 0.126471 |
|
| 721 |
+
+-------------------------------+----------+
|
| 722 |
+
| eval_rare_recall_at_15 | 0.149956 |
|
| 723 |
+
+-------------------------------+----------+
|
| 724 |
+
| eval_not_rare_f1_micro | 0.507353 |
|
| 725 |
+
+-------------------------------+----------+
|
| 726 |
+
| eval_not_rare_f1_macro | 0.506044 |
|
| 727 |
+
+-------------------------------+----------+
|
| 728 |
+
| eval_not_rare_precision | 0.507353 |
|
| 729 |
+
+-------------------------------+----------+
|
| 730 |
+
| eval_not_rare_recall | 0.507353 |
|
| 731 |
+
+-------------------------------+----------+
|
| 732 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 733 |
+
+-------------------------------+----------+
|
| 734 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 735 |
+
+-------------------------------+----------+
|
| 736 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 737 |
+
+-------------------------------+----------+
|
| 738 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 739 |
+
+-------------------------------+----------+
|
| 740 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 741 |
+
+-------------------------------+----------+
|
| 742 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 743 |
+
+-------------------------------+----------+
|
| 744 |
+
| eval_loss | 0.101633 |
|
| 745 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1855:on_evaluate]
|
| 746 |
+
2025-06-05 15:46:40,584 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-108 - [multilabel_classify.py:2093:_save]
|
| 747 |
+
2025-06-05 15:46:40,585 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-108 - [multilabel_classify.py:2098:_save]
|
| 748 |
+
2025-06-05 15:46:40,586 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-108:
|
| 749 |
+
+---------+-------------------+------------+
|
| 750 |
+
| Index | Saved File | Size |
|
| 751 |
+
+=========+===================+============+
|
| 752 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 753 |
+
+---------+-------------------+------------+
|
| 754 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 755 |
+
+---------+-------------------+------------+
|
| 756 |
+
| 3 | config.json | 0.00 MB |
|
| 757 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 758 |
+
2025-06-05 15:46:46,602 - INFO -
|
| 759 |
+
[36m🚂 Training Metrics (Step 110) 🚂
|
| 760 |
+
+---------------+----------+
|
| 761 |
+
| Metric | Value |
|
| 762 |
+
+===============+==========+
|
| 763 |
+
| loss | 0.0844 |
|
| 764 |
+
+---------------+----------+
|
| 765 |
+
| grad_norm | 0.075479 |
|
| 766 |
+
+---------------+----------+
|
| 767 |
+
| learning_rate | 5e-06 |
|
| 768 |
+
+---------------+----------+
|
| 769 |
+
| epoch | 6.11268 |
|
| 770 |
+
+---------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 771 |
+
2025-06-05 15:47:04,125 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-119 - [multilabel_classify.py:2093:_save]
|
| 772 |
+
2025-06-05 15:47:04,126 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-119 - [multilabel_classify.py:2098:_save]
|
| 773 |
+
2025-06-05 15:47:04,127 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-119:
|
| 774 |
+
+---------+-------------------+------------+
|
| 775 |
+
| Index | Saved File | Size |
|
| 776 |
+
+=========+===================+============+
|
| 777 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 778 |
+
+---------+-------------------+------------+
|
| 779 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 780 |
+
+---------+-------------------+------------+
|
| 781 |
+
| 3 | config.json | 0.00 MB |
|
| 782 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 783 |
+
2025-06-05 15:47:04,430 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:2000:evaluate]
|
| 784 |
+
2025-06-05 15:47:39,440 - INFO -
|
| 785 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 786 |
+
+-------------------------------+----------+
|
| 787 |
+
| Metric | Value |
|
| 788 |
+
+===============================+==========+
|
| 789 |
+
| eval_f1_micro | 0.037369 |
|
| 790 |
+
+-------------------------------+----------+
|
| 791 |
+
| eval_f1_macro | 0.001265 |
|
| 792 |
+
+-------------------------------+----------+
|
| 793 |
+
| eval_precision_at_5 | 0.276471 |
|
| 794 |
+
+-------------------------------+----------+
|
| 795 |
+
| eval_recall_at_5 | 0.116576 |
|
| 796 |
+
+-------------------------------+----------+
|
| 797 |
+
| eval_precision_at_8 | 0.235294 |
|
| 798 |
+
+-------------------------------+----------+
|
| 799 |
+
| eval_recall_at_8 | 0.144086 |
|
| 800 |
+
+-------------------------------+----------+
|
| 801 |
+
| eval_precision_at_15 | 0.153431 |
|
| 802 |
+
+-------------------------------+----------+
|
| 803 |
+
| eval_recall_at_15 | 0.174813 |
|
| 804 |
+
+-------------------------------+----------+
|
| 805 |
+
| eval_rare_f1_micro | 0.011508 |
|
| 806 |
+
+-------------------------------+----------+
|
| 807 |
+
| eval_rare_f1_macro | 0.000666 |
|
| 808 |
+
+-------------------------------+----------+
|
| 809 |
+
| eval_rare_precision | 0.341463 |
|
| 810 |
+
+-------------------------------+----------+
|
| 811 |
+
| eval_rare_recall | 0.005853 |
|
| 812 |
+
+-------------------------------+----------+
|
| 813 |
+
| eval_rare_precision_at_5 | 0.232353 |
|
| 814 |
+
+-------------------------------+----------+
|
| 815 |
+
| eval_rare_recall_at_5 | 0.098451 |
|
| 816 |
+
+-------------------------------+----------+
|
| 817 |
+
| eval_rare_precision_at_8 | 0.200368 |
|
| 818 |
+
+-------------------------------+----------+
|
| 819 |
+
| eval_rare_recall_at_8 | 0.12225 |
|
| 820 |
+
+-------------------------------+----------+
|
| 821 |
+
| eval_rare_precision_at_15 | 0.126471 |
|
| 822 |
+
+-------------------------------+----------+
|
| 823 |
+
| eval_rare_recall_at_15 | 0.149956 |
|
| 824 |
+
+-------------------------------+----------+
|
| 825 |
+
| eval_not_rare_f1_micro | 0.5 |
|
| 826 |
+
+-------------------------------+----------+
|
| 827 |
+
| eval_not_rare_f1_macro | 0.5 |
|
| 828 |
+
+-------------------------------+----------+
|
| 829 |
+
| eval_not_rare_precision | 0.5 |
|
| 830 |
+
+-------------------------------+----------+
|
| 831 |
+
| eval_not_rare_recall | 0.5 |
|
| 832 |
+
+-------------------------------+----------+
|
| 833 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 834 |
+
+-------------------------------+----------+
|
| 835 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 836 |
+
+-------------------------------+----------+
|
| 837 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 838 |
+
+-------------------------------+----------+
|
| 839 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 840 |
+
+-------------------------------+----------+
|
| 841 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 842 |
+
+-------------------------------+----------+
|
| 843 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 844 |
+
+-------------------------------+----------+
|
| 845 |
+
| eval_loss | 0.101503 |
|
| 846 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1855:on_evaluate]
|
| 847 |
+
2025-06-05 15:47:43,196 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-119 - [multilabel_classify.py:2093:_save]
|
| 848 |
+
2025-06-05 15:47:43,198 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-119 - [multilabel_classify.py:2098:_save]
|
| 849 |
+
2025-06-05 15:47:43,199 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-119:
|
| 850 |
+
+---------+--------------------+------------+
|
| 851 |
+
| Index | Saved File | Size |
|
| 852 |
+
+=========+====================+============+
|
| 853 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 854 |
+
+---------+--------------------+------------+
|
| 855 |
+
| 2 | optimizer.pt | 642.30 MB |
|
| 856 |
+
+---------+--------------------+------------+
|
| 857 |
+
| 3 | model.safetensors | 4267.74 MB |
|
| 858 |
+
+---------+--------------------+------------+
|
| 859 |
+
| 4 | scaler.pt | 0.00 MB |
|
| 860 |
+
+---------+--------------------+------------+
|
| 861 |
+
| 5 | config.json | 0.00 MB |
|
| 862 |
+
+---------+--------------------+------------+
|
| 863 |
+
| 6 | scheduler.pt | 0.00 MB |
|
| 864 |
+
+---------+--------------------+------------+
|
| 865 |
+
| 7 | trainer_state.json | 0.01 MB |
|
| 866 |
+
+---------+--------------------+------------+
|
| 867 |
+
| 8 | rng_state.pth | 0.01 MB |
|
| 868 |
+
+---------+--------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 869 |
+
2025-06-05 15:47:43,564 - INFO - 📂 Loading best model from ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-90 - [multilabel_classify.py:2167:_load_best_model]
|
| 870 |
+
2025-06-05 15:47:43,565 - INFO - 🖥️ Model is on device: cuda:0 - [multilabel_classify.py:2177:_load_best_model]
|
| 871 |
+
2025-06-05 15:47:43,617 - INFO - 🔑 Key order comparison:
|
| 872 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 873 |
+
| Index | Saved state_dict Keys | Model state_dict Keys |
|
| 874 |
+
+=========+============================================+======================================================================================+
|
| 875 |
+
| 1 | attention.in_proj_bias | boost_mul |
|
| 876 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 877 |
+
| 2 | attention.in_proj_weight | boost_add |
|
| 878 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 879 |
+
| 3 | attention.out_proj.bias | base_model.base_model.model.model.embed_tokens.weight |
|
| 880 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 881 |
+
| 4 | attention.out_proj.weight | base_model.base_model.model.model.layers.0.self_attn.q_proj.base_layer.weight |
|
| 882 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 883 |
+
| 5 | base_model.base_model.model.lm_head.weight | base_model.base_model.model.model.layers.0.self_attn.q_proj.base_layer.weight.absmax |
|
| 884 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+ - [multilabel_classify.py:2201:_load_best_model]
|
| 885 |
+
2025-06-05 15:47:44,629 - INFO - ✅ Loaded best model weights from ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-90/model.safetensors - [multilabel_classify.py:2218:_load_best_model]
|
| 886 |
+
2025-06-05 15:47:44,653 - INFO - ✔️ Weight for boost_mul matches between saved and loaded state_dict - [multilabel_classify.py:2230:_load_best_model]
|
| 887 |
+
2025-06-05 15:47:44,676 - INFO - ✔️ Weight for boost_add matches between saved and loaded state_dict - [multilabel_classify.py:2230:_load_best_model]
|
| 888 |
+
2025-06-05 15:47:44,691 - INFO -
|
| 889 |
+
[36m🚂 Training Metrics (Step 119) 🚂
|
| 890 |
+
+--------------------------+----------+
|
| 891 |
+
| Metric | Value |
|
| 892 |
+
+==========================+==========+
|
| 893 |
+
| train_runtime | 487.858 |
|
| 894 |
+
+--------------------------+----------+
|
| 895 |
+
| train_samples_per_second | 8.136 |
|
| 896 |
+
+--------------------------+----------+
|
| 897 |
+
| train_steps_per_second | 0.244 |
|
| 898 |
+
+--------------------------+----------+
|
| 899 |
+
| total_flos | 0 |
|
| 900 |
+
+--------------------------+----------+
|
| 901 |
+
| train_loss | 0.095432 |
|
| 902 |
+
+--------------------------+----------+
|
| 903 |
+
| epoch | 6.61972 |
|
| 904 |
+
+--------------------------+----------+[0m - [multilabel_classify.py:1836:on_log]
|
| 905 |
+
2025-06-05 15:47:44,692 - INFO - ✨ Training Completed! ✨ - [multilabel_classify.py:1709:on_train_end]
|
| 906 |
+
2025-06-05 15:47:44,763 - INFO - 📊 Training loss plot saved as '../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/train_loss_plot.png' - [multilabel_classify.py:1905:on_train_end]
|
| 907 |
+
2025-06-05 15:47:44,825 - INFO - 📊 Evaluation loss plot saved as '../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/eval_loss_plot.png' - [multilabel_classify.py:1919:on_train_end]
|
| 908 |
+
2025-06-05 15:47:44,881 - INFO - 📊 Evaluation metric plot saved as '../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/eval_precision_at_15_plot.png' - [multilabel_classify.py:1940:on_train_end]
|
| 909 |
+
2025-06-05 15:47:44,881 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 910 |
+
2025-06-05 15:47:44,882 - INFO - + ✨ MODEL SAVING + - [multilabel_classify.py:101:log_section]
|
| 911 |
+
2025-06-05 15:47:44,882 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:104:log_section]
|
| 912 |
+
2025-06-05 15:47:44,882 - INFO - 💾 Saving trained model and pushing to Hugging Face Hub... - [multilabel_classify.py:3708:main]
|
| 913 |
+
2025-06-05 15:47:44,882 - INFO - 📁 Creating/using output directory: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2693:save_and_push]
|
| 914 |
+
2025-06-05 15:47:46,066 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2093:_save]
|
| 915 |
+
2025-06-05 15:47:46,067 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2098:_save]
|
| 916 |
+
2025-06-05 15:47:46,068 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b:
|
| 917 |
+
+---------+-------------------------------+------------+
|
| 918 |
+
| Index | Saved File | Size |
|
| 919 |
+
+=========+===============================+============+
|
| 920 |
+
| 1 | eval_loss_plot.png | 0.04 MB |
|
| 921 |
+
+---------+-------------------------------+------------+
|
| 922 |
+
| 2 | training_args.bin | 0.01 MB |
|
| 923 |
+
+---------+-------------------------------+------------+
|
| 924 |
+
| 3 | model.safetensors | 4267.74 MB |
|
| 925 |
+
+---------+-------------------------------+------------+
|
| 926 |
+
| 4 | config.json | 0.00 MB |
|
| 927 |
+
+---------+-------------------------------+------------+
|
| 928 |
+
| 5 | train_loss_plot.png | 0.04 MB |
|
| 929 |
+
+---------+-------------------------------+------------+
|
| 930 |
+
| 6 | eval_precision_at_15_plot.png | 0.03 MB |
|
| 931 |
+
+---------+-------------------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 932 |
+
2025-06-05 15:47:49,635 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2093:_save]
|
| 933 |
+
2025-06-05 15:47:49,637 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2098:_save]
|
| 934 |
+
2025-06-05 15:47:49,638 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b:
|
| 935 |
+
+---------+-------------------------------+------------+
|
| 936 |
+
| Index | Saved File | Size |
|
| 937 |
+
+=========+===============================+============+
|
| 938 |
+
| 1 | eval_loss_plot.png | 0.04 MB |
|
| 939 |
+
+---------+-------------------------------+------------+
|
| 940 |
+
| 2 | training_args.bin | 0.01 MB |
|
| 941 |
+
+---------+-------------------------------+------------+
|
| 942 |
+
| 3 | model.safetensors | 4267.74 MB |
|
| 943 |
+
+---------+-------------------------------+------------+
|
| 944 |
+
| 4 | config.json | 0.00 MB |
|
| 945 |
+
+---------+-------------------------------+------------+
|
| 946 |
+
| 5 | train_loss_plot.png | 0.04 MB |
|
| 947 |
+
+---------+-------------------------------+------------+
|
| 948 |
+
| 6 | eval_precision_at_15_plot.png | 0.03 MB |
|
| 949 |
+
+---------+-------------------------------+------------+ - [multilabel_classify.py:2115:_save]
|
| 950 |
+
2025-06-05 15:49:10,190 - INFO - 💾 Model saved to: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2697:save_and_push]
|
| 951 |
+
2025-06-05 15:49:10,219 - INFO - 🖌️ Tokenizer saved to: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2701:save_and_push]
|