Trained classifier model on MIMIC-IV
Browse files- classification_log_2025-06-03_20-22-05.log +646 -0
- tokenizer.json +2 -14
classification_log_2025-06-03_20-22-05.log
ADDED
|
@@ -0,0 +1,646 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-06-03 20:22:05,457 - INFO - 🛠️ Command-line Arguments: - [multilabel_classify.py:323:print_args]
|
| 2 |
+
2025-06-03 20:22:05,457 - INFO -
|
| 3 |
+
🔹 output_dir: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b
|
| 4 |
+
🔹 source_url: XURLs.MIMIC4_DEMO
|
| 5 |
+
🔹 data: mimic4_icd10_full
|
| 6 |
+
🔹 logfile: classification_log
|
| 7 |
+
🔹 base_dir: ../tmp/MIMIC4_DEMO
|
| 8 |
+
🔹 hub_model_id: deb101/mistral-7b-instruct-v0.3-mimic4-adapt
|
| 9 |
+
🔹 model_name: mistralai/Mistral-7B-Instruct-v0.3
|
| 10 |
+
🔹 max_length: 512
|
| 11 |
+
🔹 do_fresh_training: True
|
| 12 |
+
🔹 load_from_checkpoint: False
|
| 13 |
+
🔹 task: multilabel-classify
|
| 14 |
+
🔹 num_train_epochs: 4
|
| 15 |
+
🔹 metric_for_best_model: precision_at_15
|
| 16 |
+
🔹 learning_rate: 0.0001
|
| 17 |
+
🔹 final_lr_scheduling: 1e-06
|
| 18 |
+
🔹 warmup_steps: 500
|
| 19 |
+
🔹 source: /home/ubuntu/.xcube/data/mimic4_demo
|
| 20 |
+
🔹 logfile_path: ../tmp/MIMIC4_DEMO/logs/classification_log_2025-06-03_20-22-05.log - [multilabel_classify.py:324:print_args]
|
| 21 |
+
2025-06-03 20:22:05,457 - INFO - ➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖ - [multilabel_classify.py:325:print_args]
|
| 22 |
+
2025-06-03 20:22:05,458 - INFO - +----------------------------------------------------+ - [multilabel_classify.py:133:log_with_box]
|
| 23 |
+
2025-06-03 20:22:05,458 - INFO - | 🤖🏷️ ****Multi-Label Classification Started**** 🏷️🤖 | - [multilabel_classify.py:133:log_with_box]
|
| 24 |
+
2025-06-03 20:22:05,458 - INFO - +----------------------------------------------------+ - [multilabel_classify.py:133:log_with_box]
|
| 25 |
+
2025-06-03 20:22:05,458 - INFO - ================================================================================ - [multilabel_classify.py:96:log_section]
|
| 26 |
+
2025-06-03 20:22:05,458 - INFO - = 📌 INITIALIZING TRAINING ENVIRONMENT = - [multilabel_classify.py:97:log_section]
|
| 27 |
+
2025-06-03 20:22:05,458 - INFO - ================================================================================ - [multilabel_classify.py:100:log_section]
|
| 28 |
+
2025-06-03 20:22:05,458 - INFO - Hub Model ID for this multi-label classification task: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify - [multilabel_classify.py:3227:main]
|
| 29 |
+
2025-06-03 20:22:05,504 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:96:log_section]
|
| 30 |
+
2025-06-03 20:22:05,505 - INFO - + ✨ LOADING DATASETS + - [multilabel_classify.py:97:log_section]
|
| 31 |
+
2025-06-03 20:22:05,505 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 32 |
+
2025-06-03 20:22:05,505 - INFO - 📊 Loading main datasets.... - [multilabel_classify.py:3245:main]
|
| 33 |
+
2025-06-03 20:22:13,776 - INFO - 🔍 Total unique labels in dataset: 7942 - [multilabel_classify.py:3086:sample_df_with_full_label_coverage]
|
| 34 |
+
2025-06-03 20:22:13,789 - INFO - 🧪 Attempt 1: Sampled 122 rows covering 863 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 35 |
+
2025-06-03 20:22:13,798 - INFO - 🧪 Attempt 2: Sampled 122 rows covering 816 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 36 |
+
2025-06-03 20:22:13,807 - INFO - 🧪 Attempt 3: Sampled 122 rows covering 885 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 37 |
+
2025-06-03 20:22:13,816 - INFO - 🧪 Attempt 4: Sampled 122 rows covering 828 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 38 |
+
2025-06-03 20:22:13,825 - INFO - 🧪 Attempt 5: Sampled 122 rows covering 879 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 39 |
+
2025-06-03 20:22:13,834 - INFO - 🧪 Attempt 6: Sampled 122 rows covering 852 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 40 |
+
2025-06-03 20:22:13,842 - INFO - 🧪 Attempt 7: Sampled 122 rows covering 838 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 41 |
+
2025-06-03 20:22:13,850 - INFO - 🧪 Attempt 8: Sampled 122 rows covering 851 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 42 |
+
2025-06-03 20:22:13,859 - INFO - 🧪 Attempt 9: Sampled 122 rows covering 825 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 43 |
+
2025-06-03 20:22:13,867 - INFO - 🧪 Attempt 10: Sampled 122 rows covering 833 labels. - [multilabel_classify.py:3100:sample_df_with_full_label_coverage]
|
| 44 |
+
2025-06-03 20:22:13,868 - INFO - ⚠️ Skipping label coverage fix. 7109 labels are missing. - [multilabel_classify.py:3118:sample_df_with_full_label_coverage]
|
| 45 |
+
2025-06-03 20:22:13,868 - INFO - ✅ Final row count: 122 (Valid: 20, Not-valid: 102) - [multilabel_classify.py:3121:sample_df_with_full_label_coverage]
|
| 46 |
+
2025-06-03 20:22:13,886 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:96:log_section]
|
| 47 |
+
2025-06-03 20:22:13,887 - INFO - + ✨ STARTING FRESH TRAINING + - [multilabel_classify.py:97:log_section]
|
| 48 |
+
2025-06-03 20:22:13,887 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 49 |
+
2025-06-03 20:22:13,887 - INFO - 🔄 Starting fresh training (either forced or model not found)... - [multilabel_classify.py:3266:main]
|
| 50 |
+
2025-06-03 20:22:13,902 - 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]
|
| 51 |
+
2025-06-03 20:22:13,902 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:96:log_section]
|
| 52 |
+
2025-06-03 20:22:13,902 - INFO - + ✨ LOADING BASE MODEL + - [multilabel_classify.py:97:log_section]
|
| 53 |
+
2025-06-03 20:22:13,902 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 54 |
+
2025-06-03 20:22:13,902 - INFO - 📥 Loading pretrained model and tokenizer... - [multilabel_classify.py:3296:main]
|
| 55 |
+
2025-06-03 20:22:13,902 - INFO - 🚀 Starting model and tokenizer loading process... - [multilabel_classify.py:1198:load_base_model_and_tokenizer]
|
| 56 |
+
2025-06-03 20:22:13,903 - INFO - 📊 Quantization config: BitsAndBytesConfig {
|
| 57 |
+
"_load_in_4bit": true,
|
| 58 |
+
"_load_in_8bit": false,
|
| 59 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
| 60 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 61 |
+
"bnb_4bit_quant_type": "nf4",
|
| 62 |
+
"bnb_4bit_use_double_quant": true,
|
| 63 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
| 64 |
+
"llm_int8_has_fp16_weight": false,
|
| 65 |
+
"llm_int8_skip_modules": null,
|
| 66 |
+
"llm_int8_threshold": 6.0,
|
| 67 |
+
"load_in_4bit": true,
|
| 68 |
+
"load_in_8bit": false,
|
| 69 |
+
"quant_method": "bitsandbytes"
|
| 70 |
+
}
|
| 71 |
+
- [multilabel_classify.py:1207:load_base_model_and_tokenizer]
|
| 72 |
+
2025-06-03 20:22:13,904 - INFO - 🔤 Loading tokenizer for model: deb101/mistral-7b-instruct-v0.3-mimic4-adapt... - [multilabel_classify.py:1211:load_base_model_and_tokenizer]
|
| 73 |
+
2025-06-03 20:22:14,413 - INFO - 🔍 Checking if deb101/mistral-7b-instruct-v0.3-mimic4-adapt is a PEFT model... - [multilabel_classify.py:1223:load_base_model_and_tokenizer]
|
| 74 |
+
2025-06-03 20:22:14,438 - INFO - ✅ Detected PEFT model. Base model: mistralai/Mistral-7B-Instruct-v0.3 - [multilabel_classify.py:1227:load_base_model_and_tokenizer]
|
| 75 |
+
2025-06-03 20:22:14,438 - INFO - 🔍 Loading model configuration for mistralai/Mistral-7B-Instruct-v0.3... - [multilabel_classify.py:1237:load_base_model_and_tokenizer]
|
| 76 |
+
2025-06-03 20:22:14,461 - INFO - Model type: mistral, Architectures: ['MistralForCausalLM'] - [multilabel_classify.py:1243:load_base_model_and_tokenizer]
|
| 77 |
+
2025-06-03 20:22:14,461 - INFO - 🧠 Loading base model: mistralai/Mistral-7B-Instruct-v0.3... - [multilabel_classify.py:1306:load_base_model_and_tokenizer]
|
| 78 |
+
2025-06-03 20:22:15,021 - 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]
|
| 79 |
+
2025-06-03 20:22:20,344 - INFO - 🧩 Loading PEFT adapters for deb101/mistral-7b-instruct-v0.3-mimic4-adapt... - [multilabel_classify.py:1321:load_base_model_and_tokenizer]
|
| 80 |
+
2025-06-03 20:22:20,612 - INFO - 🔧 Before enabling PEFT adapters for training - [multilabel_classify.py:1323:load_base_model_and_tokenizer]
|
| 81 |
+
2025-06-03 20:22:20,617 - INFO - 🔧 After Enabling PEFT adapters for training - [multilabel_classify.py:1330:load_base_model_and_tokenizer]
|
| 82 |
+
2025-06-03 20:22:20,620 - INFO - ✅ Model and tokenizer successfully loaded! - [multilabel_classify.py:1371:load_base_model_and_tokenizer]
|
| 83 |
+
2025-06-03 20:22:20,620 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:96:log_section]
|
| 84 |
+
2025-06-03 20:22:20,620 - INFO - + ✨ DATA PREPROCESSING + - [multilabel_classify.py:97:log_section]
|
| 85 |
+
2025-06-03 20:22:20,620 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 86 |
+
2025-06-03 20:22:20,620 - INFO - 🔄 Loading and preprocessing training data... - [multilabel_classify.py:3304:main]
|
| 87 |
+
2025-06-03 20:22:20,624 - INFO - Total number of labels: 833 - [multilabel_classify.py:999:preprocess_data]
|
| 88 |
+
2025-06-03 20:22:20,624 - INFO - Rare labels (freq < 50): 832 - [multilabel_classify.py:1000:preprocess_data]
|
| 89 |
+
2025-06-03 20:22:20,624 - INFO - Not rare labels (freq >= 50): 1 - [multilabel_classify.py:1001:preprocess_data]
|
| 90 |
+
2025-06-03 20:22:20,624 - INFO - Label partitions and classes saved to ../tmp/MIMIC4_DEMO/labels_partition.json - [multilabel_classify.py:1002:preprocess_data]
|
| 91 |
+
2025-06-03 20:22:21,956 - INFO - The size of training set: 567 - [multilabel_classify.py:1098:preprocess_data]
|
| 92 |
+
2025-06-03 20:22:21,956 - INFO - The size of Evaluation set: 136 - [multilabel_classify.py:1099:preprocess_data]
|
| 93 |
+
2025-06-03 20:22:21,961 - INFO - Number of unique ICD-10 codes: 833 - [multilabel_classify.py:3310:main]
|
| 94 |
+
2025-06-03 20:22:21,961 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:96:log_section]
|
| 95 |
+
2025-06-03 20:22:21,961 - INFO - + ✨ MODEL INITIALIZATION + - [multilabel_classify.py:97:log_section]
|
| 96 |
+
2025-06-03 20:22:21,961 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 97 |
+
2025-06-03 20:22:21,961 - INFO - 🧠 Initializing custom L2R model for outputting per-token relevance scores per ICD-10 codes. - [multilabel_classify.py:3313:main]
|
| 98 |
+
2025-06-03 20:22:21,962 - INFO - Will now start to create Multilabel-Classification Model from the base model - [multilabel_classify.py:516:__init__]
|
| 99 |
+
2025-06-03 20:22:21,965 - INFO - Trainable params: 6815744 / 3765178368 (0.18%) - [multilabel_classify.py:568:compute_trainable_params]
|
| 100 |
+
2025-06-03 20:22:22,304 - INFO - Creating the Multi-Label Classification Model from base model mistralai/Mistral-7B-Instruct-v0.3 completed!!! - [multilabel_classify.py:558:__init__]
|
| 101 |
+
2025-06-03 20:22:22,307 - INFO - Trainable params: 84177025 / 3842539649 (2.19%) - [multilabel_classify.py:568:compute_trainable_params]
|
| 102 |
+
2025-06-03 20:22:22,307 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:96:log_section]
|
| 103 |
+
2025-06-03 20:22:22,307 - INFO - + ✨ TRAINING PREPARATION + - [multilabel_classify.py:97:log_section]
|
| 104 |
+
2025-06-03 20:22:22,307 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 105 |
+
2025-06-03 20:22:22,307 - INFO - ⚙️ Preparing training components and optimizers... - [multilabel_classify.py:3320:main]
|
| 106 |
+
2025-06-03 20:22:22,367 - INFO - 🖥️ Device: NVIDIA GH200 480GB - [multilabel_classify.py:845:log_training_configuration]
|
| 107 |
+
2025-06-03 20:22:22,367 - INFO - 🔋 CUDA Available: True - [multilabel_classify.py:848:log_training_configuration]
|
| 108 |
+
2025-06-03 20:22:22,367 - INFO - 💾 CUDA Device Count: 1 - [multilabel_classify.py:849:log_training_configuration]
|
| 109 |
+
2025-06-03 20:22:22,368 - INFO -
|
| 110 |
+
📋 Training Configuration 📋
|
| 111 |
+
+----------+-----------------------------+------------------------------------------------------------------+
|
| 112 |
+
| 🌟 Emoji | 🏷️ Parameter | 📊 Value |
|
| 113 |
+
+----------+-----------------------------+------------------------------------------------------------------+
|
| 114 |
+
| 📁 | Output Directory | ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b |
|
| 115 |
+
| 🔁 | Training Epochs | 4 |
|
| 116 |
+
| 🏋️ | Train Batch Size | 8 |
|
| 117 |
+
| 🔍 | Eval Batch Size | 8 |
|
| 118 |
+
| 📊 | Gradient Accumulation Steps | 4 |
|
| 119 |
+
| 🚀 | Learning Rate | 0.0001 |
|
| 120 |
+
| 🌅 | Warmup Steps | 500 |
|
| 121 |
+
| 💾 | Save Strategy | epoch |
|
| 122 |
+
| 💾 | Save Total Limit | 10 |
|
| 123 |
+
| 📊 | Evaluation Strategy | epoch |
|
| 124 |
+
| 🎯 | Best Model Metric | precision_at_15 |
|
| 125 |
+
| 📝 | Logging Strategy | steps (every 10 steps) |
|
| 126 |
+
| 🌐 | Push to Hub | True |
|
| 127 |
+
| 🌐 | Hub Model ID | deb101/mistral-7b-instruct-v0.3-mimic4-adapt-multilabel-classify |
|
| 128 |
+
| 🔢 | Steps per Epoch | 17 |
|
| 129 |
+
| 🔢 | Total Training Steps | 68 |
|
| 130 |
+
| 🔢 | Evaluation Steps | 17 |
|
| 131 |
+
| 📊 | Training Dataset Size | 567 samples 🏋️ |
|
| 132 |
+
| 📊 | Evaluation Dataset Size | 136 samples 🔍 |
|
| 133 |
+
+----------+-----------------------------+------------------------------------------------------------------+ - [multilabel_classify.py:837:log_training_args]
|
| 134 |
+
2025-06-03 20:22:22,368 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:96:log_section]
|
| 135 |
+
2025-06-03 20:22:22,368 - INFO - + ✨ MODEL TRAINING + - [multilabel_classify.py:97:log_section]
|
| 136 |
+
2025-06-03 20:22:22,368 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 137 |
+
2025-06-03 20:22:22,368 - INFO - 🏋️ Starting model training process... - [multilabel_classify.py:3340:main]
|
| 138 |
+
2025-06-03 20:22:22,411 - INFO - We are registering the tokenizer deb101/mistral-7b-instruct-v0.3-mimic4-adapt in Custom Trainer - [multilabel_classify.py:1941:__init__]
|
| 139 |
+
2025-06-03 20:22:22,667 - INFO - 🚀 Starting Training... - [multilabel_classify.py:1595:on_train_begin]
|
| 140 |
+
2025-06-03 20:22:43,213 - INFO -
|
| 141 |
+
[36m🚂 Training Metrics (Step 10) 🚂
|
| 142 |
+
+---------------+---------+
|
| 143 |
+
| Metric | Value |
|
| 144 |
+
+===============+=========+
|
| 145 |
+
| loss | 0.624 |
|
| 146 |
+
+---------------+---------+
|
| 147 |
+
| grad_norm | 6.71916 |
|
| 148 |
+
+---------------+---------+
|
| 149 |
+
| learning_rate | 2e-06 |
|
| 150 |
+
+---------------+---------+
|
| 151 |
+
| epoch | 0.56338 |
|
| 152 |
+
+---------------+---------+[0m - [multilabel_classify.py:1789:on_log]
|
| 153 |
+
2025-06-03 20:22:57,293 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:1953:evaluate]
|
| 154 |
+
2025-06-03 20:23:32,597 - INFO -
|
| 155 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 156 |
+
+-------------------------------+----------+
|
| 157 |
+
| Metric | Value |
|
| 158 |
+
+===============================+==========+
|
| 159 |
+
| eval_f1_micro | 0 |
|
| 160 |
+
+-------------------------------+----------+
|
| 161 |
+
| eval_f1_macro | 0 |
|
| 162 |
+
+-------------------------------+----------+
|
| 163 |
+
| eval_precision_at_5 | 0.005882 |
|
| 164 |
+
+-------------------------------+----------+
|
| 165 |
+
| eval_recall_at_5 | 0.001306 |
|
| 166 |
+
+-------------------------------+----------+
|
| 167 |
+
| eval_precision_at_8 | 0.011029 |
|
| 168 |
+
+-------------------------------+----------+
|
| 169 |
+
| eval_recall_at_8 | 0.003704 |
|
| 170 |
+
+-------------------------------+----------+
|
| 171 |
+
| eval_precision_at_15 | 0.014706 |
|
| 172 |
+
+-------------------------------+----------+
|
| 173 |
+
| eval_recall_at_15 | 0.011093 |
|
| 174 |
+
+-------------------------------+----------+
|
| 175 |
+
| eval_rare_f1_micro | 0 |
|
| 176 |
+
+-------------------------------+----------+
|
| 177 |
+
| eval_rare_f1_macro | 0 |
|
| 178 |
+
+-------------------------------+----------+
|
| 179 |
+
| eval_rare_precision | 0 |
|
| 180 |
+
+-------------------------------+----------+
|
| 181 |
+
| eval_rare_recall | 0 |
|
| 182 |
+
+-------------------------------+----------+
|
| 183 |
+
| eval_rare_precision_at_5 | 0.010294 |
|
| 184 |
+
+-------------------------------+----------+
|
| 185 |
+
| eval_rare_recall_at_5 | 0.002439 |
|
| 186 |
+
+-------------------------------+----------+
|
| 187 |
+
| eval_rare_precision_at_8 | 0.012868 |
|
| 188 |
+
+-------------------------------+----------+
|
| 189 |
+
| eval_rare_recall_at_8 | 0.00592 |
|
| 190 |
+
+-------------------------------+----------+
|
| 191 |
+
| eval_rare_precision_at_15 | 0.014216 |
|
| 192 |
+
+-------------------------------+----------+
|
| 193 |
+
| eval_rare_recall_at_15 | 0.011398 |
|
| 194 |
+
+-------------------------------+----------+
|
| 195 |
+
| eval_not_rare_f1_micro | 0.595588 |
|
| 196 |
+
+-------------------------------+----------+
|
| 197 |
+
| eval_not_rare_f1_macro | 0.373272 |
|
| 198 |
+
+-------------------------------+----------+
|
| 199 |
+
| eval_not_rare_precision | 0.595588 |
|
| 200 |
+
+-------------------------------+----------+
|
| 201 |
+
| eval_not_rare_recall | 0.595588 |
|
| 202 |
+
+-------------------------------+----------+
|
| 203 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 204 |
+
+-------------------------------+----------+
|
| 205 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 206 |
+
+-------------------------------+----------+
|
| 207 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 208 |
+
+-------------------------------+----------+
|
| 209 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 210 |
+
+-------------------------------+----------+
|
| 211 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 212 |
+
+-------------------------------+----------+
|
| 213 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 214 |
+
+-------------------------------+----------+
|
| 215 |
+
| eval_loss | 0.221656 |
|
| 216 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1808:on_evaluate]
|
| 217 |
+
2025-06-03 20:23:36,490 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-18 - [multilabel_classify.py:2046:_save]
|
| 218 |
+
2025-06-03 20:23:36,493 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-18 - [multilabel_classify.py:2051:_save]
|
| 219 |
+
2025-06-03 20:23:36,494 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-18:
|
| 220 |
+
+---------+--------------------+------------+
|
| 221 |
+
| Index | Saved File | Size |
|
| 222 |
+
+=========+====================+============+
|
| 223 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 224 |
+
+---------+--------------------+------------+
|
| 225 |
+
| 2 | optimizer.pt | 642.30 MB |
|
| 226 |
+
+---------+--------------------+------------+
|
| 227 |
+
| 3 | model.safetensors | 4267.74 MB |
|
| 228 |
+
+---------+--------------------+------------+
|
| 229 |
+
| 4 | scaler.pt | 0.00 MB |
|
| 230 |
+
+---------+--------------------+------------+
|
| 231 |
+
| 5 | config.json | 0.04 MB |
|
| 232 |
+
+---------+--------------------+------------+
|
| 233 |
+
| 6 | scheduler.pt | 0.00 MB |
|
| 234 |
+
+---------+--------------------+------------+
|
| 235 |
+
| 7 | trainer_state.json | 0.00 MB |
|
| 236 |
+
+---------+--------------------+------------+
|
| 237 |
+
| 8 | rng_state.pth | 0.01 MB |
|
| 238 |
+
+---------+--------------------+------------+ - [multilabel_classify.py:2068:_save]
|
| 239 |
+
2025-06-03 20:23:42,667 - INFO -
|
| 240 |
+
[36m🚂 Training Metrics (Step 20) 🚂
|
| 241 |
+
+---------------+---------+
|
| 242 |
+
| Metric | Value |
|
| 243 |
+
+===============+=========+
|
| 244 |
+
| loss | 0.3386 |
|
| 245 |
+
+---------------+---------+
|
| 246 |
+
| grad_norm | 2.49967 |
|
| 247 |
+
+---------------+---------+
|
| 248 |
+
| learning_rate | 4e-06 |
|
| 249 |
+
+---------------+---------+
|
| 250 |
+
| epoch | 1.11268 |
|
| 251 |
+
+---------------+---------+[0m - [multilabel_classify.py:1789:on_log]
|
| 252 |
+
2025-06-03 20:24:00,820 - INFO -
|
| 253 |
+
[36m🚂 Training Metrics (Step 30) 🚂
|
| 254 |
+
+---------------+----------+
|
| 255 |
+
| Metric | Value |
|
| 256 |
+
+===============+==========+
|
| 257 |
+
| loss | 0.1249 |
|
| 258 |
+
+---------------+----------+
|
| 259 |
+
| grad_norm | 0.258035 |
|
| 260 |
+
+---------------+----------+
|
| 261 |
+
| learning_rate | 6e-06 |
|
| 262 |
+
+---------------+----------+
|
| 263 |
+
| epoch | 1.67606 |
|
| 264 |
+
+---------------+----------+[0m - [multilabel_classify.py:1789:on_log]
|
| 265 |
+
2025-06-03 20:24:11,212 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:1953:evaluate]
|
| 266 |
+
2025-06-03 20:24:46,604 - INFO -
|
| 267 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 268 |
+
+-------------------------------+----------+
|
| 269 |
+
| Metric | Value |
|
| 270 |
+
+===============================+==========+
|
| 271 |
+
| eval_f1_micro | 0 |
|
| 272 |
+
+-------------------------------+----------+
|
| 273 |
+
| eval_f1_macro | 0 |
|
| 274 |
+
+-------------------------------+----------+
|
| 275 |
+
| eval_precision_at_5 | 0.008824 |
|
| 276 |
+
+-------------------------------+----------+
|
| 277 |
+
| eval_recall_at_5 | 0.001777 |
|
| 278 |
+
+-------------------------------+----------+
|
| 279 |
+
| eval_precision_at_8 | 0.011029 |
|
| 280 |
+
+-------------------------------+----------+
|
| 281 |
+
| eval_recall_at_8 | 0.003763 |
|
| 282 |
+
+-------------------------------+----------+
|
| 283 |
+
| eval_precision_at_15 | 0.015196 |
|
| 284 |
+
+-------------------------------+----------+
|
| 285 |
+
| eval_recall_at_15 | 0.010958 |
|
| 286 |
+
+-------------------------------+----------+
|
| 287 |
+
| eval_rare_f1_micro | 0 |
|
| 288 |
+
+-------------------------------+----------+
|
| 289 |
+
| eval_rare_f1_macro | 0 |
|
| 290 |
+
+-------------------------------+----------+
|
| 291 |
+
| eval_rare_precision | 0 |
|
| 292 |
+
+-------------------------------+----------+
|
| 293 |
+
| eval_rare_recall | 0 |
|
| 294 |
+
+-------------------------------+----------+
|
| 295 |
+
| eval_rare_precision_at_5 | 0.008824 |
|
| 296 |
+
+-------------------------------+----------+
|
| 297 |
+
| eval_rare_recall_at_5 | 0.001785 |
|
| 298 |
+
+-------------------------------+----------+
|
| 299 |
+
| eval_rare_precision_at_8 | 0.011029 |
|
| 300 |
+
+-------------------------------+----------+
|
| 301 |
+
| eval_rare_recall_at_8 | 0.004688 |
|
| 302 |
+
+-------------------------------+----------+
|
| 303 |
+
| eval_rare_precision_at_15 | 0.014216 |
|
| 304 |
+
+-------------------------------+----------+
|
| 305 |
+
| eval_rare_recall_at_15 | 0.012365 |
|
| 306 |
+
+-------------------------------+----------+
|
| 307 |
+
| eval_not_rare_f1_micro | 0.595588 |
|
| 308 |
+
+-------------------------------+----------+
|
| 309 |
+
| eval_not_rare_f1_macro | 0.373272 |
|
| 310 |
+
+-------------------------------+----------+
|
| 311 |
+
| eval_not_rare_precision | 0.595588 |
|
| 312 |
+
+-------------------------------+----------+
|
| 313 |
+
| eval_not_rare_recall | 0.595588 |
|
| 314 |
+
+-------------------------------+----------+
|
| 315 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 316 |
+
+-------------------------------+----------+
|
| 317 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 318 |
+
+-------------------------------+----------+
|
| 319 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 320 |
+
+-------------------------------+----------+
|
| 321 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 322 |
+
+-------------------------------+----------+
|
| 323 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 324 |
+
+-------------------------------+----------+
|
| 325 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 326 |
+
+-------------------------------+----------+
|
| 327 |
+
| eval_loss | 0.121922 |
|
| 328 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1808:on_evaluate]
|
| 329 |
+
2025-06-03 20:24:50,330 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-36 - [multilabel_classify.py:2046:_save]
|
| 330 |
+
2025-06-03 20:24:50,333 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-36 - [multilabel_classify.py:2051:_save]
|
| 331 |
+
2025-06-03 20:24:50,334 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-36:
|
| 332 |
+
+---------+--------------------+------------+
|
| 333 |
+
| Index | Saved File | Size |
|
| 334 |
+
+=========+====================+============+
|
| 335 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 336 |
+
+---------+--------------------+------------+
|
| 337 |
+
| 2 | optimizer.pt | 642.30 MB |
|
| 338 |
+
+---------+--------------------+------------+
|
| 339 |
+
| 3 | model.safetensors | 4267.74 MB |
|
| 340 |
+
+---------+--------------------+------------+
|
| 341 |
+
| 4 | scaler.pt | 0.00 MB |
|
| 342 |
+
+---------+--------------------+------------+
|
| 343 |
+
| 5 | config.json | 0.04 MB |
|
| 344 |
+
+---------+--------------------+------------+
|
| 345 |
+
| 6 | scheduler.pt | 0.00 MB |
|
| 346 |
+
+---------+--------------------+------------+
|
| 347 |
+
| 7 | trainer_state.json | 0.00 MB |
|
| 348 |
+
+---------+--------------------+------------+
|
| 349 |
+
| 8 | rng_state.pth | 0.01 MB |
|
| 350 |
+
+---------+--------------------+------------+ - [multilabel_classify.py:2068:_save]
|
| 351 |
+
2025-06-03 20:25:00,280 - INFO -
|
| 352 |
+
[36m🚂 Training Metrics (Step 40) 🚂
|
| 353 |
+
+---------------+----------+
|
| 354 |
+
| Metric | Value |
|
| 355 |
+
+===============+==========+
|
| 356 |
+
| loss | 0.1098 |
|
| 357 |
+
+---------------+----------+
|
| 358 |
+
| grad_norm | 0.470693 |
|
| 359 |
+
+---------------+----------+
|
| 360 |
+
| learning_rate | 8e-06 |
|
| 361 |
+
+---------------+----------+
|
| 362 |
+
| epoch | 2.22535 |
|
| 363 |
+
+---------------+----------+[0m - [multilabel_classify.py:1789:on_log]
|
| 364 |
+
2025-06-03 20:25:18,391 - INFO -
|
| 365 |
+
[36m🚂 Training Metrics (Step 50) 🚂
|
| 366 |
+
+---------------+----------+
|
| 367 |
+
| Metric | Value |
|
| 368 |
+
+===============+==========+
|
| 369 |
+
| loss | 0.1086 |
|
| 370 |
+
+---------------+----------+
|
| 371 |
+
| grad_norm | 0.067761 |
|
| 372 |
+
+---------------+----------+
|
| 373 |
+
| learning_rate | 1e-05 |
|
| 374 |
+
+---------------+----------+
|
| 375 |
+
| epoch | 2.78873 |
|
| 376 |
+
+---------------+----------+[0m - [multilabel_classify.py:1789:on_log]
|
| 377 |
+
2025-06-03 20:25:25,137 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:1953:evaluate]
|
| 378 |
+
2025-06-03 20:26:00,429 - INFO -
|
| 379 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 380 |
+
+-------------------------------+----------+
|
| 381 |
+
| Metric | Value |
|
| 382 |
+
+===============================+==========+
|
| 383 |
+
| eval_f1_micro | 0 |
|
| 384 |
+
+-------------------------------+----------+
|
| 385 |
+
| eval_f1_macro | 0 |
|
| 386 |
+
+-------------------------------+----------+
|
| 387 |
+
| eval_precision_at_5 | 0.047059 |
|
| 388 |
+
+-------------------------------+----------+
|
| 389 |
+
| eval_recall_at_5 | 0.012123 |
|
| 390 |
+
+-------------------------------+----------+
|
| 391 |
+
| eval_precision_at_8 | 0.050551 |
|
| 392 |
+
+-------------------------------+----------+
|
| 393 |
+
| eval_recall_at_8 | 0.02375 |
|
| 394 |
+
+-------------------------------+----------+
|
| 395 |
+
| eval_precision_at_15 | 0.045588 |
|
| 396 |
+
+-------------------------------+----------+
|
| 397 |
+
| eval_recall_at_15 | 0.045718 |
|
| 398 |
+
+-------------------------------+----------+
|
| 399 |
+
| eval_rare_f1_micro | 0 |
|
| 400 |
+
+-------------------------------+----------+
|
| 401 |
+
| eval_rare_f1_macro | 0 |
|
| 402 |
+
+-------------------------------+----------+
|
| 403 |
+
| eval_rare_precision | 0 |
|
| 404 |
+
+-------------------------------+----------+
|
| 405 |
+
| eval_rare_recall | 0 |
|
| 406 |
+
+-------------------------------+----------+
|
| 407 |
+
| eval_rare_precision_at_5 | 0.036765 |
|
| 408 |
+
+-------------------------------+----------+
|
| 409 |
+
| eval_rare_recall_at_5 | 0.010554 |
|
| 410 |
+
+-------------------------------+----------+
|
| 411 |
+
| eval_rare_precision_at_8 | 0.04136 |
|
| 412 |
+
+-------------------------------+----------+
|
| 413 |
+
| eval_rare_recall_at_8 | 0.019845 |
|
| 414 |
+
+-------------------------------+----------+
|
| 415 |
+
| eval_rare_precision_at_15 | 0.036765 |
|
| 416 |
+
+-------------------------------+----------+
|
| 417 |
+
| eval_rare_recall_at_15 | 0.037016 |
|
| 418 |
+
+-------------------------------+----------+
|
| 419 |
+
| eval_not_rare_f1_micro | 0.595588 |
|
| 420 |
+
+-------------------------------+----------+
|
| 421 |
+
| eval_not_rare_f1_macro | 0.373272 |
|
| 422 |
+
+-------------------------------+----------+
|
| 423 |
+
| eval_not_rare_precision | 0.595588 |
|
| 424 |
+
+-------------------------------+----------+
|
| 425 |
+
| eval_not_rare_recall | 0.595588 |
|
| 426 |
+
+-------------------------------+----------+
|
| 427 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 428 |
+
+-------------------------------+----------+
|
| 429 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 430 |
+
+-------------------------------+----------+
|
| 431 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 432 |
+
+-------------------------------+----------+
|
| 433 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 434 |
+
+-------------------------------+----------+
|
| 435 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 436 |
+
+-------------------------------+----------+
|
| 437 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 438 |
+
+-------------------------------+----------+
|
| 439 |
+
| eval_loss | 0.107352 |
|
| 440 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1808:on_evaluate]
|
| 441 |
+
2025-06-03 20:26:01,710 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-54 - [multilabel_classify.py:2046:_save]
|
| 442 |
+
2025-06-03 20:26:01,713 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-54 - [multilabel_classify.py:2051:_save]
|
| 443 |
+
2025-06-03 20:26:01,714 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-54:
|
| 444 |
+
+---------+-------------------+------------+
|
| 445 |
+
| Index | Saved File | Size |
|
| 446 |
+
+=========+===================+============+
|
| 447 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 448 |
+
+---------+-------------------+------------+
|
| 449 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 450 |
+
+---------+-------------------+------------+
|
| 451 |
+
| 3 | config.json | 0.04 MB |
|
| 452 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2068:_save]
|
| 453 |
+
2025-06-03 20:26:15,035 - INFO -
|
| 454 |
+
[36m🚂 Training Metrics (Step 60) 🚂
|
| 455 |
+
+---------------+----------+
|
| 456 |
+
| Metric | Value |
|
| 457 |
+
+===============+==========+
|
| 458 |
+
| loss | 0.1033 |
|
| 459 |
+
+---------------+----------+
|
| 460 |
+
| grad_norm | 0.206313 |
|
| 461 |
+
+---------------+----------+
|
| 462 |
+
| learning_rate | 1.2e-05 |
|
| 463 |
+
+---------------+----------+
|
| 464 |
+
| epoch | 3.33803 |
|
| 465 |
+
+---------------+----------+[0m - [multilabel_classify.py:1789:on_log]
|
| 466 |
+
2025-06-03 20:26:30,807 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-68 - [multilabel_classify.py:2046:_save]
|
| 467 |
+
2025-06-03 20:26:30,810 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-68 - [multilabel_classify.py:2051:_save]
|
| 468 |
+
2025-06-03 20:26:30,811 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-68:
|
| 469 |
+
+---------+-------------------+------------+
|
| 470 |
+
| Index | Saved File | Size |
|
| 471 |
+
+=========+===================+============+
|
| 472 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 473 |
+
+---------+-------------------+------------+
|
| 474 |
+
| 2 | model.safetensors | 4267.74 MB |
|
| 475 |
+
+---------+-------------------+------------+
|
| 476 |
+
| 3 | config.json | 0.04 MB |
|
| 477 |
+
+---------+-------------------+------------+ - [multilabel_classify.py:2068:_save]
|
| 478 |
+
2025-06-03 20:26:31,136 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [multilabel_classify.py:1953:evaluate]
|
| 479 |
+
2025-06-03 20:27:06,536 - INFO -
|
| 480 |
+
[33m🔍 Evaluation Metrics 🔍
|
| 481 |
+
+-------------------------------+----------+
|
| 482 |
+
| Metric | Value |
|
| 483 |
+
+===============================+==========+
|
| 484 |
+
| eval_f1_micro | 0 |
|
| 485 |
+
+-------------------------------+----------+
|
| 486 |
+
| eval_f1_macro | 0 |
|
| 487 |
+
+-------------------------------+----------+
|
| 488 |
+
| eval_precision_at_5 | 0.167647 |
|
| 489 |
+
+-------------------------------+----------+
|
| 490 |
+
| eval_recall_at_5 | 0.059694 |
|
| 491 |
+
+-------------------------------+----------+
|
| 492 |
+
| eval_precision_at_8 | 0.137868 |
|
| 493 |
+
+-------------------------------+----------+
|
| 494 |
+
| eval_recall_at_8 | 0.077015 |
|
| 495 |
+
+-------------------------------+----------+
|
| 496 |
+
| eval_precision_at_15 | 0.092647 |
|
| 497 |
+
+-------------------------------+----------+
|
| 498 |
+
| eval_recall_at_15 | 0.099223 |
|
| 499 |
+
+-------------------------------+----------+
|
| 500 |
+
| eval_rare_f1_micro | 0 |
|
| 501 |
+
+-------------------------------+----------+
|
| 502 |
+
| eval_rare_f1_macro | 0 |
|
| 503 |
+
+-------------------------------+----------+
|
| 504 |
+
| eval_rare_precision | 0 |
|
| 505 |
+
+-------------------------------+----------+
|
| 506 |
+
| eval_rare_recall | 0 |
|
| 507 |
+
+-------------------------------+----------+
|
| 508 |
+
| eval_rare_precision_at_5 | 0.127941 |
|
| 509 |
+
+-------------------------------+----------+
|
| 510 |
+
| eval_rare_recall_at_5 | 0.046166 |
|
| 511 |
+
+-------------------------------+----------+
|
| 512 |
+
| eval_rare_precision_at_8 | 0.09375 |
|
| 513 |
+
+-------------------------------+----------+
|
| 514 |
+
| eval_rare_recall_at_8 | 0.052229 |
|
| 515 |
+
+-------------------------------+----------+
|
| 516 |
+
| eval_rare_precision_at_15 | 0.064216 |
|
| 517 |
+
+-------------------------------+----------+
|
| 518 |
+
| eval_rare_recall_at_15 | 0.069951 |
|
| 519 |
+
+-------------------------------+----------+
|
| 520 |
+
| eval_not_rare_f1_micro | 0.595588 |
|
| 521 |
+
+-------------------------------+----------+
|
| 522 |
+
| eval_not_rare_f1_macro | 0.373272 |
|
| 523 |
+
+-------------------------------+----------+
|
| 524 |
+
| eval_not_rare_precision | 0.595588 |
|
| 525 |
+
+-------------------------------+----------+
|
| 526 |
+
| eval_not_rare_recall | 0.595588 |
|
| 527 |
+
+-------------------------------+----------+
|
| 528 |
+
| eval_not_rare_precision_at_5 | 0.080882 |
|
| 529 |
+
+-------------------------------+----------+
|
| 530 |
+
| eval_not_rare_recall_at_5 | 0.404412 |
|
| 531 |
+
+-------------------------------+----------+
|
| 532 |
+
| eval_not_rare_precision_at_8 | 0.050551 |
|
| 533 |
+
+-------------------------------+----------+
|
| 534 |
+
| eval_not_rare_recall_at_8 | 0.404412 |
|
| 535 |
+
+-------------------------------+----------+
|
| 536 |
+
| eval_not_rare_precision_at_15 | 0.026961 |
|
| 537 |
+
+-------------------------------+----------+
|
| 538 |
+
| eval_not_rare_recall_at_15 | 0.404412 |
|
| 539 |
+
+-------------------------------+----------+
|
| 540 |
+
| eval_loss | 0.10503 |
|
| 541 |
+
+-------------------------------+----------+[0m - [multilabel_classify.py:1808:on_evaluate]
|
| 542 |
+
2025-06-03 20:27:10,158 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-68 - [multilabel_classify.py:2046:_save]
|
| 543 |
+
2025-06-03 20:27:10,161 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-68 - [multilabel_classify.py:2051:_save]
|
| 544 |
+
2025-06-03 20:27:10,163 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-68:
|
| 545 |
+
+---------+--------------------+------------+
|
| 546 |
+
| Index | Saved File | Size |
|
| 547 |
+
+=========+====================+============+
|
| 548 |
+
| 1 | training_args.bin | 0.01 MB |
|
| 549 |
+
+---------+--------------------+------------+
|
| 550 |
+
| 2 | optimizer.pt | 642.30 MB |
|
| 551 |
+
+---------+--------------------+------------+
|
| 552 |
+
| 3 | model.safetensors | 4267.74 MB |
|
| 553 |
+
+---------+--------------------+------------+
|
| 554 |
+
| 4 | scaler.pt | 0.00 MB |
|
| 555 |
+
+---------+--------------------+------------+
|
| 556 |
+
| 5 | config.json | 0.04 MB |
|
| 557 |
+
+---------+--------------------+------------+
|
| 558 |
+
| 6 | scheduler.pt | 0.00 MB |
|
| 559 |
+
+---------+--------------------+------------+
|
| 560 |
+
| 7 | trainer_state.json | 0.01 MB |
|
| 561 |
+
+---------+--------------------+------------+
|
| 562 |
+
| 8 | rng_state.pth | 0.01 MB |
|
| 563 |
+
+---------+--------------------+------------+ - [multilabel_classify.py:2068:_save]
|
| 564 |
+
2025-06-03 20:27:10,994 - INFO - 📂 Loading best model from ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-68 - [multilabel_classify.py:2120:_load_best_model]
|
| 565 |
+
2025-06-03 20:27:10,994 - INFO - 🖥️ Model is on device: cuda:0 - [multilabel_classify.py:2130:_load_best_model]
|
| 566 |
+
2025-06-03 20:27:11,049 - INFO - 🔑 Key order comparison:
|
| 567 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 568 |
+
| Index | Saved state_dict Keys | Model state_dict Keys |
|
| 569 |
+
+=========+============================================+======================================================================================+
|
| 570 |
+
| 1 | attention.in_proj_bias | boost_mul |
|
| 571 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 572 |
+
| 2 | attention.in_proj_weight | boost_add |
|
| 573 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 574 |
+
| 3 | attention.out_proj.bias | base_model.base_model.model.model.embed_tokens.weight |
|
| 575 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 576 |
+
| 4 | attention.out_proj.weight | base_model.base_model.model.model.layers.0.self_attn.q_proj.base_layer.weight |
|
| 577 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+
|
| 578 |
+
| 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 |
|
| 579 |
+
+---------+--------------------------------------------+--------------------------------------------------------------------------------------+ - [multilabel_classify.py:2154:_load_best_model]
|
| 580 |
+
2025-06-03 20:27:12,079 - INFO - ✅ Loaded best model weights from ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/checkpoint-68/model.safetensors - [multilabel_classify.py:2171:_load_best_model]
|
| 581 |
+
2025-06-03 20:27:12,105 - INFO - ✔️ Weight for boost_mul matches between saved and loaded state_dict - [multilabel_classify.py:2183:_load_best_model]
|
| 582 |
+
2025-06-03 20:27:12,129 - INFO - ✔️ Weight for boost_add matches between saved and loaded state_dict - [multilabel_classify.py:2183:_load_best_model]
|
| 583 |
+
2025-06-03 20:27:12,152 - INFO -
|
| 584 |
+
[36m🚂 Training Metrics (Step 68) 🚂
|
| 585 |
+
+--------------------------+---------+
|
| 586 |
+
| Metric | Value |
|
| 587 |
+
+==========================+=========+
|
| 588 |
+
| train_runtime | 289.486 |
|
| 589 |
+
+--------------------------+---------+
|
| 590 |
+
| train_samples_per_second | 7.835 |
|
| 591 |
+
+--------------------------+---------+
|
| 592 |
+
| train_steps_per_second | 0.235 |
|
| 593 |
+
+--------------------------+---------+
|
| 594 |
+
| total_flos | 0 |
|
| 595 |
+
+--------------------------+---------+
|
| 596 |
+
| train_loss | 0.21949 |
|
| 597 |
+
+--------------------------+---------+
|
| 598 |
+
| epoch | 3.78873 |
|
| 599 |
+
+--------------------------+---------+[0m - [multilabel_classify.py:1789:on_log]
|
| 600 |
+
2025-06-03 20:27:12,152 - INFO - ✨ Training Completed! ✨ - [multilabel_classify.py:1662:on_train_end]
|
| 601 |
+
2025-06-03 20:27:12,224 - INFO - 📊 Training loss plot saved as '../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/train_loss_plot.png' - [multilabel_classify.py:1858:on_train_end]
|
| 602 |
+
2025-06-03 20:27:12,286 - INFO - 📊 Evaluation loss plot saved as '../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/eval_loss_plot.png' - [multilabel_classify.py:1872:on_train_end]
|
| 603 |
+
2025-06-03 20:27:12,348 - INFO - 📊 Evaluation metric plot saved as '../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b/eval_precision_at_15_plot.png' - [multilabel_classify.py:1893:on_train_end]
|
| 604 |
+
2025-06-03 20:27:12,348 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:96:log_section]
|
| 605 |
+
2025-06-03 20:27:12,348 - INFO - + ✨ MODEL SAVING + - [multilabel_classify.py:97:log_section]
|
| 606 |
+
2025-06-03 20:27:12,348 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [multilabel_classify.py:100:log_section]
|
| 607 |
+
2025-06-03 20:27:12,349 - INFO - 💾 Saving trained model and pushing to Hugging Face Hub... - [multilabel_classify.py:3354:main]
|
| 608 |
+
2025-06-03 20:27:12,349 - INFO - 📁 Creating/using output directory: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2646:save_and_push]
|
| 609 |
+
2025-06-03 20:27:13,614 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2046:_save]
|
| 610 |
+
2025-06-03 20:27:13,617 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2051:_save]
|
| 611 |
+
2025-06-03 20:27:13,618 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b:
|
| 612 |
+
+---------+-------------------------------+------------+
|
| 613 |
+
| Index | Saved File | Size |
|
| 614 |
+
+=========+===============================+============+
|
| 615 |
+
| 1 | eval_loss_plot.png | 0.03 MB |
|
| 616 |
+
+---------+-------------------------------+------------+
|
| 617 |
+
| 2 | training_args.bin | 0.01 MB |
|
| 618 |
+
+---------+-------------------------------+------------+
|
| 619 |
+
| 3 | model.safetensors | 4267.74 MB |
|
| 620 |
+
+---------+-------------------------------+------------+
|
| 621 |
+
| 4 | config.json | 0.04 MB |
|
| 622 |
+
+---------+-------------------------------+------------+
|
| 623 |
+
| 5 | train_loss_plot.png | 0.02 MB |
|
| 624 |
+
+---------+-------------------------------+------------+
|
| 625 |
+
| 6 | eval_precision_at_15_plot.png | 0.03 MB |
|
| 626 |
+
+---------+-------------------------------+------------+ - [multilabel_classify.py:2068:_save]
|
| 627 |
+
2025-06-03 20:27:17,349 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2046:_save]
|
| 628 |
+
2025-06-03 20:27:17,352 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2051:_save]
|
| 629 |
+
2025-06-03 20:27:17,354 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b:
|
| 630 |
+
+---------+-------------------------------+------------+
|
| 631 |
+
| Index | Saved File | Size |
|
| 632 |
+
+=========+===============================+============+
|
| 633 |
+
| 1 | eval_loss_plot.png | 0.03 MB |
|
| 634 |
+
+---------+-------------------------------+------------+
|
| 635 |
+
| 2 | training_args.bin | 0.01 MB |
|
| 636 |
+
+---------+-------------------------------+------------+
|
| 637 |
+
| 3 | model.safetensors | 4267.74 MB |
|
| 638 |
+
+---------+-------------------------------+------------+
|
| 639 |
+
| 4 | config.json | 0.04 MB |
|
| 640 |
+
+---------+-------------------------------+------------+
|
| 641 |
+
| 5 | train_loss_plot.png | 0.02 MB |
|
| 642 |
+
+---------+-------------------------------+------------+
|
| 643 |
+
| 6 | eval_precision_at_15_plot.png | 0.03 MB |
|
| 644 |
+
+---------+-------------------------------+------------+ - [multilabel_classify.py:2068:_save]
|
| 645 |
+
2025-06-03 20:28:44,815 - INFO - 💾 Model saved to: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2650:save_and_push]
|
| 646 |
+
2025-06-03 20:28:44,845 - INFO - 🖌️ Tokenizer saved to: ../tmp/MIMIC4_DEMO/mimic4_classify_mistral7b - [multilabel_classify.py:2654:save_and_push]
|
tokenizer.json
CHANGED
|
@@ -1,19 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
-
"truncation":
|
| 4 |
-
|
| 5 |
-
"max_length": 512,
|
| 6 |
-
"strategy": "LongestFirst",
|
| 7 |
-
"stride": 0
|
| 8 |
-
},
|
| 9 |
-
"padding": {
|
| 10 |
-
"strategy": "BatchLongest",
|
| 11 |
-
"direction": "Left",
|
| 12 |
-
"pad_to_multiple_of": null,
|
| 13 |
-
"pad_id": 2,
|
| 14 |
-
"pad_type_id": 0,
|
| 15 |
-
"pad_token": "</s>"
|
| 16 |
-
},
|
| 17 |
"added_tokens": [
|
| 18 |
{
|
| 19 |
"id": 0,
|
|
|
|
| 1 |
{
|
| 2 |
"version": "1.0",
|
| 3 |
+
"truncation": null,
|
| 4 |
+
"padding": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"added_tokens": [
|
| 6 |
{
|
| 7 |
"id": 0,
|