training_l2r_log_2025-06-10_16-45-59.log

2025-06-10 16:45:59,389 - INFO - 📝 Logging initialized. Log file created at: ../tmp/logs/training_l2r_log_2025-06-10_16-45-59.log - [learning2rank.py:288:setup_logging]
2025-06-10 16:45:59,389 - INFO - ================================================================================ - [learning2rank.py:110:log_section]
2025-06-10 16:45:59,390 - INFO - =                      📌 INITIALIZING TRAINING ENVIRONMENT                       = - [learning2rank.py:111:log_section]
2025-06-10 16:45:59,390 - INFO - ================================================================================ - [learning2rank.py:114:log_section]
2025-06-10 16:45:59,390 - INFO - 🚀 Setting up data paths and environment variables... - [learning2rank.py:3920:main]
2025-06-10 16:45:59,390 - INFO - 🛠️ Command-line Arguments: - [learning2rank.py:304:print_args]
2025-06-10 16:45:59,390 - INFO - 
  🔹 output_dir: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b
  🔹 source_url: XURLs.MIMIC4_DEMO
  🔹 data: mimic4_icd10_full
  🔹 data_l2r_fname_prefix: mimic4_icd10
  🔹 logfile: training_l2r_log
  🔹 base_dir: ../tmp/MIMIC4_DEMO
  🔹 l2r_boot_dir: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b
  🔹 hub_model_id: deb101/mistral-7b-instruct-v0.3-mimic4-adapt
  🔹 model_name: mistralai/Mistral-7B-Instruct-v0.3
  🔹 max_length: 512
  🔹 do_fresh_training: True
  🔹 load_from_checkpoint: False
  🔹 task: l2r
  🔹 num_train_epochs: 2
  🔹 metric_for_best_model: ndcg@25
  🔹 learning_rate: 0.0001
  🔹 warmup_steps: 0
  🔹 generate_report: False
  🔹 logfile_path: ../tmp/logs/training_l2r_log_2025-06-10_16-45-59.log
  🔹 source: /home/ubuntu/.xcube/data/mimic4_demo - [learning2rank.py:305:print_args]
2025-06-10 16:45:59,390 - INFO - ➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖ - [learning2rank.py:306:print_args]
2025-06-10 16:45:59,391 - INFO - 📁 Using L2R boot directory: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b - [learning2rank.py:3934:main]
2025-06-10 16:45:59,391 - INFO - 📂 Using output directory: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3936:main]
2025-06-10 16:45:59,391 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:110:log_section]
2025-06-10 16:45:59,391 - INFO - +                               ✨ LOADING DATASETS                               + - [learning2rank.py:111:log_section]
2025-06-10 16:45:59,391 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:114:log_section]
2025-06-10 16:45:59,391 - INFO - 📊 Loading main dataset and L2R dataset... - [learning2rank.py:3939:main]
2025-06-10 16:45:59,391 - INFO - 📂 Loading main data from: /home/ubuntu/.xcube/data/mimic4_demo/mimic4_icd10_full.csv - [learning2rank.py:334:get_data]
2025-06-10 16:46:09,016 - INFO - ✅ Successfully loaded main data: 122279 rows - [learning2rank.py:347:get_data]
2025-06-10 16:46:09,017 - INFO - 📂 Loading L2R data from: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b/mimic4_icd10_tok_rank_per_lbl.ft - [learning2rank.py:353:get_data]
2025-06-10 16:46:09,017 - INFO - 📂 Loading L2R tokens from: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b/mimic4_icd10_tok.ft - [learning2rank.py:356:get_data]
2025-06-10 16:46:09,017 - INFO - 📂 Loading L2R labels from: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b/mimic4_icd10_lbl.ft - [learning2rank.py:359:get_data]
2025-06-10 16:46:14,824 - INFO - ✅ Successfully loaded L2R data: 260243456 rows - [learning2rank.py:366:get_data]
2025-06-10 16:46:14,824 - INFO - ✅ Successfully loaded L2R tokens: 32768 tokens - [learning2rank.py:367:get_data]
2025-06-10 16:46:14,824 - INFO - ✅ Successfully loaded L2R labels: 7942 rows - [learning2rank.py:368:get_data]
2025-06-10 16:46:14,824 - INFO - 🔄 Total data loaded: 122279 main rows, 260243456 L2R rows - [learning2rank.py:375:get_data]
2025-06-10 16:46:14,824 - INFO - ✨ Successfully loaded both datasets: - [learning2rank.py:399:load_datasets]
2025-06-10 16:46:14,825 - INFO -    - 📄 Main dataset: 122279 records - [learning2rank.py:400:load_datasets]
2025-06-10 16:46:14,825 - INFO -    - 📄 L2R dataset: 260243456 records - [learning2rank.py:401:load_datasets]
2025-06-10 16:46:14,825 - INFO -    - 🔤 Tokens: 32768 items - [learning2rank.py:402:load_datasets]
2025-06-10 16:46:14,825 - INFO -    - 🏷️ Labels: 7942 items - [learning2rank.py:403:load_datasets]
2025-06-10 16:46:14,835 - INFO - ✅ Data loading completed successfully - [learning2rank.py:411:load_datasets]
2025-06-10 16:46:16,280 - INFO - ⏳ Starting quantization of ranks for DataFrame with 260243456 rows. Containing 32768 unique tokens & 7942 unique labels - [learning2rank.py:530:quantize_l2r_data]
2025-06-10 16:46:16,281 - INFO - 🔄 Quantizing those 32768 unique token ranks into 101 quantization levels for each label - [learning2rank.py:555:quantize_l2r_data]
2025-06-10 16:47:24,678 - INFO - ✅ Completed quantization: Produced tensor of shape torch.Size([7942, 32768, 4]) with 101 quantization levels per label - [learning2rank.py:609:quantize_l2r_data]
2025-06-10 16:47:25,253 - WARNING - Label 1553: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:725:test_scored_tokens]
2025-06-10 16:47:25,263 - WARNING - Label 5088: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:725:test_scored_tokens]
2025-06-10 16:47:25,269 - WARNING - Label 6323: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:725:test_scored_tokens]
2025-06-10 16:47:25,275 - WARNING - Label 5572: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:725:test_scored_tokens]
2025-06-10 16:47:25,280 - WARNING - Label 4876: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:725:test_scored_tokens]
2025-06-10 16:47:25,453 - INFO - ******************************************************************************** - [learning2rank.py:110:log_section]
2025-06-10 16:47:25,454 - INFO - *                   🌟 STARTING LEARNING TO RANK MODEL TRAINING                   * - [learning2rank.py:111:log_section]
2025-06-10 16:47:25,454 - INFO - ******************************************************************************** - [learning2rank.py:114:log_section]
2025-06-10 16:47:25,454 - INFO - 🔐 Loaded authentication token from environment - [learning2rank.py:3968:main]
2025-06-10 16:47:25,454 - INFO - 🏷️ Hub Model ID for this Learning to Rank task: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-l2r - [learning2rank.py:3972:main]
2025-06-10 16:47:25,454 - INFO - -------------------------------------------------------------------------------- - [learning2rank.py:110:log_section]
2025-06-10 16:47:25,454 - INFO - -                            📋 MODEL EXISTENCE CHECK                             - - [learning2rank.py:111:log_section]
2025-06-10 16:47:25,454 - INFO - -------------------------------------------------------------------------------- - [learning2rank.py:114:log_section]
2025-06-10 16:47:25,454 - INFO - 🔍 Checking model existence locally and on Hugging Face Hub... - [learning2rank.py:431:check_model_existence]
2025-06-10 16:47:25,454 - INFO - ❌ Model not found locally at: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:438:check_model_existence]
2025-06-10 16:47:25,598 - INFO - ✅ Model exists on Hugging Face Hub with ID: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-l2r - [learning2rank.py:450:check_model_existence]
2025-06-10 16:47:25,598 - INFO - 📁 Model exists either locally or on Hub - [learning2rank.py:476:check_model_existence]
2025-06-10 16:47:25,598 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:110:log_section]
2025-06-10 16:47:25,598 - INFO - +                           ✨ STARTING FRESH TRAINING                            + - [learning2rank.py:111:log_section]
2025-06-10 16:47:25,598 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:114:log_section]
2025-06-10 16:47:25,598 - INFO - 🔄 Starting fresh training (either forced or model not found)... - [learning2rank.py:3985:main]
2025-06-10 16:47:25,669 - 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]
2025-06-10 16:47:25,669 - INFO - 🔑 Successfully authenticated with Hugging Face Hub - [learning2rank.py:312:authenticate_hf]
2025-06-10 16:47:25,669 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:110:log_section]
2025-06-10 16:47:25,669 - INFO - +                              ✨ LOADING BASE MODEL                              + - [learning2rank.py:111:log_section]
2025-06-10 16:47:25,669 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:114:log_section]
2025-06-10 16:47:25,670 - INFO - 📥 Loading pretrained model and tokenizer... - [learning2rank.py:4027:main]
2025-06-10 16:47:25,670 - INFO - 🚀 Starting model and tokenizer loading process... - [learning2rank.py:940:load_base_model_and_tokenizer]
2025-06-10 16:47:25,671 - INFO - 📊 Quantization config: BitsAndBytesConfig {
  "_load_in_4bit": true,
  "_load_in_8bit": false,
  "bnb_4bit_compute_dtype": "bfloat16",
  "bnb_4bit_quant_storage": "uint8",
  "bnb_4bit_quant_type": "nf4",
  "bnb_4bit_use_double_quant": true,
  "llm_int8_enable_fp32_cpu_offload": false,
  "llm_int8_has_fp16_weight": false,
  "llm_int8_skip_modules": null,
  "llm_int8_threshold": 6.0,
  "load_in_4bit": true,
  "load_in_8bit": false,
  "quant_method": "bitsandbytes"
}
 - [learning2rank.py:949:load_base_model_and_tokenizer]
2025-06-10 16:47:25,671 - INFO - 🔤 Loading tokenizer for model: mistralai/Mistral-7B-Instruct-v0.3... - [learning2rank.py:953:load_base_model_and_tokenizer]
2025-06-10 16:47:26,343 - INFO - 📝 Setting pad token to eos token... - [learning2rank.py:957:load_base_model_and_tokenizer]
2025-06-10 16:47:26,343 - INFO - 🧠 Loading base model with quantization... - [learning2rank.py:965:load_base_model_and_tokenizer]
2025-06-10 16:47:27,012 - 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]
2025-06-10 16:47:42,365 - INFO - 🔧 Setting up default LoRA configuration... - [learning2rank.py:988:load_base_model_and_tokenizer]
2025-06-10 16:47:42,365 - INFO - 🔍 LoRA config: r=16, alpha=32, targets={'o_proj', 'v_proj', 'q_proj', 'k_proj'}, dropout=0.05 - [learning2rank.py:1011:load_base_model_and_tokenizer]
2025-06-10 16:47:42,365 - INFO - 🧩 Applying LoRA adapters to model... - [learning2rank.py:1018:load_base_model_and_tokenizer]
2025-06-10 16:47:42,588 - INFO - 📊 trainable params: 13,631,488 || all params: 7,261,655,040 || trainable%: 0.1877 - [learning2rank.py:136:log_print_output]
2025-06-10 16:47:42,588 - INFO - ✅ Model and tokenizer successfully loaded! - [learning2rank.py:1025:load_base_model_and_tokenizer]
2025-06-10 16:47:42,588 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:110:log_section]
2025-06-10 16:47:42,588 - INFO - +                              ✨ DATA PREPROCESSING                              + - [learning2rank.py:111:log_section]
2025-06-10 16:47:42,589 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:114:log_section]
2025-06-10 16:47:42,589 - INFO - 🔄 Loading and preprocessing training data... - [learning2rank.py:4035:main]
2025-06-10 16:47:44,687 - INFO - 🔍 Verifying uniqueness of token IDs per label in scored_tokens... - [learning2rank.py:1303:preprocess_data]
2025-06-10 16:47:45,783 - INFO - ✅ All labels have unique token IDs in scored_tokens. - [learning2rank.py:1315:preprocess_data]
2025-06-10 16:47:45,783 - INFO - 🚀 Building a 2D lookup table for efficient token-to-relevance mapping across all labels! 🚀 - [learning2rank.py:1318:preprocess_data]
2025-06-10 16:47:45,784 - INFO - 🔢 Total labels = 7942 - [learning2rank.py:1321:preprocess_data]
2025-06-10 16:47:45,784 - INFO - 🔍 Precomputing token indices and corresponding relevance_values for each label... - [learning2rank.py:1322:preprocess_data]
2025-06-10 16:47:45,792 - INFO - 📊 Lookup table dimensions: 32768 vocabulary size × 7942 labels - [learning2rank.py:1329:preprocess_data]
2025-06-10 16:47:45,792 - INFO - ⚡ This approach eliminates token comparison broadcasting and provides O(1) lookup time for relevance scores! - [learning2rank.py:1332:preprocess_data]
2025-06-10 16:47:45,792 - INFO - 🧮 Processing relevance calculations vectorized for maximum speed 🔥 - [learning2rank.py:1335:preprocess_data]
2025-06-10 16:47:45,970 - INFO - 🔍 Verifying token mappings with 10 samples... - [learning2rank.py:1368:verify_token_mappings]
2025-06-10 16:47:46,153 - INFO - ✅ Token mappings verification completed successfully! 🎉 - [learning2rank.py:1459:verify_token_mappings]
2025-06-10 16:47:46,160 - INFO - 🔄 Processing dataset with map... - [learning2rank.py:1523:preprocess_data]
2025-06-10 16:47:46,545 - INFO - ✅ Dataset built in 0h 0m 0.39s - [learning2rank.py:1546:preprocess_data]
2025-06-10 16:47:46,560 - INFO - The size of Training set: 173 🏋️ (Training Data) - [learning2rank.py:1577:preprocess_data]
2025-06-10 16:47:46,560 - INFO - The size of Evaluation set: 35 🔍 (Test Data) - [learning2rank.py:1578:preprocess_data]
2025-06-10 16:47:46,560 - INFO - 🚀 Created HuggingFace Dataset with 208 samples, 7942 labels - [learning2rank.py:1586:preprocess_data]
2025-06-10 16:47:46,561 - INFO - 🏷️ Number of unique ICD-10 codes: 7942 - [learning2rank.py:4048:main]
2025-06-10 16:47:46,561 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:110:log_section]
2025-06-10 16:47:46,561 - INFO - +                             ✨ MODEL INITIALIZATION                             + - [learning2rank.py:111:log_section]
2025-06-10 16:47:46,561 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:114:log_section]
2025-06-10 16:47:46,561 - INFO - 🧠 Initializing custom L2R model for outputting per-token relevance scores per ICD-10 codes. - [learning2rank.py:4051:main]
2025-06-10 16:47:46,964 - INFO - 🔧 Registering LTRModel with transformers AutoModel 🚀 - [learning2rank.py:1902:define_model]
2025-06-10 16:47:46,964 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:110:log_section]
2025-06-10 16:47:46,964 - INFO - +                             ✨ TRAINING PREPARATION                             + - [learning2rank.py:111:log_section]
2025-06-10 16:47:46,964 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:114:log_section]
2025-06-10 16:47:46,964 - INFO - ⚙️ Preparing training components and optimizers... - [learning2rank.py:4058:main]
2025-06-10 16:47:47,038 - INFO - 🖥️ Device: NVIDIA A100 80GB PCIe - [learning2rank.py:2064:log_training_configuration]
2025-06-10 16:47:47,038 - INFO - 🔋 CUDA Available: True - [learning2rank.py:2067:log_training_configuration]
2025-06-10 16:47:47,038 - INFO - 💾 CUDA Device Count: 1 - [learning2rank.py:2068:log_training_configuration]
2025-06-10 16:47:47,039 - INFO - 
📋 Training Configuration 📋
+----------+-----------------------------+--------------------------------------------------+
| 🌟 Emoji |        🏷️ Parameter         |                     📊 Value                     |
+----------+-----------------------------+--------------------------------------------------+
|    📁    |      Output Directory       |   ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b   |
|    🔁    |       Training Epochs       |                        2                         |
|    🏋️    |      Train Batch Size       |                        1                         |
|    🔍    |       Eval Batch Size       |                        1                         |
|    📊    | Gradient Accumulation Steps |                        4                         |
|    🚀    |        Learning Rate        |                      0.0001                      |
|    🌅    |        Warmup Steps         |                        0                         |
|    💾    |        Save Strategy        |                      epoch                       |
|    💾    |      Save Total Limit       |                        10                        |
|    📊    |     Evaluation Strategy     |                      epoch                       |
|    🎯    |      Best Model Metric      |                     ndcg@25                      |
|    📝    |      Logging Strategy       |              steps (every 10 steps)              |
|    🌐    |         Push to Hub         |                       True                       |
|    🌐    |        Hub Model ID         | deb101/mistral-7b-instruct-v0.3-mimic4-adapt-l2r |
|    🔢    |       Steps per Epoch       |                        43                        |
|    🔢    |    Total Training Steps     |                        86                        |
|    🔢    |      Evaluation Steps       |                        35                        |
|    📊    |    Training Dataset Size    |                  173 samples 🏋️                  |
|    📊    |   Evaluation Dataset Size   |                  35 samples 🔍                   |
+----------+-----------------------------+--------------------------------------------------+ - [learning2rank.py:2056:log_training_args]
2025-06-10 16:47:47,039 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:110:log_section]
2025-06-10 16:47:47,039 - INFO - +                                ✨ MODEL TRAINING                                + - [learning2rank.py:111:log_section]
2025-06-10 16:47:47,039 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:114:log_section]
2025-06-10 16:47:47,039 - INFO - 🏋️ Starting model training process... - [learning2rank.py:4078:main]
2025-06-10 16:47:47,040 - INFO - 🏁 Preparing Custom Trainer 🛠️ - [learning2rank.py:3257:train_model]
2025-06-10 16:47:47,097 - INFO - We are registering the tokenizer mistralai/Mistral-7B-Instruct-v0.3 in Custom Trainer - [learning2rank.py:2696:__init__]
2025-06-10 16:47:47,097 - INFO - 🏋️ Commencing Model Training 💪 - [learning2rank.py:3298:train_model]
2025-06-10 16:47:47,441 - INFO - 🚀 Starting Training... - [learning2rank.py:2446:on_train_begin]
2025-06-10 16:48:09,541 - INFO - 
🚂 Training Metrics (Step 10) 🚂
+---------------+--------------+
| Metric        |        Value |
+===============+==============+
| loss          | -4.73061e+16 |
+---------------+--------------+
| grad_norm     |          nan |
+---------------+--------------+
| learning_rate |       0.0001 |
+---------------+--------------+
| epoch         |     0.231214 |
+---------------+--------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:48:31,336 - INFO - 
🚂 Training Metrics (Step 20) 🚂
+---------------+--------------+
| Metric        |        Value |
+===============+==============+
| loss          | -3.42152e+17 |
+---------------+--------------+
| grad_norm     |          nan |
+---------------+--------------+
| learning_rate |       0.0001 |
+---------------+--------------+
| epoch         |     0.462428 |
+---------------+--------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:48:53,161 - INFO - 
🚂 Training Metrics (Step 30) 🚂
+---------------+-------------+
| Metric        |       Value |
+===============+=============+
| loss          | -6.4721e+16 |
+---------------+-------------+
| grad_norm     |         nan |
+---------------+-------------+
| learning_rate |      0.0001 |
+---------------+-------------+
| epoch         |    0.693642 |
+---------------+-------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:49:14,996 - INFO - 
🚂 Training Metrics (Step 40) 🚂
+---------------+-------------+
| Metric        |       Value |
+===============+=============+
| loss          | -2.3308e+18 |
+---------------+-------------+
| grad_norm     |         nan |
+---------------+-------------+
| learning_rate |      0.0001 |
+---------------+-------------+
| epoch         |    0.924855 |
+---------------+-------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:49:22,093 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [learning2rank.py:2812:evaluate]
2025-06-10 16:51:02,814 - INFO - 
🔍 Evaluation Metrics 🔍
+-------------------+--------------+
| Metric            |        Value |
+===================+==============+
| eval_loss         | -9.27285e+16 |
+-------------------+--------------+
| eval_ndcg         |     0.955594 |
+-------------------+--------------+
| eval_ndcg@25      |     0.201397 |
+-------------------+--------------+
| eval_precision@25 |         0.04 |
+-------------------+--------------+ - [learning2rank.py:2383:on_evaluate]
2025-06-10 16:51:10,862 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-44 - [learning2rank.py:2936:_save]
2025-06-10 16:51:10,896 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-44 - [learning2rank.py:2941:_save]
2025-06-10 16:51:10,900 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-44:
+---------+-------------------+------------+
|   Index | Saved File        | Size       |
+=========+===================+============+
|       1 | training_args.bin | 0.01 MB    |
+---------+-------------------+------------+
|       2 | config.json       | 0.38 MB    |
+---------+-------------------+------------+
|       3 | model.safetensors | 4122.74 MB |
+---------+-------------------+------------+ - [learning2rank.py:2958:_save]
2025-06-10 16:51:23,955 - INFO - 
🚂 Training Metrics (Step 50) 🚂
+---------------+--------------+
| Metric        |        Value |
+===============+==============+
| loss          | -1.39594e+18 |
+---------------+--------------+
| grad_norm     |          nan |
+---------------+--------------+
| learning_rate |       0.0001 |
+---------------+--------------+
| epoch         |      1.13873 |
+---------------+--------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:51:45,870 - INFO - 
🚂 Training Metrics (Step 60) 🚂
+---------------+--------------+
| Metric        |        Value |
+===============+==============+
| loss          | -1.11813e+17 |
+---------------+--------------+
| grad_norm     |          nan |
+---------------+--------------+
| learning_rate |      9.7e-05 |
+---------------+--------------+
| epoch         |      1.36994 |
+---------------+--------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:52:07,824 - INFO - 
🚂 Training Metrics (Step 70) 🚂
+---------------+--------------+
| Metric        |        Value |
+===============+==============+
| loss          | -6.66649e+18 |
+---------------+--------------+
| grad_norm     |  2.15935e+18 |
+---------------+--------------+
| learning_rate |      8.7e-05 |
+---------------+--------------+
| epoch         |      1.60116 |
+---------------+--------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:52:29,799 - INFO - 
🚂 Training Metrics (Step 80) 🚂
+---------------+--------------+
| Metric        |        Value |
+===============+==============+
| loss          | -9.84298e+16 |
+---------------+--------------+
| grad_norm     |  1.43927e+18 |
+---------------+--------------+
| learning_rate |      7.6e-05 |
+---------------+--------------+
| epoch         |      1.83237 |
+---------------+--------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:52:51,795 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2936:_save]
2025-06-10 16:52:51,829 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2941:_save]
2025-06-10 16:52:51,831 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86:
+---------+-------------------+------------+
|   Index | Saved File        | Size       |
+=========+===================+============+
|       1 | training_args.bin | 0.01 MB    |
+---------+-------------------+------------+
|       2 | config.json       | 0.38 MB    |
+---------+-------------------+------------+
|       3 | model.safetensors | 4122.74 MB |
+---------+-------------------+------------+ - [learning2rank.py:2958:_save]
2025-06-10 16:52:52,469 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [learning2rank.py:2812:evaluate]
2025-06-10 16:54:33,647 - INFO - 
🔍 Evaluation Metrics 🔍
+-------------------+--------------+
| Metric            |        Value |
+===================+==============+
| eval_loss         | -2.40602e+17 |
+-------------------+--------------+
| eval_ndcg         |     0.956305 |
+-------------------+--------------+
| eval_ndcg@25      |     0.442205 |
+-------------------+--------------+
| eval_precision@25 |     0.221429 |
+-------------------+--------------+ - [learning2rank.py:2383:on_evaluate]
2025-06-10 16:54:45,111 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2936:_save]
2025-06-10 16:54:45,151 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2941:_save]
2025-06-10 16:54:45,153 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86:
+---------+--------------------+------------+
|   Index | Saved File         | Size       |
+=========+====================+============+
|       1 | trainer_state.json | 0.00 MB    |
+---------+--------------------+------------+
|       2 | training_args.bin  | 0.01 MB    |
+---------+--------------------+------------+
|       3 | config.json        | 0.38 MB    |
+---------+--------------------+------------+
|       4 | model.safetensors  | 4122.74 MB |
+---------+--------------------+------------+
|       5 | scheduler.pt       | 0.00 MB    |
+---------+--------------------+------------+
|       6 | scaler.pt          | 0.00 MB    |
+---------+--------------------+------------+
|       7 | rng_state.pth      | 0.01 MB    |
+---------+--------------------+------------+
|       8 | optimizer.pt       | 352.39 MB  |
+---------+--------------------+------------+ - [learning2rank.py:2958:_save]
2025-06-10 16:54:46,066 - INFO - 📂 Loading best model from ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:3014:_load_best_model]
2025-06-10 16:54:46,066 - INFO - 🖥️ Model is on device: cuda:0 - [learning2rank.py:3024:_load_best_model]
2025-06-10 16:54:46,136 - INFO - 🔑 Key order comparison:
+---------+--------------------------------------------------------------------------+-------------------------------------------------------------------------------------------+
|   Index | Saved state_dict Keys                                                    | Model state_dict Keys                                                                     |
+=========+==========================================================================+===========================================================================================+
|       1 | ground_model.base_model.model.lm_head.weight                             | label_embeddings.weight                                                                   |
+---------+--------------------------------------------------------------------------+-------------------------------------------------------------------------------------------+
|       2 | ground_model.base_model.model.model.embed_tokens.weight                  | ground_model.base_model.model.model.embed_tokens.weight                                   |
+---------+--------------------------------------------------------------------------+-------------------------------------------------------------------------------------------+
|       3 | ground_model.base_model.model.model.layers.0.input_layernorm.weight      | ground_model.base_model.model.model.layers.0.self_attn.q_proj.base_layer.weight           |
+---------+--------------------------------------------------------------------------+-------------------------------------------------------------------------------------------+
|       4 | ground_model.base_model.model.model.layers.0.mlp.down_proj.weight        | ground_model.base_model.model.model.layers.0.self_attn.q_proj.base_layer.weight.absmax    |
+---------+--------------------------------------------------------------------------+-------------------------------------------------------------------------------------------+
|       5 | ground_model.base_model.model.model.layers.0.mlp.down_proj.weight.absmax | ground_model.base_model.model.model.layers.0.self_attn.q_proj.base_layer.weight.quant_map |
+---------+--------------------------------------------------------------------------+-------------------------------------------------------------------------------------------+ - [learning2rank.py:3048:_load_best_model]
2025-06-10 16:54:46,827 - INFO - ✅ Loaded best model weights from ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86/model.safetensors - [learning2rank.py:3065:_load_best_model]
2025-06-10 16:54:46,869 - INFO - ✔️ Weight for label_embeddings.weight matches between saved and loaded state_dict - [learning2rank.py:3077:_load_best_model]
2025-06-10 16:54:46,913 - INFO - ✔️ Weight for ground_model.base_model.model.model.embed_tokens.weight matches between saved and loaded state_dict - [learning2rank.py:3077:_load_best_model]
2025-06-10 16:54:47,092 - INFO - 
🚂 Training Metrics (Step 86) 🚂
+--------------------------+--------------+
| Metric                   |        Value |
+==========================+==============+
| train_runtime            |      419.653 |
+--------------------------+--------------+
| train_samples_per_second |        0.824 |
+--------------------------+--------------+
| train_steps_per_second   |        0.205 |
+--------------------------+--------------+
| total_flos               |  3.81076e+15 |
+--------------------------+--------------+
| train_loss               | -1.35419e+18 |
+--------------------------+--------------+
| epoch                    |       1.9711 |
+--------------------------+--------------+ - [learning2rank.py:2364:on_log]
2025-06-10 16:54:47,094 - INFO - ✨ Training Completed! ✨ - [learning2rank.py:2513:on_train_end]
2025-06-10 16:54:47,223 - INFO - 📊 Training loss plot saved as '../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/train_loss_plot.png' - [learning2rank.py:2616:on_train_end]
2025-06-10 16:54:47,315 - INFO - 📊 Evaluation loss plot saved as '../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/eval_loss_plot.png' - [learning2rank.py:2630:on_train_end]
2025-06-10 16:54:47,405 - INFO - 📊 Evaluation metric plot saved as '../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/eval_ndcg@25_plot.png' - [learning2rank.py:2651:on_train_end]
2025-06-10 16:54:47,406 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:110:log_section]
2025-06-10 16:54:47,406 - INFO - +                                 ✨ MODEL SAVING                                 + - [learning2rank.py:111:log_section]
2025-06-10 16:54:47,406 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:114:log_section]
2025-06-10 16:54:47,406 - INFO - 💾 Saving trained model and pushing to Hugging Face Hub... - [learning2rank.py:4093:main]
2025-06-10 16:54:56,139 - INFO - 📁 Creating/using output directory: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3332:save_and_push]
2025-06-10 16:55:04,924 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:2936:_save]
2025-06-10 16:55:04,958 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:2941:_save]
2025-06-10 16:55:04,959 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b:
+---------+-----------------------+------------+
|   Index | Saved File            | Size       |
+=========+=======================+============+
|       1 | training_args.bin     | 0.01 MB    |
+---------+-----------------------+------------+
|       2 | eval_ndcg@25_plot.png | 0.03 MB    |
+---------+-----------------------+------------+
|       3 | eval_loss_plot.png    | 0.03 MB    |
+---------+-----------------------+------------+
|       4 | config.json           | 0.38 MB    |
+---------+-----------------------+------------+
|       5 | model.safetensors     | 4122.74 MB |
+---------+-----------------------+------------+
|       6 | train_loss_plot.png   | 0.03 MB    |
+---------+-----------------------+------------+ - [learning2rank.py:2958:_save]
2025-06-10 16:55:14,012 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:2936:_save]
2025-06-10 16:55:14,046 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:2941:_save]
2025-06-10 16:55:14,048 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b:
+---------+-----------------------+------------+
|   Index | Saved File            | Size       |
+=========+=======================+============+
|       1 | training_args.bin     | 0.01 MB    |
+---------+-----------------------+------------+
|       2 | eval_ndcg@25_plot.png | 0.03 MB    |
+---------+-----------------------+------------+
|       3 | eval_loss_plot.png    | 0.03 MB    |
+---------+-----------------------+------------+
|       4 | config.json           | 0.38 MB    |
+---------+-----------------------+------------+
|       5 | model.safetensors     | 4122.74 MB |
+---------+-----------------------+------------+
|       6 | train_loss_plot.png   | 0.03 MB    |
+---------+-----------------------+------------+ - [learning2rank.py:2958:_save]
2025-06-10 16:56:44,068 - INFO - 💾 Model saved to: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3336:save_and_push]
2025-06-10 16:56:44,106 - INFO - 🖌️ Tokenizer saved to: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3340:save_and_push]
2025-06-10 16:56:45,539 - INFO - 📊 Lookup table saved to: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/lookup_table.pt - [learning2rank.py:3345:save_and_push]