| 2025-06-11 15:29:36,035 - INFO - 📝 Logging initialized. Log file created at: ../tmp/logs/training_l2r_log_2025-06-11_15-29-36.log - [learning2rank.py:287:setup_logger] | |
| 2025-06-11 15:29:36,035 - INFO - ================================================================================ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:29:36,036 - INFO - = 📌 INITIALIZING TRAINING ENVIRONMENT = - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:29:36,036 - INFO - ================================================================================ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:29:36,036 - INFO - 🚀 Setting up data paths and environment variables... - [learning2rank.py:3751:main] | |
| 2025-06-11 15:29:36,036 - INFO - 🛠️ Command-line Arguments: - [learning2rank.py:303:print_args] | |
| 2025-06-11 15:29:36,036 - 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_2025-06-11_15-29-36.log | |
| 🔹 source: /home/ubuntu/.xcube/data/mimic4_demo - [learning2rank.py:304:print_args] | |
| 2025-06-11 15:29:36,036 - INFO - ➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖➖ - [learning2rank.py:305:print_args] | |
| 2025-06-11 15:29:36,036 - INFO - 📁 Using L2R boot directory: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b - [learning2rank.py:3765:main] | |
| 2025-06-11 15:29:36,036 - INFO - 📂 Using output directory: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3767:main] | |
| 2025-06-11 15:29:36,037 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:29:36,037 - INFO - + ✨ LOADING DATASETS + - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:29:36,037 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:29:36,037 - INFO - 📊 Loading main dataset and L2R dataset... - [learning2rank.py:3770:main] | |
| 2025-06-11 15:29:36,037 - INFO - 📂 Loading main data from: /home/ubuntu/.xcube/data/mimic4_demo/mimic4_icd10_full.csv - [learning2rank.py:333:get_data] | |
| 2025-06-11 15:29:44,310 - INFO - ✅ Successfully loaded main data: 122279 rows - [learning2rank.py:346:get_data] | |
| 2025-06-11 15:29:44,310 - INFO - 📂 Loading L2R data from: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b/mimic4_icd10_tok_rank_per_lbl.ft - [learning2rank.py:352:get_data] | |
| 2025-06-11 15:29:44,310 - INFO - 📂 Loading L2R tokens from: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b/mimic4_icd10_tok.ft - [learning2rank.py:355:get_data] | |
| 2025-06-11 15:29:44,310 - INFO - 📂 Loading L2R labels from: ../tmp/MIMIC4_DEMO/mimic4_l2rboot_mistral7b/mimic4_icd10_lbl.ft - [learning2rank.py:358:get_data] | |
| 2025-06-11 15:29:46,024 - INFO - ✅ Successfully loaded L2R data: 260243456 rows - [learning2rank.py:365:get_data] | |
| 2025-06-11 15:29:46,024 - INFO - ✅ Successfully loaded L2R tokens: 32768 tokens - [learning2rank.py:366:get_data] | |
| 2025-06-11 15:29:46,024 - INFO - ✅ Successfully loaded L2R labels: 7942 rows - [learning2rank.py:367:get_data] | |
| 2025-06-11 15:29:46,024 - INFO - 🔄 Total data loaded: 122279 main rows, 260243456 L2R rows - [learning2rank.py:374:get_data] | |
| 2025-06-11 15:29:46,025 - INFO - ✨ Successfully loaded both datasets: - [learning2rank.py:398:load_datasets] | |
| 2025-06-11 15:29:46,025 - INFO - - 📄 Main dataset: 122279 records - [learning2rank.py:399:load_datasets] | |
| 2025-06-11 15:29:46,025 - INFO - - 📄 L2R dataset: 260243456 records - [learning2rank.py:400:load_datasets] | |
| 2025-06-11 15:29:46,025 - INFO - - 🔤 Tokens: 32768 items - [learning2rank.py:401:load_datasets] | |
| 2025-06-11 15:29:46,025 - INFO - - 🏷️ Labels: 7942 items - [learning2rank.py:402:load_datasets] | |
| 2025-06-11 15:29:46,032 - INFO - ✅ Data loading completed successfully - [learning2rank.py:410:load_datasets] | |
| 2025-06-11 15:29:47,314 - INFO - ⏳ Starting quantization of ranks for DataFrame with 260243456 rows. Containing 32768 unique tokens & 7942 unique labels - [learning2rank.py:529:quantize_l2r_data] | |
| 2025-06-11 15:29:47,314 - INFO - 🔄 Quantizing those 32768 unique token ranks into 101 quantization levels for each label - [learning2rank.py:554:quantize_l2r_data] | |
| 2025-06-11 15:30:33,284 - INFO - ✅ Completed quantization: Produced tensor of shape torch.Size([7942, 32768, 4]) with 101 quantization levels per label - [learning2rank.py:608:quantize_l2r_data] | |
| 2025-06-11 15:30:33,309 - WARNING - Label 457: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:724:test_scored_tokens] | |
| 2025-06-11 15:30:33,314 - WARNING - Label 2871: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:724:test_scored_tokens] | |
| 2025-06-11 15:30:33,318 - WARNING - Label 4330: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:724:test_scored_tokens] | |
| 2025-06-11 15:30:33,322 - WARNING - Label 3865: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:724:test_scored_tokens] | |
| 2025-06-11 15:30:33,325 - WARNING - Label 6626: Only 1 tokens with top relevance score (need 50) - [learning2rank.py:724:test_scored_tokens] | |
| 2025-06-11 15:30:33,353 - INFO - ******************************************************************************** - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:30:33,353 - INFO - * 🌟 STARTING LEARNING TO RANK MODEL TRAINING * - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:30:33,353 - INFO - ******************************************************************************** - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:30:33,353 - INFO - 🔐 Loaded authentication token from environment - [learning2rank.py:3799:main] | |
| 2025-06-11 15:30:33,353 - INFO - 🏷️ Hub Model ID for this Learning to Rank task: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-l2r - [learning2rank.py:3803:main] | |
| 2025-06-11 15:30:33,353 - INFO - -------------------------------------------------------------------------------- - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:30:33,353 - INFO - - 📋 MODEL EXISTENCE CHECK - - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:30:33,353 - INFO - -------------------------------------------------------------------------------- - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:30:33,353 - INFO - 🔍 Checking model existence locally and on Hugging Face Hub... - [learning2rank.py:430:check_model_existence] | |
| 2025-06-11 15:30:33,354 - INFO - ❌ Model not found locally at: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:437:check_model_existence] | |
| 2025-06-11 15:30:33,411 - INFO - ✅ Model exists on Hugging Face Hub with ID: deb101/mistral-7b-instruct-v0.3-mimic4-adapt-l2r - [learning2rank.py:449:check_model_existence] | |
| 2025-06-11 15:30:33,411 - INFO - 📁 Model exists either locally or on Hub - [learning2rank.py:475:check_model_existence] | |
| 2025-06-11 15:30:33,411 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:30:33,411 - INFO - + ✨ STARTING FRESH TRAINING + - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:30:33,411 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:30:33,411 - INFO - 🔄 Starting fresh training (either forced or model not found)... - [learning2rank.py:3816:main] | |
| 2025-06-11 15:30:33,681 - 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] | |
| 2025-06-11 15:30:33,681 - INFO - 🔑 Successfully authenticated with Hugging Face Hub - [learning2rank.py:311:authenticate_hf] | |
| 2025-06-11 15:30:33,681 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:30:33,681 - INFO - + ✨ LOADING BASE MODEL + - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:30:33,681 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:30:33,681 - INFO - 📥 Loading pretrained model and tokenizer... - [learning2rank.py:3858:main] | |
| 2025-06-11 15:30:33,681 - INFO - 🚀 Starting model and tokenizer loading process... - [learning2rank.py:939:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:33,682 - 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:948:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:33,682 - INFO - 🔤 Loading tokenizer for model: mistralai/Mistral-7B-Instruct-v0.3... - [learning2rank.py:952:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:33,981 - INFO - 📝 Setting pad token to eos token... - [learning2rank.py:956:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:33,981 - INFO - 🧠 Loading base model with quantization... - [learning2rank.py:964:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:34,497 - 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] | |
| 2025-06-11 15:30:39,731 - INFO - 🔧 Setting up default LoRA configuration... - [learning2rank.py:987:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:39,731 - INFO - 🔍 LoRA config: r=16, alpha=32, targets={'k_proj', 'q_proj', 'o_proj', 'v_proj'}, dropout=0.05 - [learning2rank.py:1010:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:39,731 - INFO - 🧩 Applying LoRA adapters to model... - [learning2rank.py:1017:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:39,906 - INFO - 📊 trainable params: 13,631,488 || all params: 7,261,655,040 || trainable%: 0.1877 - [learning2rank.py:135:log_print_output] | |
| 2025-06-11 15:30:39,907 - INFO - ✅ Model and tokenizer successfully loaded! - [learning2rank.py:1024:load_base_model_and_tokenizer] | |
| 2025-06-11 15:30:39,907 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:30:39,907 - INFO - + ✨ DATA PREPROCESSING + - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:30:39,907 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:30:39,907 - INFO - 🔄 Loading and preprocessing training data... - [learning2rank.py:3866:main] | |
| 2025-06-11 15:30:40,975 - INFO - 🔍 Verifying uniqueness of token IDs per label in scored_tokens... - [learning2rank.py:1302:preprocess_data] | |
| 2025-06-11 15:30:41,803 - INFO - ✅ All labels have unique token IDs in scored_tokens. - [learning2rank.py:1314:preprocess_data] | |
| 2025-06-11 15:30:41,803 - INFO - 🚀 Building a 2D lookup table for efficient token-to-relevance mapping across all labels! 🚀 - [learning2rank.py:1317:preprocess_data] | |
| 2025-06-11 15:30:41,803 - INFO - 🔢 Total labels = 7942 - [learning2rank.py:1320:preprocess_data] | |
| 2025-06-11 15:30:41,803 - INFO - 🔍 Precomputing token indices and corresponding relevance_values for each label... - [learning2rank.py:1321:preprocess_data] | |
| 2025-06-11 15:30:42,036 - INFO - 📊 Lookup table dimensions: 32768 vocabulary size × 7942 labels - [learning2rank.py:1328:preprocess_data] | |
| 2025-06-11 15:30:42,036 - INFO - ⚡ This approach eliminates token comparison broadcasting and provides O(1) lookup time for relevance scores! - [learning2rank.py:1331:preprocess_data] | |
| 2025-06-11 15:30:42,036 - INFO - 🧮 Processing relevance calculations vectorized for maximum speed 🔥 - [learning2rank.py:1334:preprocess_data] | |
| 2025-06-11 15:30:42,202 - INFO - 🔍 Verifying token mappings with 10 samples... - [learning2rank.py:1367:verify_token_mappings] | |
| 2025-06-11 15:30:42,375 - INFO - ✅ Token mappings verification completed successfully! 🎉 - [learning2rank.py:1458:verify_token_mappings] | |
| 2025-06-11 15:30:42,380 - INFO - 🔄 Processing dataset with map... - [learning2rank.py:1522:preprocess_data] | |
| 2025-06-11 15:30:42,697 - INFO - ✅ Dataset built in 0h 0m 0.32s - [learning2rank.py:1545:preprocess_data] | |
| 2025-06-11 15:30:42,709 - INFO - The size of Training set: 173 🏋️ (Training Data) - [learning2rank.py:1576:preprocess_data] | |
| 2025-06-11 15:30:42,709 - INFO - The size of Evaluation set: 35 🔍 (Test Data) - [learning2rank.py:1577:preprocess_data] | |
| 2025-06-11 15:30:42,709 - INFO - 🚀 Created HuggingFace Dataset with 208 samples, 7942 labels - [learning2rank.py:1585:preprocess_data] | |
| 2025-06-11 15:30:42,710 - INFO - 🏷️ Number of unique ICD-10 codes: 7942 - [learning2rank.py:3879:main] | |
| 2025-06-11 15:30:42,710 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:30:42,710 - INFO - + ✨ MODEL INITIALIZATION + - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:30:42,710 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:30:42,710 - INFO - 🧠 Initializing custom L2R model for outputting per-token relevance scores per ICD-10 codes. - [learning2rank.py:3882:main] | |
| 2025-06-11 15:30:44,257 - INFO - 🔧 Registering LTRModel with transformers AutoModel 🚀 - [learning2rank.py:1725:define_model] | |
| 2025-06-11 15:30:44,258 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:30:44,258 - INFO - + ✨ TRAINING PREPARATION + - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:30:44,258 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:30:44,258 - INFO - ⚙️ Preparing training components and optimizers... - [learning2rank.py:3889:main] | |
| 2025-06-11 15:30:44,289 - INFO - 🖥️ Device: NVIDIA GH200 480GB - [learning2rank.py:1887:log_training_configuration] | |
| 2025-06-11 15:30:44,289 - INFO - 🔋 CUDA Available: True - [learning2rank.py:1890:log_training_configuration] | |
| 2025-06-11 15:30:44,289 - INFO - 💾 CUDA Device Count: 1 - [learning2rank.py:1891:log_training_configuration] | |
| 2025-06-11 15:30:44,290 - 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:1879:log_training_args] | |
| 2025-06-11 15:30:44,290 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:30:44,290 - INFO - + ✨ MODEL TRAINING + - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:30:44,290 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:30:44,290 - INFO - 🏋️ Starting model training process... - [learning2rank.py:3909:main] | |
| 2025-06-11 15:30:44,290 - INFO - 🏁 Preparing Custom Trainer 🛠️ - [learning2rank.py:3080:train_model] | |
| 2025-06-11 15:30:44,329 - INFO - We are registering the tokenizer mistralai/Mistral-7B-Instruct-v0.3 in Custom Trainer - [learning2rank.py:2519:__init__] | |
| 2025-06-11 15:30:44,329 - INFO - 🏋️ Commencing Model Training 💪 - [learning2rank.py:3121:train_model] | |
| 2025-06-11 15:30:44,595 - INFO - 🚀 Starting Training... - [learning2rank.py:2269:on_train_begin] | |
| 2025-06-11 15:31:06,023 - INFO - | |
| [36m🚂 Training Metrics (Step 10) 🚂 | |
| +---------------+--------------+ | |
| | Metric | Value | | |
| +===============+==============+ | |
| | loss | -4.73834e+16 | | |
| +---------------+--------------+ | |
| | grad_norm | nan | | |
| +---------------+--------------+ | |
| | learning_rate | 0.0001 | | |
| +---------------+--------------+ | |
| | epoch | 0.231214 | | |
| +---------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:31:20,408 - INFO - | |
| [36m🚂 Training Metrics (Step 20) 🚂 | |
| +---------------+--------------+ | |
| | Metric | Value | | |
| +===============+==============+ | |
| | loss | -3.59497e+17 | | |
| +---------------+--------------+ | |
| | grad_norm | nan | | |
| +---------------+--------------+ | |
| | learning_rate | 0.0001 | | |
| +---------------+--------------+ | |
| | epoch | 0.462428 | | |
| +---------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:31:34,863 - INFO - | |
| [36m🚂 Training Metrics (Step 30) 🚂 | |
| +---------------+--------------+ | |
| | Metric | Value | | |
| +===============+==============+ | |
| | loss | -6.37696e+16 | | |
| +---------------+--------------+ | |
| | grad_norm | nan | | |
| +---------------+--------------+ | |
| | learning_rate | 0.0001 | | |
| +---------------+--------------+ | |
| | epoch | 0.693642 | | |
| +---------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:31:49,256 - INFO - | |
| [36m🚂 Training Metrics (Step 40) 🚂 | |
| +---------------+--------------+ | |
| | Metric | Value | | |
| +===============+==============+ | |
| | loss | -8.24219e+17 | | |
| +---------------+--------------+ | |
| | grad_norm | nan | | |
| +---------------+--------------+ | |
| | learning_rate | 0.0001 | | |
| +---------------+--------------+ | |
| | epoch | 0.924855 | | |
| +---------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:31:53,966 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [learning2rank.py:2635:evaluate] | |
| 2025-06-11 15:35:08,172 - WARNING - No valid samples found for metric 'precision@25'. - [learning2rank.py:2739:evaluate] | |
| 2025-06-11 15:35:08,172 - INFO - | |
| [33m🔍 Evaluation Metrics 🔍 | |
| +-------------------+-------------+ | |
| | Metric | Value | | |
| +===================+=============+ | |
| | eval_loss | -1.2019e+17 | | |
| +-------------------+-------------+ | |
| | eval_ndcg | 0.955488 | | |
| +-------------------+-------------+ | |
| | eval_ndcg@25 | 0.212566 | | |
| +-------------------+-------------+ | |
| | eval_precision@25 | 0 | | |
| +-------------------+-------------+[0m - [learning2rank.py:2206:on_evaluate] | |
| 2025-06-11 15:35:09,356 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-44 - [learning2rank.py:2759:_save] | |
| 2025-06-11 15:35:09,383 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-44 - [learning2rank.py:2764:_save] | |
| 2025-06-11 15:35:09,384 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-44: | |
| +---------+-------------------+------------+ | |
| | Index | Saved File | Size | | |
| +=========+===================+============+ | |
| | 1 | training_args.bin | 0.01 MB | | |
| +---------+-------------------+------------+ | |
| | 2 | model.safetensors | 4122.74 MB | | |
| +---------+-------------------+------------+ | |
| | 3 | config.json | 0.38 MB | | |
| +---------+-------------------+------------+ - [learning2rank.py:2781:_save] | |
| 2025-06-11 15:35:22,810 - INFO - | |
| [36m🚂 Training Metrics (Step 50) 🚂 | |
| +---------------+--------------+ | |
| | Metric | Value | | |
| +===============+==============+ | |
| | loss | -1.24339e+18 | | |
| +---------------+--------------+ | |
| | grad_norm | nan | | |
| +---------------+--------------+ | |
| | learning_rate | 0.0001 | | |
| +---------------+--------------+ | |
| | epoch | 1.13873 | | |
| +---------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:35:37,308 - INFO - | |
| [36m🚂 Training Metrics (Step 60) 🚂 | |
| +---------------+--------------+ | |
| | Metric | Value | | |
| +===============+==============+ | |
| | loss | -8.42207e+16 | | |
| +---------------+--------------+ | |
| | grad_norm | nan | | |
| +---------------+--------------+ | |
| | learning_rate | 9.7e-05 | | |
| +---------------+--------------+ | |
| | epoch | 1.36994 | | |
| +---------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:35:51,736 - INFO - | |
| [36m🚂 Training Metrics (Step 70) 🚂 | |
| +---------------+--------------+ | |
| | Metric | Value | | |
| +===============+==============+ | |
| | loss | -6.54621e+18 | | |
| +---------------+--------------+ | |
| | grad_norm | 1.53209e+18 | | |
| +---------------+--------------+ | |
| | learning_rate | 8.8e-05 | | |
| +---------------+--------------+ | |
| | epoch | 1.60116 | | |
| +---------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:36:06,062 - INFO - | |
| [36m🚂 Training Metrics (Step 80) 🚂 | |
| +---------------+--------------+ | |
| | Metric | Value | | |
| +===============+==============+ | |
| | loss | -9.11572e+16 | | |
| +---------------+--------------+ | |
| | grad_norm | nan | | |
| +---------------+--------------+ | |
| | learning_rate | 7.8e-05 | | |
| +---------------+--------------+ | |
| | epoch | 1.83237 | | |
| +---------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:36:15,873 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2759:_save] | |
| 2025-06-11 15:36:15,900 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2764:_save] | |
| 2025-06-11 15:36:15,901 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86: | |
| +---------+-------------------+------------+ | |
| | Index | Saved File | Size | | |
| +=========+===================+============+ | |
| | 1 | training_args.bin | 0.01 MB | | |
| +---------+-------------------+------------+ | |
| | 2 | model.safetensors | 4122.74 MB | | |
| +---------+-------------------+------------+ | |
| | 3 | config.json | 0.38 MB | | |
| +---------+-------------------+------------+ - [learning2rank.py:2781:_save] | |
| 2025-06-11 15:36:16,120 - INFO - Removing 'token_type_ids' from eval_dataset as they are not needed. - [learning2rank.py:2635:evaluate] | |
| 2025-06-11 15:39:29,774 - INFO - | |
| [33m🔍 Evaluation Metrics 🔍 | |
| +-------------------+--------------+ | |
| | Metric | Value | | |
| +===================+==============+ | |
| | eval_loss | -2.89696e+17 | | |
| +-------------------+--------------+ | |
| | eval_ndcg | 0.956037 | | |
| +-------------------+--------------+ | |
| | eval_ndcg@25 | 0.620746 | | |
| +-------------------+--------------+ | |
| | eval_precision@25 | 0.2464 | | |
| +-------------------+--------------+[0m - [learning2rank.py:2206:on_evaluate] | |
| 2025-06-11 15:39:33,224 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2759:_save] | |
| 2025-06-11 15:39:33,252 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2764:_save] | |
| 2025-06-11 15:39:33,253 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86: | |
| +---------+--------------------+------------+ | |
| | Index | Saved File | Size | | |
| +=========+====================+============+ | |
| | 1 | training_args.bin | 0.01 MB | | |
| +---------+--------------------+------------+ | |
| | 2 | optimizer.pt | 352.39 MB | | |
| +---------+--------------------+------------+ | |
| | 3 | model.safetensors | 4122.74 MB | | |
| +---------+--------------------+------------+ | |
| | 4 | scaler.pt | 0.00 MB | | |
| +---------+--------------------+------------+ | |
| | 5 | config.json | 0.38 MB | | |
| +---------+--------------------+------------+ | |
| | 6 | scheduler.pt | 0.00 MB | | |
| +---------+--------------------+------------+ | |
| | 7 | trainer_state.json | 0.00 MB | | |
| +---------+--------------------+------------+ | |
| | 8 | rng_state.pth | 0.01 MB | | |
| +---------+--------------------+------------+ - [learning2rank.py:2781:_save] | |
| 2025-06-11 15:39:33,735 - INFO - 📂 Loading best model from ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86 - [learning2rank.py:2837:_load_best_model] | |
| 2025-06-11 15:39:33,735 - INFO - 🖥️ Model is on device: cuda:0 - [learning2rank.py:2847:_load_best_model] | |
| 2025-06-11 15:39:33,798 - 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:2871:_load_best_model] | |
| 2025-06-11 15:39:34,630 - INFO - ✅ Loaded best model weights from ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/checkpoint-86/model.safetensors - [learning2rank.py:2888:_load_best_model] | |
| 2025-06-11 15:39:34,670 - INFO - ✔️ Weight for label_embeddings.weight matches between saved and loaded state_dict - [learning2rank.py:2900:_load_best_model] | |
| 2025-06-11 15:39:34,725 - INFO - ✔️ Weight for ground_model.base_model.model.model.embed_tokens.weight matches between saved and loaded state_dict - [learning2rank.py:2900:_load_best_model] | |
| 2025-06-11 15:39:34,740 - INFO - | |
| [36m🚂 Training Metrics (Step 86) 🚂 | |
| +--------------------------+--------------+ | |
| | Metric | Value | | |
| +==========================+==============+ | |
| | train_runtime | 530.146 | | |
| +--------------------------+--------------+ | |
| | train_samples_per_second | 0.653 | | |
| +--------------------------+--------------+ | |
| | train_steps_per_second | 0.162 | | |
| +--------------------------+--------------+ | |
| | total_flos | 3.81076e+15 | | |
| +--------------------------+--------------+ | |
| | train_loss | -1.15684e+18 | | |
| +--------------------------+--------------+ | |
| | epoch | 1.9711 | | |
| +--------------------------+--------------+[0m - [learning2rank.py:2187:on_log] | |
| 2025-06-11 15:39:34,740 - INFO - ✨ Training Completed! ✨ - [learning2rank.py:2336:on_train_end] | |
| 2025-06-11 15:39:34,814 - INFO - 📊 Training loss plot saved as '../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/train_loss_plot.png' - [learning2rank.py:2439:on_train_end] | |
| 2025-06-11 15:39:34,870 - INFO - 📊 Evaluation loss plot saved as '../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/eval_loss_plot.png' - [learning2rank.py:2453:on_train_end] | |
| 2025-06-11 15:39:34,928 - INFO - 📊 Evaluation metric plot saved as '../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/eval_ndcg@25_plot.png' - [learning2rank.py:2474:on_train_end] | |
| 2025-06-11 15:39:34,928 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:109:log_section] | |
| 2025-06-11 15:39:34,928 - INFO - + ✨ MODEL SAVING + - [learning2rank.py:110:log_section] | |
| 2025-06-11 15:39:34,928 - INFO - ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - [learning2rank.py:113:log_section] | |
| 2025-06-11 15:39:34,928 - INFO - 💾 Saving trained model and pushing to Hugging Face Hub... - [learning2rank.py:3924:main] | |
| 2025-06-11 15:39:34,928 - INFO - 📁 Creating/using output directory: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3155:save_and_push] | |
| 2025-06-11 15:39:36,118 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:2759:_save] | |
| 2025-06-11 15:39:36,144 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:2764:_save] | |
| 2025-06-11 15:39:36,145 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b: | |
| +---------+-----------------------+------------+ | |
| | Index | Saved File | Size | | |
| +=========+=======================+============+ | |
| | 1 | eval_loss_plot.png | 0.03 MB | | |
| +---------+-----------------------+------------+ | |
| | 2 | training_args.bin | 0.01 MB | | |
| +---------+-----------------------+------------+ | |
| | 3 | model.safetensors | 4122.74 MB | | |
| +---------+-----------------------+------------+ | |
| | 4 | eval_ndcg@25_plot.png | 0.04 MB | | |
| +---------+-----------------------+------------+ | |
| | 5 | config.json | 0.38 MB | | |
| +---------+-----------------------+------------+ | |
| | 6 | train_loss_plot.png | 0.03 MB | | |
| +---------+-----------------------+------------+ - [learning2rank.py:2781:_save] | |
| 2025-06-11 15:39:39,545 - INFO - 💾 Model weights saved in safetensors format: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:2759:_save] | |
| 2025-06-11 15:39:39,573 - INFO - ⚙️ Config saved in checkpoint: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:2764:_save] | |
| 2025-06-11 15:39:39,575 - INFO - 📋 Saved files in ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b: | |
| +---------+-----------------------+------------+ | |
| | Index | Saved File | Size | | |
| +=========+=======================+============+ | |
| | 1 | eval_loss_plot.png | 0.03 MB | | |
| +---------+-----------------------+------------+ | |
| | 2 | training_args.bin | 0.01 MB | | |
| +---------+-----------------------+------------+ | |
| | 3 | model.safetensors | 4122.74 MB | | |
| +---------+-----------------------+------------+ | |
| | 4 | eval_ndcg@25_plot.png | 0.04 MB | | |
| +---------+-----------------------+------------+ | |
| | 5 | config.json | 0.38 MB | | |
| +---------+-----------------------+------------+ | |
| | 6 | train_loss_plot.png | 0.03 MB | | |
| +---------+-----------------------+------------+ - [learning2rank.py:2781:_save] | |
| 2025-06-11 15:41:03,613 - INFO - 💾 Model saved to: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3159:save_and_push] | |
| 2025-06-11 15:41:07,306 - INFO - 💾 Model and config explicitly saved to: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3163:save_and_push] | |
| 2025-06-11 15:41:07,337 - INFO - 🖌️ Tokenizer saved to: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b - [learning2rank.py:3167:save_and_push] | |
| 2025-06-11 15:41:07,771 - INFO - 📊 Lookup table saved to: ../tmp/MIMIC4_DEMO/mimic4_l2rtrain_mistral7b/lookup_table.pt - [learning2rank.py:3172:save_and_push] | |