# Quark Profiling Results memory_usage: - step: "Start" timestamp: 1778206913.5518243 relative_time_secs: 0.0 cpu_memory_mb: 4589.59 gpu_memory_mb: 777.22 disk_read_mb: 0.0 disk_write_mb: 0.0 - step: "Model Loading Start" timestamp: 1778206913.7514918 relative_time_secs: 0.19966745376586914 cpu_memory_mb: 4589.59 gpu_memory_mb: 777.22 disk_read_mb: 0.0 disk_write_mb: 0.0 - step: "Model Loading End" timestamp: 1778207378.5582051 relative_time_secs: 465.0063807964325 cpu_memory_mb: 6537.77 gpu_memory_mb: 63960.58 disk_read_mb: 436296.35 disk_write_mb: 0.09 - step: "Dataset Loading Start" timestamp: 1778207394.8392925 relative_time_secs: 481.28746819496155 cpu_memory_mb: 27673.3 gpu_memory_mb: 65073.95 disk_read_mb: 437001.61 disk_write_mb: 8369.02 - step: "Dataset Loading End" timestamp: 1778207398.4055986 relative_time_secs: 484.8537743091583 cpu_memory_mb: 27678.77 gpu_memory_mb: 65073.95 disk_read_mb: 437349.88 disk_write_mb: 8369.02 - step: "Model Quantization Start" timestamp: 1778207398.6073127 relative_time_secs: 485.0554883480072 cpu_memory_mb: 27678.77 gpu_memory_mb: 65073.95 disk_read_mb: 437349.88 disk_write_mb: 8369.02 - step: "Model Preparation Start" timestamp: 1778207398.749175 relative_time_secs: 485.19735074043274 cpu_memory_mb: 27678.77 gpu_memory_mb: 65073.95 disk_read_mb: 437349.88 disk_write_mb: 8369.02 - step: "Model Preparation End" timestamp: 1778207436.6778612 relative_time_secs: 523.1260368824005 cpu_memory_mb: 35183.42 gpu_memory_mb: 65119.96 disk_read_mb: 437349.88 disk_write_mb: 8369.02 - step: "Advanced Algorithms Start" timestamp: 1778207436.84824 relative_time_secs: 523.2964155673981 cpu_memory_mb: 35183.42 gpu_memory_mb: 65119.96 disk_read_mb: 437349.88 disk_write_mb: 8369.02 - step: "Advanced Algorithms End" timestamp: 1778207437.0057657 relative_time_secs: 523.4539413452148 cpu_memory_mb: 35183.42 gpu_memory_mb: 65119.96 disk_read_mb: 437349.88 disk_write_mb: 8369.02 - step: "Calibration Start" timestamp: 1778207437.273917 relative_time_secs: 523.722092628479 cpu_memory_mb: 35183.42 gpu_memory_mb: 65119.96 disk_read_mb: 437349.88 disk_write_mb: 8369.02 - step: "Calibration End" timestamp: 1778210047.9661236 relative_time_secs: 3134.414299249649 cpu_memory_mb: 36125.95 gpu_memory_mb: 87323.7 disk_read_mb: 437351.9 disk_write_mb: 8399.16 - step: "Model Quantization End" timestamp: 1778210176.061249 relative_time_secs: 3262.509424686432 cpu_memory_mb: 107186.82 gpu_memory_mb: 87275.89 disk_read_mb: 437352.28 disk_write_mb: 8399.16 - step: "Freeze Model Start" timestamp: 1778210176.2332122 relative_time_secs: 3262.681387901306 cpu_memory_mb: 107186.82 gpu_memory_mb: 87275.89 disk_read_mb: 437352.28 disk_write_mb: 8399.16 - step: "Freeze Model End" timestamp: 1778210190.9314125 relative_time_secs: 3277.3795881271362 cpu_memory_mb: 107521.27 gpu_memory_mb: 87275.89 disk_read_mb: 437352.28 disk_write_mb: 8399.16 - step: "Export HF Safetensors Start" timestamp: 1778210191.0775836 relative_time_secs: 3277.5257592201233 cpu_memory_mb: 107521.52 gpu_memory_mb: 87275.89 disk_read_mb: 437352.28 disk_write_mb: 8399.16 - step: "Export HF Safetensors End" timestamp: 1778210393.5681884 relative_time_secs: 3480.016364097595 cpu_memory_mb: 112519.61 gpu_memory_mb: 86791.54 disk_read_mb: 437357.31 disk_write_mb: 136647.9 - step: "Model Evaluation Start" timestamp: 1778210393.7849188 relative_time_secs: 3480.2330944538116 cpu_memory_mb: 112519.61 gpu_memory_mb: 86791.54 disk_read_mb: 437357.31 disk_write_mb: 136647.9 - step: "Model Evaluation End" timestamp: 1778214657.600004 relative_time_secs: 7744.048179626465 cpu_memory_mb: 113494.72 gpu_memory_mb: 87319.95 disk_read_mb: 437365.14 disk_write_mb: 136653.94 - step: "End" timestamp: 1778214657.857704 relative_time_secs: 7744.3058795928955 cpu_memory_mb: 113492.76 gpu_memory_mb: 87319.95 disk_read_mb: 437365.14 disk_write_mb: 136653.94 # Summary Metrics total_quantization_time_seconds: 7744.3059 peak_memory_mb: 113495.66 peak_gpu_memory_mb: 87323.7 total_disk_read_mb: 437365.14 total_disk_write_mb: 136653.94 # Metric Definitions: # # Checkpoint Metrics (per record): # - step: Name of the profiling checkpoint. Common steps include: # - "Start": Initial state when profiling begins # - "Model Loaded": After loading the ONNX model into memory # - "Pre-process Start/End": Before and after model preprocessing # - "Calibration Start/End": Before and after calibration data collection # - "Quantization (MatMulNBits) Start/End": MatMulNBits quantization phase # - "Quantization (Static) Start/End": Static quantization phase # - "Post-process Start/End": Before and after post-processing # - "Fast Finetune Start/End": Before and after fast finetuning (if enabled) # - timestamp: Unix timestamp (seconds since epoch) when this measurement was taken. Useful for correlating with external logs or events. # - relative_time_secs: Time elapsed (in seconds) since the "Start" step. Useful for understanding the duration of each phase relative to the beginning of profiling. # - cpu_memory_mb: Current Resident Set Size (RSS) in megabytes at this step. This includes memory from the main process and all child processes. RSS represents the portion of memory held in RAM (not swapped out). # - gpu_memory_mb: Current GPU memory usage in megabytes. This represents actual GPU memory used by the process, including allocations from PyTorch, ONNX Runtime, TensorRT, and other frameworks. Only available when PyTorch with CUDA/ROCm is installed and GPU is available. # - disk_read_mb: Cumulative disk bytes read (in megabytes) since the start of profiling. Measured relative to the baseline captured at the 'Start' checkpoint, including I/O from the main process and all child processes. Only available when psutil is installed and the OS exposes per-process I/O counters (Linux /proc//io, Windows; not available on macOS without root). # - disk_write_mb: Cumulative disk bytes written (in megabytes) since the start of profiling. Measured relative to the baseline captured at the 'Start' checkpoint, including I/O from the main process and all child processes. Only available when psutil is installed and the OS exposes per-process I/O counters (Linux /proc//io, Windows; not available on macOS without root). # # Summary Metrics (overall): # - total_quantization_time_seconds: Total elapsed time (in seconds) from the start of profiling to the end of the quantization process. # - peak_memory_mb: Peak resident set size (RSS) in megabytes for the main process during the entire profiling session. On Linux, this is read from VmHWM (high water mark) in /proc//status. On Windows, this is the peak working set size. This metric may not be available on all platforms. # - peak_gpu_memory_mb: Peak GPU memory usage in megabytes during the entire profiling session. This is the maximum GPU memory used, including allocations from PyTorch, ONNX Runtime, TensorRT, and other frameworks. Only available when PyTorch with CUDA/ROCm is installed and GPU is available. # - total_disk_read_mb: Total disk bytes read (in megabytes) during the entire profiling session. Computed as the difference between the final and baseline cumulative read counters, including I/O from the main process and all child processes. Only available when psutil is installed and the OS exposes per-process I/O counters (Linux /proc//io, Windows; not available on macOS without root). # - total_disk_write_mb: Total disk bytes written (in megabytes) during the entire profiling session. Computed as the difference between the final and baseline cumulative write counters, including I/O from the main process and all child processes. Only available when psutil is installed and the OS exposes per-process I/O counters (Linux /proc//io, Windows; not available on macOS without root). # - peak_cache_dir_disk_usage_mb: Highest peak increase in disk usage (in megabytes) among all cache directories created during the profiling session, relative to each cache directory's size when monitoring started. Sampled every 1 second by recursively summing file sizes with os.scandir().