ptyrad.utils.common

General purpose utility tools for logging, converting types, parsing dicts and strings, etc.

Classes

CustomLogger([log_file, log_dir, ...])

Functions

expand_presets(input_list, presets)
get_nested(d, key[, delimiter, safe, default])	Get a value from a nested dictionary either safely (return default if not found) or stricly to fail early.
get_time([time_format])	Returns a formatted timestamp string based on time_format.
handle_hdf5_types(x)	Convert data to native Python or NumPy types.
is_mig_enabled()	Detects if any GPU on the system is operating in MIG (Multi-Instance GPU) mode.
list_nested_keys(hobj[, delimiter, prefix])	Recursively list all keys in an HDF5 file, HDF5 group, or dict, including hierarchical paths.
ndarrays_to_tensors(data[, device])	Recursively convert all numpy.ndarray instances in any nested structure (including lists, dicts, and tuples) into torch.Tensor on the specified device.
normalize_constraint_params(constraint_params)	Convert old constraint param format {freq} (pre v0.1.0b11) to {start_iter, step, end_iter}.
parse_hypertune_params_to_str(hypertune_params)
parse_sec_to_time_str(seconds)
print_gpu_info()
print_packages_info()
print_system_info()
resolve_seed_priority(args_seed, ...)
safe_filename(filepath[, verbose])	Ensures a filepath is safe across platforms by: 1.
set_accelerator()
set_gpu_device([gpuid])	Sets the GPU device based on the input.
set_random_seed(seed[, deterministic])	Set the random seeds for numpy and pytorch operations.
tensors_to_ndarrays(data)	Recursively convert all torch.Tensor instances in any nested structure (including lists, dicts, and tuples) into numpy.ndarray.
time_sync()
vprint(*args[, verbose])	Verbose print/logging with individual control, only for rank 0 in DDP.
vprint_nested_dict(d[, indent, verbose, ...])