Enum TorshDistributedError 

pub enum TorshDistributedError {
    BackendNotInitialized,
    InvalidArgument {
        arg: String,
        reason: String,
        expected: String,
    },
    CommunicationError {
        operation: String,
        cause: String,
    },
    BackendError {
        backend: String,
        message: String,
    },
    RankOutOfBounds {
        rank: u32,
        world_size: u32,
    },
    FeatureNotAvailable {
        feature: String,
        required_features: String,
    },
    ProcessGroupNotFound {
        group_id: String,
        available_groups: Vec<String>,
    },
    TensorShapeMismatch {
        expected: Vec<usize>,
        actual: Vec<usize>,
    },
    OperationTimeout {
        operation: String,
        timeout_secs: u64,
    },
    ProcessFailure {
        rank: u32,
        operation: String,
        cause: String,
    },
    MemoryAllocationFailed {
        requested_bytes: usize,
        context: String,
    },
    SerializationError(String),
    IoError(String),
    InternalError(String),
    ConfigurationError {
        message: String,
    },
    CheckpointError {
        operation: String,
        cause: String,
    },
}

Distributed training specific errors with detailed context

Variants

BackendNotInitialized

InvalidArgument

Fields

arg: String

reason: String

expected: String

CommunicationError

Fields

operation: String

cause: String

BackendError

Fields

backend: String

message: String

RankOutOfBounds

Fields

rank: u32

world_size: u32

FeatureNotAvailable

Fields

feature: String

required_features: String

ProcessGroupNotFound

Fields

group_id: String

available_groups: Vec<String>

TensorShapeMismatch

Fields

expected: Vec<usize>

actual: Vec<usize>

OperationTimeout

Fields

operation: String

timeout_secs: u64

ProcessFailure

Fields

rank: u32

operation: String

cause: String

MemoryAllocationFailed

Fields

requested_bytes: usize

context: String

SerializationError(String)

IoError(String)

InternalError(String)

ConfigurationError

Fields

message: String

CheckpointError

Fields

operation: String

cause: String

Implementations

impl TorshDistributedError

pub fn invalid_argument( arg: impl Into<String>, reason: impl Into<String>, expected: impl Into<String>, ) -> Self

Create an invalid argument error with context

pub fn communication_error( operation: impl Into<String>, cause: impl Into<String>, ) -> Self

Create a communication error with operation context

pub fn backend_error( backend: impl Into<String>, message: impl Into<String>, ) -> Self

Create a backend error with backend type

pub fn feature_not_available( feature: impl Into<String>, required_features: impl Into<String>, ) -> Self

Create a feature not available error with required features

pub fn process_group_not_found( group_id: impl Into<String>, available_groups: Vec<String>, ) -> Self

Create a process group not found error

pub fn tensor_shape_mismatch(expected: Vec<usize>, actual: Vec<usize>) -> Self

Create a tensor shape mismatch error

pub fn operation_timeout( operation: impl Into<String>, timeout_secs: u64, ) -> Self

Create an operation timeout error

pub fn process_failure( rank: u32, operation: impl Into<String>, cause: impl Into<String>, ) -> Self

Create a process failure error

pub fn memory_allocation_failed( requested_bytes: usize, context: impl Into<String>, ) -> Self

Create a memory allocation failure error

pub fn serialization_error(message: impl Into<String>) -> Self

Create a serialization error

pub fn io_error(message: impl Into<String>) -> Self

Create an I/O error

pub fn internal_error(message: impl Into<String>) -> Self

Create an internal error

pub fn configuration_error(message: impl Into<String>) -> Self

Create a configuration error

pub fn checkpoint_error( operation: impl Into<String>, cause: impl Into<String>, ) -> Self

Create a checkpoint error

pub fn not_implemented(feature: impl Into<String>) -> Self

Create a not implemented error

pub fn is_retryable(&self) -> bool

Check if this error is retryable

pub fn recovery_suggestions(&self) -> Vec<&'static str>

Get suggested recovery actions for this error

Trait Implementations

impl Debug for TorshDistributedError

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Display for TorshDistributedError

fn fmt(&self, __formatter: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Error for TorshDistributedError

fn source(&self) -> Option<&(dyn Error + 'static)>

Returns the lower-level source of this error, if any.

fn description(&self) -> &str

👎Deprecated since 1.42.0: use the Display impl or to_string()

fn cause(&self) -> Option<&dyn Error>

👎Deprecated since 1.33.0: replaced by Error::source, which can support downcasting

fn provide<'a>(&'a self, request: &mut Request<'a>)

🔬This is a nightly-only experimental API. (error_generic_member_access)

Provides type-based access to context intended for error reports.

impl From<TorshDistributedError> for TorshError

fn from(err: TorshDistributedError) -> Self

Converts to this type from the input type.

impl From<TorshError> for TorshDistributedError

fn from(err: TorshError) -> Self

Converts to this type from the input type.