Python mxnet.model.load_checkpoint() Examples

The following are 30 code examples of mxnet.model.load_checkpoint(). You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may also want to check out all available functions/classes of the module mxnet.model , or try the search function .
Example #1
Source File: module.py    From Faster_RCNN_for_DOTA with Apache License 2.0 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #2
Source File: module.py    From Deformable-ConvNets with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #3
Source File: module.py    From RoITransformer_DOTA with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #4
Source File: module.py    From RoITransformer_DOTA with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #5
Source File: module.py    From RoITransformer_DOTA with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #6
Source File: module.py    From RoITransformer_DOTA with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #7
Source File: module.py    From Relation-Networks-for-Object-Detection with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #8
Source File: module.py    From Relation-Networks-for-Object-Detection with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #9
Source File: module.py    From Accel with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #10
Source File: module.py    From Accel with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #11
Source File: module.py    From Accel with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #12
Source File: module.py    From Accel with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #13
Source File: module.py    From Accel with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #14
Source File: module.py    From Accel with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #15
Source File: module.py    From Accel with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #16
Source File: module.py    From Deformable-ConvNets with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #17
Source File: module.py    From Faster_RCNN_for_DOTA with Apache License 2.0 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True, allow_extra=False):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init, allow_extra=allow_extra)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params, allow_extra=allow_extra)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #18
Source File: module.py    From Sequence-Level-Semantics-Aggregation with Apache License 2.0 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #19
Source File: module.py    From Sequence-Level-Semantics-Aggregation with Apache License 2.0 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True, allow_extra=False):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init, allow_extra=allow_extra)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #20
Source File: detection_module.py    From groupsoftmax-simpledet with Apache License 2.0 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Creates a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is ``cpu()``.
        work_load_list : list of number
            Default ``None``, indicating uniform workload.
        fixed_param_names: list of str
            Default ``None``, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = DetModule(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #21
Source File: detection_module.py    From groupsoftmax-simpledet with Apache License 2.0 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True,
                   allow_extra=False):
        """Assigns parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to `NDArray`.
        aux_params : dict
            Dictionary of name to `NDArray`.
        allow_missing : bool
            If ``True``, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If ``True``, will force re-initialize even if already initialized.
        allow_extra : boolean, optional
            Whether allow extra parameters that are not needed by symbol.
            If this is True, no error will be thrown when arg_params or aux_params
            contain extra parameters that is not needed by the executor.
        Examples
        --------
        >>> # An example of setting module parameters.
        >>> sym, arg_params, aux_params = mx.model.load_checkpoint(model_prefix, n_epoch_load)
        >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init,
                             allow_extra=allow_extra)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params, allow_extra=allow_extra)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #22
Source File: module.py    From Decoupled-Classification-Refinement with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #23
Source File: module.py    From Decoupled-Classification-Refinement with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #24
Source File: module.py    From Decoupled-Classification-Refinement with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #25
Source File: module.py    From Decoupled-Classification-Refinement with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #26
Source File: module.py    From Decoupled-Classification-Refinement with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #27
Source File: module.py    From Decoupled-Classification-Refinement with MIT License 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Create a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is `cpu()`.
        work_load_list : list of number
            Default `None`, indicating uniform workload.
        fixed_param_names: list of str
            Default `None`, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = Module(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #28
Source File: module.py    From Decoupled-Classification-Refinement with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True 
Example #29
Source File: custom_module.py    From SNIPER-mxnet with Apache License 2.0 5 votes vote down vote up
def load(prefix, epoch, load_optimizer_states=False, **kwargs):
        """Creates a model from previously saved checkpoint.

        Parameters
        ----------
        prefix : str
            path prefix of saved model files. You should have
            "prefix-symbol.json", "prefix-xxxx.params", and
            optionally "prefix-xxxx.states", where xxxx is the
            epoch number.
        epoch : int
            epoch to load.
        load_optimizer_states : bool
            whether to load optimizer states. Checkpoint needs
            to have been made with save_optimizer_states=True.
        data_names : list of str
            Default is `('data')` for a typical model used in image classification.
        label_names : list of str
            Default is `('softmax_label')` for a typical model used in image
            classification.
        logger : Logger
            Default is `logging`.
        context : Context or list of Context
            Default is ``cpu()``.
        work_load_list : list of number
            Default ``None``, indicating uniform workload.
        fixed_param_names: list of str
            Default ``None``, indicating no network parameters are fixed.
        """
        sym, args, auxs = load_checkpoint(prefix, epoch)
        mod = CustomModule(symbol=sym, **kwargs)
        mod._arg_params = args
        mod._aux_params = auxs
        mod.params_initialized = True
        if load_optimizer_states:
            mod._preload_opt_states = '%s-%04d.states'%(prefix, epoch)
        return mod 
Example #30
Source File: module.py    From kaggle-rsna18 with MIT License 5 votes vote down vote up
def set_params(self, arg_params, aux_params, allow_missing=False, force_init=True):
        """Assign parameter and aux state values.

        Parameters
        ----------
        arg_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        aux_params : dict
            Dictionary of name to value (`NDArray`) mapping.
        allow_missing : bool
            If true, params could contain missing values, and the initializer will be
            called to fill those missing params.
        force_init : bool
            If true, will force re-initialize even if already initialized.

        Examples
        --------
        An example of setting module parameters::
            >>> sym, arg_params, aux_params = \
            >>>     mx.model.load_checkpoint(model_prefix, n_epoch_load)
            >>> mod.set_params(arg_params=arg_params, aux_params=aux_params)
        """
        if not allow_missing:
            self.init_params(initializer=None, arg_params=arg_params, aux_params=aux_params,
                             allow_missing=allow_missing, force_init=force_init)
            return

        if self.params_initialized and not force_init:
            warnings.warn("Parameters already initialized and force_init=False. "
                          "set_params call ignored.", stacklevel=2)
            return

        self._exec_group.set_params(arg_params, aux_params)

        # because we didn't update self._arg_params, they are dirty now.
        self._params_dirty = True
        self.params_initialized = True