Parameter#

class zfit.param.Parameter(name, value, lower=None, upper=None, step_size=None, floating=True, *, dtype=None, lower_limit=None, upper_limit=None)[source]#

Bases: ZfitParameterMixin, TFBaseVariable, BaseParameter, SerializableMixin, ZfitIndependentParameter

Class for fit parameters that has a default state.

Fit Parameter that has a default state (value) and limits (lower, upper). (deprecated arguments) (deprecated arguments)

Deprecated: SOME ARGUMENTS ARE DEPRECATED: (lower_limit). They will be removed in a future version. Instructions for updating: Use lower instead.

Deprecated: SOME ARGUMENTS ARE DEPRECATED: (upper_limit). They will be removed in a future version. Instructions for updating: Use upper instead.

The name identifies the parameter. Multiple parameters with the same name can exist, however, they cannot be in the same PDF/func/loss as the value would not be uniquely defined.

Parameters:
  • name (str) – name of the parameter

  • value (ztyping.NumericalScalarType) – starting value

  • lower (ztyping.NumericalScalarType | None) – lower limit

  • upper (ztyping.NumericalScalarType | None) – upper limit

  • step_size (ztyping.NumericalScalarType | None) – step size

property has_limits: bool#

If the parameter has limits set or not.

property at_limit: Tensor#

If the value is at the limit (or over it).

Returns:

Boolean tf.Tensor that tells whether the value is at the limits.

read_value()[source]#

DEPRECATED FUNCTION

Deprecated: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use value instead.

property step_size: Tensor#

Step size of the parameter, the estimated order of magnitude of the uncertainty.

This can be crucial to tune for the minimization. A too large step_size can produce NaNs, a too small won’t converge.

If the step size is not set, the DEFAULT_STEP_SIZE is used.

Returns:

The step size

set_value(value)[source]#

Set the Parameter to value (temporarily if used in a context manager).

This operation won’t, compared to the assign, return the read value but an object that can act as a context manager.

Parameters:

value (ztyping.NumericalScalarType) – The value the parameter will take on.

Raises:
  • ValueError – If the value is not inside the limits (in normal Python/eager mode)

  • InvalidArgumentError – If the value is not inside the limits (in JIT/traced/graph mode)

assign(value, use_locking=None, name=None, read_value=False)[source]#

Set the Parameter to value without any checks.

Compared to set_value, this method cannot be used with a context manager and won’t raise an error

Parameters:

value – The value the parameter will take on.

randomize(minval=None, maxval=None, sampler=<built-in method uniform of numpy.random.mtrand.RandomState object>)[source]#

Update the parameter with a randomised value between minval and maxval and return it.

Parameters:
  • minval (ztyping.NumericalScalarType | None) – The lower bound of the sampler. If not given, lower_limit is used.

  • maxval (ztyping.NumericalScalarType | None) – The upper bound of the sampler. If not given, upper_limit is used.

  • sampler (Callable) – A sampler with the same interface as np.random.uniform

Return type:

tf.Tensor

Returns:

The sampled value

class SaveSliceInfo(full_name=None, full_shape=None, var_offset=None, var_shape=None, save_slice_info_def=None, import_scope=None)#

Bases: object

Information on how to save this Variable as a slice.

Provides internal support for saving variables as slices of a larger variable. This API is not public and is subject to change.

Available properties:

  • full_name

  • full_shape

  • var_offset

  • var_shape

Create a SaveSliceInfo.

Parameters:
  • full_name – Name of the full variable of which this Variable is a slice.

  • full_shape – Shape of the full variable, as a list of int.

  • var_offset – Offset of this Variable into the full variable, as a list of int.

  • var_shape – Shape of this Variable, as a list of int.

  • save_slice_info_defSaveSliceInfoDef protocol buffer. If not None, recreates the SaveSliceInfo object its contents. save_slice_info_def and other arguments are mutually exclusive.

  • import_scope – Optional string. Name scope to add. Only used when initializing from protocol buffer.

property spec#

Computes the spec string used for saving.

to_proto(export_scope=None)#

Returns a SaveSliceInfoDef() proto.

Parameters:

export_scope – Optional string. Name scope to remove.

Returns:

A SaveSliceInfoDef protocol buffer, or None if the Variable is not in the specified name scope.

__array__(dtype=None)#

Allows direct conversion to a numpy array.

>>> np.array(tf.Variable([1.0]))
array([1.], dtype=float32)
Returns:

The variable value as a numpy array.

__iter__()#

When executing eagerly, iterates over the value of the variable.

__ne__(other)#

Compares two variables element-wise for equality.

add_cache_deps(cache_deps, allow_non_cachable=True)#

Add dependencies that render the cache invalid if they change.

Parameters:
  • cache_deps (ztyping.CacherOrCachersType) –

  • allow_non_cachable (bool) – If True, allow cache_dependents to be non-cachables. If False, any cache_dependents that is not a ZfitGraphCachable will raise an error.

Raises:

TypeError – if one of the cache_dependents is not a ZfitGraphCachable _and_ allow_non_cachable if False.

assign_add(delta, use_locking=None, name=None, read_value=True)#

Adds a value to this variable.

Parameters:
  • delta – A Tensor. The value to add to this variable.

  • use_locking – If True, use locking during the operation.

  • name – The name to use for the operation.

  • read_value – A bool. Whether to read and return the new value of the variable or not.

Returns:

If read_value is True, this method will return the new value of the variable after the assignment has completed. Otherwise, when in graph mode it will return the Operation that does the assignment, and when in eager mode it will return None.

assign_sub(delta, use_locking=None, name=None, read_value=True)#

Subtracts a value from this variable.

Parameters:
  • delta – A Tensor. The value to subtract from this variable.

  • use_locking – If True, use locking during the operation.

  • name – The name to use for the operation.

  • read_value – A bool. Whether to read and return the new value of the variable or not.

Returns:

If read_value is True, this method will return the new value of the variable after the assignment has completed. Otherwise, when in graph mode it will return the Operation that does the assignment, and when in eager mode it will return None.

batch_scatter_update(sparse_delta, use_locking=False, name=None)#

Assigns tf.IndexedSlices to this variable batch-wise.

Analogous to batch_gather. This assumes that this variable and the sparse_delta IndexedSlices have a series of leading dimensions that are the same for all of them, and the updates are performed on the last dimension of indices. In other words, the dimensions should be the following:

num_prefix_dims = sparse_delta.indices.ndims - 1 batch_dim = num_prefix_dims + 1 `sparse_delta.updates.shape = sparse_delta.indices.shape + var.shape[

batch_dim:]`

where

sparse_delta.updates.shape[:num_prefix_dims] == sparse_delta.indices.shape[:num_prefix_dims] == var.shape[:num_prefix_dims]

And the operation performed can be expressed as:

`var[i_1, …, i_n,
sparse_delta.indices[i_1, …, i_n, j]] = sparse_delta.updates[

i_1, …, i_n, j]`

When sparse_delta.indices is a 1D tensor, this operation is equivalent to scatter_update.

To avoid this operation one can looping over the first ndims of the variable and using scatter_update on the subtensors that result of slicing the first dimension. This is a valid option for ndims = 1, but less efficient than this implementation.

Parameters:
  • sparse_deltatf.IndexedSlices to be assigned to this variable.

  • use_locking – If True, use locking during the operation.

  • name – the name of the operation.

Returns:

The updated variable.

Raises:

TypeError – if sparse_delta is not an IndexedSlices.

property constraint#

Returns the constraint function associated with this variable.

Returns:

The constraint function that was passed to the variable constructor. Can be None if no constraint was passed.

count_up_to(limit)#

Increments this variable until it reaches limit. (deprecated)

Deprecated: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Prefer Dataset.range instead.

When that Op is run it tries to increment the variable by 1. If incrementing the variable would bring it above limit then the Op raises the exception OutOfRangeError.

If no error is raised, the Op outputs the value of the variable before the increment.

This is essentially a shortcut for count_up_to(self, limit).

Parameters:

limit – value at which incrementing the variable raises an error.

Returns:

A Tensor that will hold the variable value before the increment. If no other Op modifies this variable, the values produced will all be distinct.

property create#

The op responsible for initializing this variable.

property device#

The device this variable is on.

property dtype: DType#

The dtype of the object.

eval(session=None)#

Evaluates and returns the value of this variable.

experimental_ref()#

DEPRECATED FUNCTION

Deprecated: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use ref() instead.

classmethod from_asdf(asdf_obj, *, reuse_params=None)#

Load an object from an asdf file.

Args#

asdf_obj: Object reuse_params:​If parameters, the parameters

will be reused if they are given. If a parameter is given, it will be used as the parameter with the same name. If a parameter is not given, a new parameter will be created.​

classmethod from_dict(dict_, *, reuse_params=None)#

Creates an object from a dictionary structure as generated by to_dict.

Parameters:
  • dict – Dictionary structure.

  • reuse_params – ​If parameters, the parameters will be reused if they are given. If a parameter is given, it will be used as the parameter with the same name. If a parameter is not given, a new parameter will be created.​

Returns:

The deserialized object.

classmethod from_json(cls, json, *, reuse_params=None)#

Load an object from a json string.

Parameters:
  • json (str) – Serialized object in a JSON string.

  • reuse_params – ​If parameters, the parameters will be reused if they are given. If a parameter is given, it will be used as the parameter with the same name. If a parameter is not given, a new parameter will be created.​

Return type:

object

Returns:

The deserialized object.

gather_nd(indices, name=None)#

Reads the value of this variable sparsely, using gather_nd.

get_cache_deps(only_floating=True)#

Return a set of all independent Parameter that this object depends on.

Parameters:

only_floating (bool) – If True, only return floating Parameter

Return type:

OrderedSet

get_dependencies(only_floating: bool = True) ztyping.DependentsType#

DEPRECATED FUNCTION

Deprecated: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use get_params instead if you want to retrieve the independent parameters or get_cache_deps in case you need the numerical cache dependents (advanced). :rtype: OrderedSet

classmethod get_repr()#

Abstract representation of the object for serialization.

This objects knows how to serialize and deserialize the object and is used by the to_json, from_json, to_dict and from_dict methods.

Returns:

The representation of the object.

Return type:

pydantic.BaseModel

get_shape()#

Alias of Variable.shape. :rtype: TensorShape

property graph#

The Graph of this variable.

property handle#

The handle by which this variable can be accessed.

property initial_value#

Returns the Tensor used as the initial value for the variable.

initialized_value()#

Returns the value of the initialized variable. (deprecated)

Deprecated: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Use Variable.read_value. Variables in 2.X are initialized automatically both in eager and graph (inside tf.defun) contexts.

You should use this instead of the variable itself to initialize another variable with a value that depends on the value of this variable.

`python # Initialize 'v' with a random tensor. v = tf.Variable(tf.random.truncated_normal([10, 40])) # Use `initialized_value` to guarantee that `v` has been # initialized before its value is used to initialize `w`. # The random values are picked only once. w = tf.Variable(v.initialized_value() * 2.0) `

Returns:

A Tensor holding the value of this variable after its initializer has run.

property initializer#

The op responsible for initializing this variable.

is_initialized(name=None)#

Checks whether a resource variable has been initialized.

Outputs boolean scalar indicating whether the tensor has been initialized.

Parameters:

name – A name for the operation (optional).

Returns:

A Tensor of type bool.

load(value, session=None)#

Load new value into this variable. (deprecated)

Deprecated: THIS FUNCTION IS DEPRECATED. It will be removed in a future version. Instructions for updating: Prefer Variable.assign which has equivalent behavior in 2.X.

Writes new value to variable’s memory. Doesn’t add ops to the graph.

This convenience method requires a session where the graph containing this variable has been launched. If no session is passed, the default session is used. See tf.compat.v1.Session for more information on launching a graph and on sessions.

```python v = tf.Variable([1, 2]) init = tf.compat.v1.global_variables_initializer()

with tf.compat.v1.Session() as sess:

sess.run(init) # Usage passing the session explicitly. v.load([2, 3], sess) print(v.eval(sess)) # prints [2 3] # Usage with the default session. The ‘with’ block # above makes ‘sess’ the default session. v.load([3, 4], sess) print(v.eval()) # prints [3 4]

```

Parameters:
  • value – New variable value

  • session – The session to use to evaluate this variable. If none, the default session is used.

Raises:

ValueError – Session is not passed and no default session

property op: Operation#

The op for this variable.

read_value_no_copy()#

Constructs an op which reads the value of this variable without copy.

The variable is read without making a copy even when it has been sparsely accessed. Variables in copy-on-read mode will be converted to copy-on-write mode.

Returns:

The value of the variable.

ref()#

Returns a hashable reference object to this Variable.

The primary use case for this API is to put variables in a set/dictionary. We can’t put variables in a set/dictionary as variable.__hash__() is no longer available starting Tensorflow 2.0.

The following will raise an exception starting 2.0

>>> x = tf.Variable(5)
>>> y = tf.Variable(10)
>>> z = tf.Variable(10)
>>> variable_set = {x, y, z}
Traceback (most recent call last):
  ...
TypeError: Variable is unhashable. Instead, use tensor.ref() as the key.
>>> variable_dict = {x: 'five', y: 'ten'}
Traceback (most recent call last):
  ...
TypeError: Variable is unhashable. Instead, use tensor.ref() as the key.

Instead, we can use variable.ref().

>>> variable_set = {x.ref(), y.ref(), z.ref()}
>>> x.ref() in variable_set
True
>>> variable_dict = {x.ref(): 'five', y.ref(): 'ten', z.ref(): 'ten'}
>>> variable_dict[y.ref()]
'ten'

Also, the reference object provides deref() function that returns the original Variable.

>>> x = tf.Variable(5)
>>> x.ref().deref()
<tf.Variable 'Variable:0' shape=() dtype=int32, numpy=5>
register_cacher(cacher)#

Register a cacher that caches values produces by this instance; a dependent.

Parameters:

cacher (ztyping.CacherOrCachersType) –

reset_cache_self()#

Clear the cache of self and all dependent cachers.

scatter_add(sparse_delta, use_locking=False, name=None)#

Adds tf.IndexedSlices to this variable.

Parameters:
  • sparse_deltatf.IndexedSlices to be added to this variable.

  • use_locking – If True, use locking during the operation.

  • name – the name of the operation.

Returns:

The updated variable.

Raises:

TypeError – if sparse_delta is not an IndexedSlices.

scatter_div(sparse_delta, use_locking=False, name=None)#

Divide this variable by tf.IndexedSlices.

Parameters:
  • sparse_deltatf.IndexedSlices to divide this variable by.

  • use_locking – If True, use locking during the operation.

  • name – the name of the operation.

Returns:

The updated variable.

Raises:

TypeError – if sparse_delta is not an IndexedSlices.

scatter_max(sparse_delta, use_locking=False, name=None)#

Updates this variable with the max of tf.IndexedSlices and itself.

Parameters:
  • sparse_deltatf.IndexedSlices to use as an argument of max with this variable.

  • use_locking – If True, use locking during the operation.

  • name – the name of the operation.

Returns:

The updated variable.

Raises:

TypeError – if sparse_delta is not an IndexedSlices.

scatter_min(sparse_delta, use_locking=False, name=None)#

Updates this variable with the min of tf.IndexedSlices and itself.

Parameters:
  • sparse_deltatf.IndexedSlices to use as an argument of min with this variable.

  • use_locking – If True, use locking during the operation.

  • name – the name of the operation.

Returns:

The updated variable.

Raises:

TypeError – if sparse_delta is not an IndexedSlices.

scatter_mul(sparse_delta, use_locking=False, name=None)#

Multiply this variable by tf.IndexedSlices.

Parameters:
  • sparse_deltatf.IndexedSlices to multiply this variable by.

  • use_locking – If True, use locking during the operation.

  • name – the name of the operation.

Returns:

The updated variable.

Raises:

TypeError – if sparse_delta is not an IndexedSlices.

scatter_nd_add(indices, updates, name=None)#

Applies sparse addition to individual values or slices in a Variable.

ref is a Tensor with rank P and indices is a Tensor of rank Q.

indices must be integer tensor, containing indices into ref. It must be shape [d_0, ..., d_{Q-2}, K] where 0 < K <= P.

The innermost dimension of indices (with length K) corresponds to indices into elements (if K = P) or slices (if K < P) along the K`th dimension of `ref.

updates is Tensor of rank Q-1+P-K with shape:

` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. `

For example, say we want to add 4 scattered elements to a rank-1 tensor to 8 elements. In Python, that update would look like this:

```python

ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8]) indices = tf.constant([[4], [3], [1] ,[7]]) updates = tf.constant([9, 10, 11, 12]) add = ref.scatter_nd_add(indices, updates) with tf.compat.v1.Session() as sess:

print sess.run(add)

```

The resulting update to ref would look like this:

[1, 13, 3, 14, 14, 6, 7, 20]

See tf.scatter_nd for more details about how to make updates to slices.

Parameters:
  • indices – The indices to be used in the operation.

  • updates – The values to be used in the operation.

  • name – the name of the operation.

Returns:

The updated variable.

scatter_nd_max(indices, updates, name=None)#

Updates this variable with the max of tf.IndexedSlices and itself.

ref is a Tensor with rank P and indices is a Tensor of rank Q.

indices must be integer tensor, containing indices into ref. It must be shape [d_0, ..., d_{Q-2}, K] where 0 < K <= P.

The innermost dimension of indices (with length K) corresponds to indices into elements (if K = P) or slices (if K < P) along the K`th dimension of `ref.

updates is Tensor of rank Q-1+P-K with shape:

` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. `

See tf.scatter_nd for more details about how to make updates to slices.

Parameters:
  • indices – The indices to be used in the operation.

  • updates – The values to be used in the operation.

  • name – the name of the operation.

Returns:

The updated variable.

scatter_nd_min(indices, updates, name=None)#

Updates this variable with the min of tf.IndexedSlices and itself.

ref is a Tensor with rank P and indices is a Tensor of rank Q.

indices must be integer tensor, containing indices into ref. It must be shape [d_0, ..., d_{Q-2}, K] where 0 < K <= P.

The innermost dimension of indices (with length K) corresponds to indices into elements (if K = P) or slices (if K < P) along the K`th dimension of `ref.

updates is Tensor of rank Q-1+P-K with shape:

` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. `

See tf.scatter_nd for more details about how to make updates to slices.

Parameters:
  • indices – The indices to be used in the operation.

  • updates – The values to be used in the operation.

  • name – the name of the operation.

Returns:

The updated variable.

scatter_nd_sub(indices, updates, name=None)#

Applies sparse subtraction to individual values or slices in a Variable.

ref is a Tensor with rank P and indices is a Tensor of rank Q.

indices must be integer tensor, containing indices into ref. It must be shape [d_0, ..., d_{Q-2}, K] where 0 < K <= P.

The innermost dimension of indices (with length K) corresponds to indices into elements (if K = P) or slices (if K < P) along the K`th dimension of `ref.

updates is Tensor of rank Q-1+P-K with shape:

` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. `

For example, say we want to add 4 scattered elements to a rank-1 tensor to 8 elements. In Python, that update would look like this:

```python

ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8]) indices = tf.constant([[4], [3], [1] ,[7]]) updates = tf.constant([9, 10, 11, 12]) op = ref.scatter_nd_sub(indices, updates) with tf.compat.v1.Session() as sess:

print sess.run(op)

```

The resulting update to ref would look like this:

[1, -9, 3, -6, -6, 6, 7, -4]

See tf.scatter_nd for more details about how to make updates to slices.

Parameters:
  • indices – The indices to be used in the operation.

  • updates – The values to be used in the operation.

  • name – the name of the operation.

Returns:

The updated variable.

scatter_nd_update(indices, updates, name=None)#

Applies sparse assignment to individual values or slices in a Variable.

ref is a Tensor with rank P and indices is a Tensor of rank Q.

indices must be integer tensor, containing indices into ref. It must be shape [d_0, ..., d_{Q-2}, K] where 0 < K <= P.

The innermost dimension of indices (with length K) corresponds to indices into elements (if K = P) or slices (if K < P) along the K`th dimension of `ref.

updates is Tensor of rank Q-1+P-K with shape:

` [d_0, ..., d_{Q-2}, ref.shape[K], ..., ref.shape[P-1]]. `

For example, say we want to add 4 scattered elements to a rank-1 tensor to 8 elements. In Python, that update would look like this:

```python

ref = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8]) indices = tf.constant([[4], [3], [1] ,[7]]) updates = tf.constant([9, 10, 11, 12]) op = ref.scatter_nd_update(indices, updates) with tf.compat.v1.Session() as sess:

print sess.run(op)

```

The resulting update to ref would look like this:

[1, 11, 3, 10, 9, 6, 7, 12]

See tf.scatter_nd for more details about how to make updates to slices.

Parameters:
  • indices – The indices to be used in the operation.

  • updates – The values to be used in the operation.

  • name – the name of the operation.

Returns:

The updated variable.

scatter_sub(sparse_delta, use_locking=False, name=None)#

Subtracts tf.IndexedSlices from this variable.

Parameters:
  • sparse_deltatf.IndexedSlices to be subtracted from this variable.

  • use_locking – If True, use locking during the operation.

  • name – the name of the operation.

Returns:

The updated variable.

Raises:

TypeError – if sparse_delta is not an IndexedSlices.

scatter_update(sparse_delta, use_locking=False, name=None)#

Assigns tf.IndexedSlices to this variable.

Parameters:
  • sparse_deltatf.IndexedSlices to be assigned to this variable.

  • use_locking – If True, use locking during the operation.

  • name – the name of the operation.

Returns:

The updated variable.

Raises:

TypeError – if sparse_delta is not an IndexedSlices.

property shape#

The shape of this variable.

sparse_read(indices, name=None)#

Reads the value of this variable sparsely, using gather.

to_asdf()#

Convert the object to an asdf file.

to_dict()#

Convert the object to a nested dictionary structure.

Returns:

The dictionary structure.

Return type:

dict

to_json()#

Convert the object to a json string.

Returns:

The json string.

Return type:

str

to_proto(export_scope=None)#

Converts a ResourceVariable to a VariableDef protocol buffer.

Parameters:

export_scope – Optional string. Name scope to remove.

Raises:

RuntimeError – If run in EAGER mode.

Returns:

A VariableDef protocol buffer, or None if the Variable is not in the specified name scope.

to_yaml()#

Convert the object to a yaml string.

Returns:

The yaml string.

Return type:

str