def visitSearchSpaceSum(self, sum: SearchSpaceSum, path: str, counter=None):
if len(sum.sub_spaces) == 1:
return accept(sum.sub_spaces[0], self, "")
else:
unique_name: str = self.get_unique_name("choice")
search_spaces = hp.choice(
unique_name,
[{str(i): accept(m, self, "")} for i, m in enumerate(sum.sub_spaces)],
)
return search_spaces
def visitSearchSpaceSum(self,
op: SearchSpaceSum) -> SearchSpaceGridInternalType:
sub_spaces: List[SearchSpace] = op.sub_spaces
sub_grids: Iterable[SearchSpaceGridInternalType] = [
accept(cur_space, self) for cur_space in sub_spaces
]
if len(sub_spaces) == 1:
return list(sub_grids)[0]
else:
fixed_grids: Iterable[List[SearchSpaceGrid]] = (
SearchSpaceToGridVisitor.fixupDegenerateSearchSpaces(grid)
for grid in sub_grids)
final_grids: List[SearchSpaceGrid] = []
for i, grids in enumerate(fixed_grids):
if not grids:
grids = [{}]
else:
# we need to add in this nesting
# in case a higher order operator directly contains
# another
grids = nest_choice_all_HPparams(grids)
discriminated_grids: List[SearchSpaceGrid] = [{
**d, discriminant_name:
SearchSpaceConstant(i)
} for d in grids]
final_grids.extend(discriminated_grids)
return final_grids
def visitSearchSpaceObject(
self, space: SearchSpaceObject) -> List[SearchSpaceGrid]:
keys = space.keys
keys_len = len(keys)
final_choices: List[SearchSpaceGrid] = []
for c in space.choices:
assert keys_len == len(c)
kvs_complex: List[List[SearchSpaceGrid]] = []
kvs_simple: SearchSpaceGrid = {}
for k, v in zip(keys, c):
vspace: Union[List[SearchSpaceGrid],
SearchSpacePrimitive] = accept(v, self)
if isinstance(vspace, SearchSpacePrimitive):
kvs_simple[k] = vspace
else:
nested_vspace: List[SearchSpaceGrid] = nest_all_HPparams(
k, vspace)
if nested_vspace:
kvs_complex.append(nested_vspace)
nested_space_choices: Iterable[
Iterable[SearchSpaceGrid]] = itertools.product(*kvs_complex)
nested_space_choices_lists: List[List[SearchSpaceGrid]] = list(
map((lambda x: list(x)), nested_space_choices))
nested_space_choices_filtered: List[List[SearchSpaceGrid]] = [
ll for ll in nested_space_choices_lists if ll
]
if nested_space_choices_filtered:
chained_grids: Iterable[SearchSpaceGrid] = [
dict(ChainMap(*nested_choice, kvs_simple))
for nested_choice in nested_space_choices_filtered
]
final_choices.extend(chained_grids)
else:
final_choices.append(kvs_simple)
return final_choices
def _searchSpaceList(self, space: SearchSpaceArray, *,
size: int) -> List[SearchSpaceGrid]:
sub_spaces = space.items(max=size)
param_grids: List[List[SearchSpaceGrid]] = [
nest_all_HPparams(
str(index), self.fixupDegenerateSearchSpaces(accept(sub,
self)))
for index, sub in enumerate(sub_spaces)
]
param_grids_product: Iterable[
Iterable[SearchSpaceGrid]] = itertools.product(*param_grids)
chained_grids: List[SearchSpaceGrid] = [
dict(ChainMap(*gridline, )) for gridline in param_grids_product
]
if space.is_tuple:
st_val = structure_type_tuple
else:
st_val = structure_type_list
discriminated_grids: List[SearchSpaceGrid] = [{
**d, structure_type_name:
SearchSpaceConstant(st_val)
} for d in chained_grids]
return discriminated_grids
def visitSearchSpaceProduct(
self, prod: SearchSpaceProduct, path: str, counter=None
):
search_spaces = [
accept(space, self, self.get_unique_name(make_indexed_name(name, index)))
for name, index, space in prod.get_indexed_spaces()
]
return search_spaces
def visitSearchSpaceDict(
self, sd: SearchSpaceDict, path: str, counter=None, useCounter=True
):
search_spaces = (
name + ":" + accept(space, self, path + "_" + name)
for name, space in sd.space_dict.items()
)
return "{" + ",".join(search_spaces) + "}"
def visitSearchSpaceProduct(
self, prod: SearchSpaceProduct, path: str, counter=None, useCounter=True
):
search_spaces = (
accept(space, self, self.get_unique_name(make_indexed_name(name, index)))
for name, index, space in prod.get_indexed_spaces()
)
return "[" + ",".join(search_spaces) + "]"
def visitSearchSpaceDict(self,
sd: SearchSpaceDict,
path: str,
counter=None):
search_spaces = {
name: accept(space, self, path + "_" + name)
for name, space in sd.space_dict.items()
}
return search_spaces
def visitOperatorChoice(self, op: Ops.OperatorChoice) -> Dict[str, Any]:
defaults_list: Iterable[Dict[str, Any]] = (accept(s, self)
for s in op.steps())
defaults: Dict[str, Any] = {}
for d in defaults_list:
defaults.update(d)
return defaults
def visitSearchSpaceSum(
self, sum_space: SearchSpaceSum, path: str, counter=None, useCounter=True
):
unique_name: str = self.get_unique_name("choice")
sub_str: Iterable[str] = (
'"' + str(i) + '"' + " : " + '"' + accept(m, self, "") + '"'
for i, m in enumerate(sum_space.sub_spaces)
)
sub_spaces_str: str = "[" + ",".join(sub_str) + "]"
return f"hp.choice({unique_name}, {sub_spaces_str})"
def visitPipeline(self, op: Ops.PlannedPipeline) -> Dict[str, Any]:
defaults_list: Iterable[Dict[str, Any]] = (nest_HPparams(
s.name(), accept(s, self)) for s in op.steps())
defaults: Dict[str, Any] = {}
for d in defaults_list:
defaults.update(d)
return defaults
def asexpr(key, e):
nonlocal child_counter
if e is None:
return None
else:
ee = accept(e, self, path + "_" + key, counter=child_counter)
if child_counter is None:
child_counter = 1
else:
child_counter = child_counter + 1
return ee
def array_single_str_(self,
space: SearchSpaceArray,
path: str,
num,
useCounter=True) -> str:
p = _mk_label(path, num, useCounter=useCounter) + "_"
ret = "(" if space.is_tuple else "["
items: Iterable[SearchSpace] = space.items()
ret += ",".join((accept(sub, self, p, counter=x, useCounter=useCounter)
for x, sub in enumerate(items)))
ret += ")" if space.is_tuple else "]"
return ret
def run(cls, space: SearchSpace, name: str, counter=None, useCounter=True):
visitor = cls(name)
ret: str = ""
body = accept(space, visitor, name, counter=counter, useCounter=useCounter)
if visitor.pgo_header is not None:
ret += visitor.pgo_header
if visitor.nested_header is not None:
ret += visitor.nested_header
if visitor.decls:
ret += visitor.decls + "\n"
ret += "return " + body
return ret
def cstr(key, x):
nonlocal child_counter
if x is None:
return "None"
else:
s = accept(
x, self, path + "_" + key, child_counter, useCounter=useCounter
)
if child_counter is None:
child_counter = 1
else:
child_counter = child_counter + 1
return s
def visitSearchSpaceProduct(
self, op: SearchSpaceProduct) -> SearchSpaceGridInternalType:
sub_spaces = op.get_indexed_spaces()
param_grids: List[List[SearchSpaceGrid]] = [
nest_all_HPparams(
make_indexed_name(name, index),
self.fixupDegenerateSearchSpaces(accept(space, self)),
) for name, index, space in sub_spaces
]
param_grids_product: Iterable[
Iterable[SearchSpaceGrid]] = itertools.product(*param_grids)
chained_grids: List[SearchSpaceGrid] = [
dict(ChainMap(*gridline)) for gridline in param_grids_product
]
return chained_grids
def visitSearchSpaceDict(
self, op: SearchSpaceDict) -> SearchSpaceGridInternalType:
sub_spaces = op.space_dict.items()
param_grids: List[List[SearchSpaceGrid]] = [
nest_all_HPparams(
name,
self.fixupDegenerateSearchSpaces(accept(space, self)),
) for name, space in sub_spaces
]
param_grids_product: Iterable[
Iterable[SearchSpaceGrid]] = itertools.product(*param_grids)
chained_grids: List[SearchSpaceGrid] = [
dict(ChainMap(*gridline)) for gridline in param_grids_product
]
discriminated_grids: List[SearchSpaceGrid] = [{
**d, structure_type_name:
SearchSpaceConstant(structure_type_dict)
} for d in chained_grids]
return discriminated_grids
def visitOperatorChoice(self, op: "OperatorChoice") -> SearchSpace:
spaces: List[SearchSpace] = [accept(s, self) for s in op.steps()]
return SearchSpaceSum(spaces)
def visitPlannedPipeline(self, op: "PlannedPipeline") -> SearchSpace:
spaces: List[Tuple[str, SearchSpace]] = [(s.name(), accept(s, self))
for s in op.steps()]
return SearchSpaceProduct(spaces)
def run(cls,
op: PlannedOperator,
pgo: Optional[PGO] = None,
data_schema={}) -> SearchSpace:
visitor = cls(pgo=pgo, data_schema=data_schema)
return accept(op, visitor)
def run(cls, space: SearchSpace, name: str):
visitor = cls(name)
return accept(space, visitor, name)
def array_single_expr_(self, space: SearchSpaceArray, path: str, num):
p = _mk_label(path, num) + "_"
items: Iterable[SearchSpace] = space.items()
ret = [accept(sub, self, p, counter=x) for x, sub in enumerate(items)]
return tuple(ret) if space.is_tuple else ret
def run(cls, space: SearchSpace) -> List[SearchSpaceGrid]:
visitor = cls()
grids: SearchSpaceGridInternalType = accept(space, visitor)
fixed_grids = cls.fixupDegenerateSearchSpaces(grids)
return fixed_grids
def schemaToSearchSpaceHelper_(
self,
longName,
path: str,
schema: JsonSchema,
relevantFields: Optional[Set[str]],
pgo_freqs: pgo_part = None,
sub_space: bool = True,
) -> Optional[SearchSpace]:
# TODO: handle degenerate cases
# right now, this handles only a very fixed form
if is_false_schema(schema):
return None
typ: Optional[str] = None
typ = schema.get("laleType", None)
if typ is None:
typ = schema.get("type", None)
else:
typ = typ
if "enum" in schema and typ != "operator":
vals = schema["enum"]
return SearchSpaceEnum(vals,
pgo=asFreqs(pgo_freqs),
default=get_default(schema))
if typ is not None:
if typ == "boolean":
return SearchSpaceBool(pgo=asFreqs(pgo_freqs),
default=get_default(schema))
elif typ == "number" or typ == "integer":
exclusive_minimum = False
minimum = schema.get("minimumForOptimizer", None)
if minimum is not None:
exclusive_minimum = schema.get(
"exclusiveMinimumForOptimizer", False)
else:
minimum = schema.get("minimum", None)
if minimum is not None:
exclusive_minimum = schema.get("exclusiveMinimum",
False)
exclusive_maximum = False
maximum = schema.get("maximumForOptimizer", None)
if maximum is not None:
exclusive_maximum = schema.get(
"exclusiveMaximumForOptimizer", False)
else:
maximum = schema.get("maximum", None)
if maximum is not None:
exclusive_maximum = schema.get("exclusiveMaximum",
False)
distribution = schema.get("distribution", None)
laleType = schema.get("laleType", None)
if laleType is None:
laleType = typ
if laleType == "number":
discrete = False
elif laleType == "integer":
discrete = True
else:
raise OperatorSchemaError(
path,
f"specified laleType should be a number or integer, not: {laleType}.",
)
pgo: Freqs
return SearchSpaceNumber(
minimum=minimum,
exclusiveMinimum=exclusive_minimum,
maximum=maximum,
exclusiveMaximum=exclusive_maximum,
discrete=discrete,
distribution=distribution,
pgo=asFreqs(pgo_freqs),
default=get_default(schema),
)
elif typ == "array" or typ == "tuple":
laleType = schema.get("laleType", None)
if laleType is None:
laleType = typ
is_tuple: bool = laleType == "tuple"
min_items = schema.get("minItemsForOptimizer", None)
if min_items is None:
min_items = schema.get("minItems", None)
if min_items is None:
min_items = 0
max_items = schema.get("maxItemsForOptimizer", None)
if max_items is None:
max_items = schema.get("maxItems", None)
items_schema = schema.get("itemsForOptimizer", None)
if items_schema is None:
items_schema = schema.get("items", None)
if items_schema is None:
raise OperatorSchemaError(
path,
f"An array type was found without a provided schema for the items in the schema {schema}. Please provide a schema for the items (consider using itemsForOptimizer)",
)
# we can search an empty list even without schemas
if max_items == 0:
if is_tuple:
return SearchSpaceConstant([()])
else:
return SearchSpaceConstant([[]])
prefix: Optional[List[SearchSpace]] = None
additional: Optional[SearchSpace] = None
if isinstance(items_schema, list):
prefix = []
for i, sub_schema in enumerate(items_schema):
sub = self.schemaToSearchSpaceHelper_(
longName, path + "_" + str(i), sub_schema,
relevantFields)
if sub is None:
return None
else:
prefix.append(sub)
prefix_len = len(prefix)
additional_items_schema = schema.get(
"additionalItemsForOptimizer", None)
if additional_items_schema is None:
additional_items_schema = schema.get(
"additionalItems", None)
if additional_items_schema is None:
if max_items is None or max_items > prefix_len:
raise OperatorSchemaError(
path,
f"An array type was found with provided schemas for {prefix_len} elements, but either an unspecified or too high a maxItems, and no schema for the additionalItems. Please constraing maxItems to <= {prefix_len} (you can set maxItemsForOptimizer), or provide a schema for additionalItems",
)
elif additional_items_schema is False:
if max_items is None:
max_items = prefix_len
else:
max_items = min(max_items, prefix_len)
else:
additional = self.schemaToSearchSpaceHelper_(
longName,
path + "-",
additional_items_schema,
relevantFields,
)
# if items_schema is None:
# raise ValueError(f"an array type was found without a provided schema for the items in the schema {schema}. Please provide a schema for the items (consider using itemsForOptimizer)")
else:
additional = self.schemaToSearchSpaceHelper_(
longName, path + "-", items_schema, relevantFields)
if max_items is None:
raise OperatorSchemaError(
path,
f"An array type was found without a provided maximum number of items in the schema {schema}, and it is not a list with 'additionalItems' set to False. Please provide a maximum (consider using maxItemsForOptimizer), or, if you are using a list, set additionalItems to False",
)
return SearchSpaceArray(
prefix=prefix,
minimum=min_items,
maximum=max_items,
additional=additional,
is_tuple=is_tuple,
)
elif typ == "object":
if "properties" not in schema:
return SearchSpaceObject(longName, [], [])
o = self.JsonSchemaToSearchSpaceHelper(
longName,
path,
schema,
relevantFields,
pgo_freqs=pgo_freqs,
sub_space=sub_space,
)
if sub_space:
return SearchSpaceDict(o)
else:
all_keys = list(o.keys())
all_keys.sort()
o_choice = tuple([o.get(k, None) for k in all_keys])
return SearchSpaceObject(longName, all_keys, [o_choice])
elif typ == "string":
pass
elif typ == "operator":
# TODO: If there is a default, we could use it
vals = schema.get("enum", None)
if vals is None:
logger.error(
"An operator is required by the schema but was not provided"
)
return None
sub_schemas: List[SearchSpace] = [
accept(op, self)
if isinstance(op, Operator) else SearchSpaceConstant(op)
for op in vals
]
combined_sub_schema: SearchSpace
if len(sub_schemas) == 1:
combined_sub_schema = sub_schemas[0]
if isinstance(combined_sub_schema, SearchSpaceConstant):
return combined_sub_schema
else:
combined_sub_schema = SearchSpaceSum(sub_schemas)
if all((isinstance(x, SearchSpaceConstant)
for x in sub_schemas)):
return combined_sub_schema
return SearchSpaceOperator(combined_sub_schema)
elif typ == "Any":
raise OperatorSchemaError(
path,
f"A search space was found with laleType ({typ}), which is not searchable. Please mark the relevant hyperparameter as not relevant for the optimizer. schema: {schema}",
)
else:
raise OperatorSchemaError(
path,
f"An unknown type ({typ}) was found in the schema {schema}"
)
if "anyOf" in schema:
objs = []
for s_obj in schema["anyOf"]:
if "type" in s_obj and s_obj["type"] == "object":
o = self.JsonSchemaToSearchSpaceHelper(
longName,
path,
s_obj,
relevantFields,
pgo_freqs=pgo_freqs,
sub_space=sub_space,
)
if o:
objs.append(o)
if objs:
# First, gather a list of all the properties
keys_list = [set(o.keys()) for o in objs]
# make sure the iterator is deterministic
all_keys = list(set.union(*keys_list))
# and we might as well make it sorted
all_keys.sort()
def as_str(k, c):
if c is None:
return "None"
else:
return search_space_to_str_for_comparison(
c, path + "_" + k)
anys: Dict[str, Any] = {}
for o in objs:
o_choice = tuple([o.get(k, None) for k in all_keys])
k = str([
as_str(all_keys[idx], c)
for idx, c in enumerate(o_choice)
])
if k in anys:
logger.info(
f"Ignoring Duplicate SearchSpace entry {k}")
anys[k] = o_choice
return SearchSpaceObject(longName, all_keys, anys.values())
else:
return SearchSpaceObject(longName, [], [])
if "allOf" in schema:
# if all but one are negated constraints, we will just ignore them
pos_sub_schema: List[JsonSchema] = []
for sub_schema in schema["allOf"]:
if "not" not in sub_schema:
pos_sub_schema.append(sub_schema)
if len(pos_sub_schema) > 1:
raise OperatorSchemaError(
path,
f"schemaToSearchSpaceHelper does not yet know how to compile the given schema {schema}, because it is an allOf with more than one non-negated schemas ({pos_sub_schema})",
)
if len(pos_sub_schema) == 0:
raise OperatorSchemaError(
path,
f"schemaToSearchSpaceHelper does not yet know how to compile the given schema {schema}, because it is an allOf with only negated schemas",
)
logger.debug(
f"[{path}]: schemaToSearchSpaceHelper: ignoring negated schemas in the conjunction {schema}"
)
return self.schemaToSearchSpaceHelper_(
longName,
path,
pos_sub_schema[0],
relevantFields,
pgo_freqs=pgo_freqs,
sub_space=sub_space,
)
# TODO: handle degenerate cases
raise OperatorSchemaError(
path,
f"schemaToSearchSpaceHelper does not yet know how to compile the given schema {schema}",
)
def run(cls, op: Ops.Operator) -> Dict[str, Any]:
visitor = cls()
return accept(op, visitor)
def visitSearchSpaceOperator(self,
op: SearchSpaceOperator,
path: str,
counter=None):
return scope.make_nested_hyperopt(accept(op.sub_space, self, path))
def run(cls,
op: PlannedOperator,
pgo: Optional[PGO] = None) -> SearchSpace:
visitor = cls(pgo=pgo)
return accept(op, visitor)
def visitSearchSpaceOperator(
self, op: SearchSpaceOperator) -> SearchSpaceGridInternalType:
return accept(op.sub_space, self)