pm4py.algo.querying.llm.abstractions package#

This program is free software: you can redistribute it and/or modify it under the terms of the GNU Affero General Public License as published by the Free Software Foundation, either version 3 of the License, or any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more details.

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

Submodules#

pm4py.algo.querying.llm.abstractions.case_to_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.case_to_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

INCLUDE_CASE_ATTRIBUTES = 'include_case_attributes'#

INCLUDE_EVENT_ATTRIBUTES = 'include_event_attributes'#

INCLUDE_TIMESTAMP = 'include_timestamp'#

INCLUDE_HEADER = 'include_header'#

ACTIVITY_KEY = 'pm4py:param:activity_key'#

TIMESTAMP_KEY = 'pm4py:param:timestamp_key'#

pm4py.algo.querying.llm.abstractions.case_to_descr.apply(case: Trace, parameters: Dict[Any, Any] | None = None) → str[source]#

Provides a textual abstraction of a single case (Trace object) of a traditional event log

Parameters#

case: Single case (Trace object) of a traditional event log
parameters: Parameters of the method, including: - Parameters.INCLUDING_CASE_ATTRIBUTES - Parameters.INCLUDE_EVENT_ATTRIBUTES - Parameters.INCLUDE_TIMESTAMP - Parameters.INCLUDE_HEADER => includes the header (or not) in the response) - Parameters.ACTIVITY_KEY => the attribute to be used as activity - Parameters.TIMESTAMP_KEY => the attribute to be used as timestamp

Returns#

stru: Textual abstraction of the case

pm4py.algo.querying.llm.abstractions.declare_to_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.declare_to_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

INCLUDE_HEADER = 'include_header'#

pm4py.algo.querying.llm.abstractions.declare_to_descr.get_model_description()[source]#

pm4py.algo.querying.llm.abstractions.declare_to_descr.get_model_implementation()[source]#

pm4py.algo.querying.llm.abstractions.declare_to_descr.apply(declare: Dict[str, Dict[Any, Dict[str, int]]], parameters: Dict[Any, Any] | None = None) → str[source]#

Gets a textual abstraction of a DECLARE model

Parameters#

declare: DECLARE model
parameters: Possible parameters of the algorithm, including: - Parameters.INCLUDE_HEADER => include the header of the response

Returns#

stru: Textual abstraction

pm4py.algo.querying.llm.abstractions.log_to_cols_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.log_to_cols_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

MAX_LEN = 'max_len'#

CASE_ID_KEY = 'pm4py:param:case_id_key'#

pm4py.algo.querying.llm.abstractions.log_to_cols_descr.apply(log_obj: EventLog | EventStream | DataFrame, parameters: Dict[Any, Any] | None = None) → str[source]#

pm4py.algo.querying.llm.abstractions.log_to_dfg_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.log_to_dfg_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

INCLUDE_FREQUENCY = 'include_frequency'#

INCLUDE_PERFORMANCE = 'include_performance'#

MAX_LEN = 'max_len'#

RELATIVE_FREQUENCY = 'relative_frequency'#

RESPONSE_HEADER = 'response_header'#

PRIMARY_PERFORMANCE_AGGREGATION = 'primary_performance_aggregation'#

SECONDARY_PERFORMANCE_AGGREGATION = 'secondary_performance_aggregation'#

ACTIVITY_KEY = 'pm4py:param:activity_key'#

TIMESTAMP_KEY = 'pm4py:param:timestamp_key'#

CASE_ID_KEY = 'pm4py:param:case_id_key'#

pm4py.algo.querying.llm.abstractions.log_to_dfg_descr.abstraction_from_frequency_performance_dfg(freq_dfg: Dict[Tuple[str, str], int], perf_dfg: Dict[Tuple[str, str], Dict[str, float]], parameters: Dict[Any, Any] | None = None) → str[source]#

Obtains the abstraction starting from the knowledge of the frequency of the paths, and their performance.

Minimal viable example:

Parameters#

freq_dfg

Dictionary associating to each path its frequency

perf_dfg

Dictionary associating to each path its performance. A path (‘A’, ‘B’) is associated to a dictionary containing performance metrics, i.e. (‘A’, ‘B’): {‘mean’: 86400, ‘stdev’: 86400} means that: - the average time between the activities A and B is 1 day - also the standard deviation of the times between A and B is 1 day

parameters

Optional parameters of the algorithm, including:

Parameters.RELATIVE_FREQUENCY => (boolean) decides if the frequency DFG should be normalized to a relative
frequency
Parameters.INCLUDE_FREQUENCY => includes the frequency of the arcs in the textual abstraction
Parameters.INCLUDE_PERFORMANCE => includes the performance of the arcs in the textual abstraction
Parameters.MAX_LEN => desidered length of the textual abstraction
Parameters.RESPONSE_HEADER => includes an header in the textual abstraction, which explains the context
Parameters.PRIMARY_PERFORMANCE_AGGREGATION => primary performance metric to be used to express the performance of the arcs (e.g., mean). Available options: mean, median, stdev, min, max, sum
Parameters.SECONDARY_PERFORMANCE_AGGREGATION => secondary performance metric to be used to express the performance of the arcs (e.g., stdev). Available options: mean, median, stdev, min, max, sum

Returns#

textual_abstraction: Textual abstraction

pm4py.algo.querying.llm.abstractions.log_to_dfg_descr.apply(log_obj: EventLog | EventStream | DataFrame, parameters: Dict[Any, Any] | None = None) → str[source]#

Gets the textual abstraction of the directly-follows graph computed on the provided log object.

Minimal viable example:

import pm4py from pm4py.algo.querying.llm.abstractions import log_to_dfg_descr

log = pm4py.read_xes(‘tests/input_data/running-example.xes’) print(log_to_dfg_descr.apply(log))

Parameters#

log_obj

Log object (event log / Pandas dataframe)

parameters

Optional Parameters of the algorithm, including:

Parameters.ACTIVITY_KEY => the attribute to be used as activity
Parameters.TIMESTAMP_KEY => the attribute to be used as timestamp
Parameters.CASE_ID_KEY => the attribute to be used as case ID
Parameters.RELATIVE_FREQUENCY => (boolean) decides if the frequency DFG should be normalized to a relative
frequency
Parameters.INCLUDE_FREQUENCY => includes the frequency of the arcs in the textual abstraction
Parameters.INCLUDE_PERFORMANCE => includes the performance of the arcs in the textual abstraction
Parameters.MAX_LEN => desidered length of the textual abstraction
Parameters.RESPONSE_HEADER => includes an header in the textual abstraction, which explains the context
Parameters.PRIMARY_PERFORMANCE_AGGREGATION => primary performance metric to be used to express the performance of the arcs (e.g., mean). Available options: mean, median, stdev, min, max, sum
Parameters.SECONDARY_PERFORMANCE_AGGREGATION => secondary performance metric to be used to express the performance of the arcs (e.g., stdev). Available options: mean, median, stdev, min, max, sum

Returns#

textual_abstraction: Textual abstraction

pm4py.algo.querying.llm.abstractions.log_to_fea_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.log_to_fea_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

INCLUDE_HEADER = 'include_header'#

MAX_LEN = 'max_len'#

pm4py.algo.querying.llm.abstractions.log_to_fea_descr.textual_abstraction_from_fea_df(fea_df: DataFrame, parameters: Dict[Any, Any] | None = None) → str[source]#

Returns the textual abstraction of ML features already encoded in a feature table

Minimum viable example:

Parameters#

fea_df: Feature table (numeric features; stored as Pandas dataframe)
parameters: Parameters that should be provided to the feature extraction, plus: - Parameters.INCLUDE_HEADER => includes a descriptive header in the returned text - Parameters.MAX_LEN => maximum length of the provided text (if necessary, only the most meaningful features are kept)

Returns#

stru: Textual abstraction

pm4py.algo.querying.llm.abstractions.log_to_fea_descr.apply(log: EventLog | EventStream | DataFrame, parameters: Dict[Any, Any] | None = None) → str[source]#

Returns the textual abstraction of ML features extracted from a traditional event log object.

Minimum viable example:

Parameters#

log: Event log / Pandas dataframe
parameters: Parameters that should be provided to the feature extraction, plus: - Parameters.INCLUDE_HEADER => includes a descriptive header in the returned text - Parameters.MAX_LEN => maximum length of the provided text (if necessary, only the most meaningful features are kept)

Returns#

stru: Textual abstraction

pm4py.algo.querying.llm.abstractions.log_to_variants_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.log_to_variants_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

INCLUDE_FREQUENCY = 'include_frequency'#

INCLUDE_PERFORMANCE = 'include_performance'#

MAX_LEN = 'max_len'#

RELATIVE_FREQUENCY = 'relative_frequency'#

RESPONSE_HEADER = 'response_header'#

PRIMARY_PERFORMANCE_AGGREGATION = 'primary_performance_aggregation'#

SECONDARY_PERFORMANCE_AGGREGATION = 'secondary_performance_aggregation'#

ACTIVITY_KEY = 'pm4py:param:activity_key'#

TIMESTAMP_KEY = 'pm4py:param:timestamp_key'#

CASE_ID_KEY = 'pm4py:param:case_id_key'#

pm4py.algo.querying.llm.abstractions.log_to_variants_descr.abstraction_from_variants_freq_perf_list(vars_list: List[Tuple[List[str], int, float, float]], parameters: Dict[Any, Any] | None = None) → str[source]#

Obtains a textual abstraction from a list of variants provided along their frequency and performance values. Each variant of the list is expressed in the form:

where (‘A’, ‘B’, ‘C’) is the tuple of activities executed in the variant, 1000 is the number of occurrences of this variant in the event log, 86400.0 is an aggregation (mean) of the throughput times of the cases belonging to this variant, 172800.0 is an aggregation (stdev, so standard deviation) of the throughput times of these cases.

Minimal viable example:

Parameters#

vars_list

List of variants, expressed as explained above

parameters

Optional parameters of the algorithm, including:

Parameters.RELATIVE_FREQUENCY => decides if the the frequency of the variants should be normalized to a relative
frequency
Parameters.PRIMARY_PERFORMANCE_AGGREGATION => primary performance metric to be used to express the performance of the arcs (e.g., mean). Available options: mean, median, stdev, min, max, sum
Parameters.SECONDARY_PERFORMANCE_AGGREGATION => secondary performance metric to be used to express the performance of the arcs (e.g., stdev). Available options: mean, median, stdev, min, max, sum
Parameters.MAX_LEN => desidered length of the textual abstraction
Parameters.RESPONSE_HEADER => includes an header in the textual abstraction, which explains the context
Parameters.INCLUDE_FREQUENCY => includes the frequency of the arcs in the textual abstraction
Parameters.INCLUDE_PERFORMANCE => includes the performance of the arcs in the textual abstraction

Returns#

textual_abstraction: Textual abstraction of the variants

pm4py.algo.querying.llm.abstractions.log_to_variants_descr.compute_perf_aggregation(perf_values: List[float], perf_agg: str) → float[source]#

Computes an aggregation of a list of performance values

Minimal viable example:: compute_perf_aggregation([3600.0, 7200.0], ‘mean’)

Parameters#

perf_values: List of performance values
perf_agg: Desired aggregation (mean, median, stdev, sum, min, max)

Returns#

agg_value: Aggregated value

pm4py.algo.querying.llm.abstractions.log_to_variants_descr.apply(log_obj: EventLog | EventStream | DataFrame, parameters: Dict[Any, Any] | None = None) → str[source]#

Gets the textual abstraction of the variants of a specified log object.

Minimal viable example:

Parameters#

log_obj

Log object

parameters

Optional parameters of the algorithm, including: - Parameters.ACTIVITY_KEY => the attribute of the log to be used as activity - Parameters.TIMESTAMP_KEY => the attribute of the log to be used as timestamp - Parameters.CASE_ID_KEY => the attribute of the log to be used as case identifier - Parameters.RELATIVE_FREQUENCY => decides if the the frequency of the variants should be normalized to a relative

Parameters.PRIMARY_PERFORMANCE_AGGREGATION => primary performance metric to be used to express the performance of the arcs (e.g., mean). Available options: mean, median, stdev, min, max, sum
Parameters.SECONDARY_PERFORMANCE_AGGREGATION => secondary performance metric to be used to express the performance of the arcs (e.g., stdev). Available options: mean, median, stdev, min, max, sum

Returns#

textual_abstraction: Textual abstraction of the variants of an event log object

pm4py.algo.querying.llm.abstractions.logske_to_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.logske_to_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

INCLUDE_HEADER = 'include_header'#

pm4py.algo.querying.llm.abstractions.logske_to_descr.get_model_description()[source]#

pm4py.algo.querying.llm.abstractions.logske_to_descr.get_model_implementation()[source]#

pm4py.algo.querying.llm.abstractions.logske_to_descr.apply(lsk: Dict[str, Any], parameters: Dict[Any, Any] | None = None) → str[source]#

pm4py.algo.querying.llm.abstractions.net_to_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.net_to_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

RESPONSE_HEADER = 'response_header'#

pm4py.algo.querying.llm.abstractions.net_to_descr.apply(net: PetriNet, im: Marking, fm: Marking, parameters: Dict[Any, Any] | None = None) → str[source]#

Provides the description of an accepting Petri net

Parameters#

net: Petri net
im: Initial marking
fm: Final marking
parameters: Possible parameters of the algorithm, including: - Parameters.INCLUDE_HEADER => includes the header

Returns#

stru: String representation of the given accepting Petri net

pm4py.algo.querying.llm.abstractions.ocel_fea_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.ocel_fea_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

INCLUDE_HEADER = 'include_header'#

MAX_LEN = 'max_len'#

DEBUG = 'debug'#

ENABLE_OBJECT_LIFECYCLE_PATHS = 'enable_object_lifecycle_paths'#

pm4py.algo.querying.llm.abstractions.ocel_fea_descr.textual_abstraction_from_fea_df(fea_df: DataFrame, parameters: Dict[Any, Any] | None = None) → str[source]#

pm4py.algo.querying.llm.abstractions.ocel_fea_descr.apply(ocel: OCEL, obj_type: str, parameters: Dict[Any, Any] | None = None) → str[source]#

pm4py.algo.querying.llm.abstractions.ocel_ocdfg_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.ocel_ocdfg_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

MAX_LEN = 'max_len'#

INCLUDE_HEADER = 'include_header'#

INCLUDE_PERFORMANCE = 'include_performance'#

pm4py.algo.querying.llm.abstractions.ocel_ocdfg_descr.apply(ocel: OCEL, parameters: Dict[Any, Any] | None = None) → str[source]#

pm4py.algo.querying.llm.abstractions.stream_to_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.stream_to_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

RESPONSE_HEADER = 'response_header'#

TIMESTAMP_KEY = 'pm4py:param:timestamp_key'#

MAX_LEN = 'max_len'#

pm4py.algo.querying.llm.abstractions.stream_to_descr.apply(log_obj: EventLog | EventStream | DataFrame, parameters: Dict[Any, Any] | None = None) → str[source]#

Given a log object, returns a representation of the (last) events of a stream corresponding to the log object.

Parameters#

log_obj: Log object
parameters: Parameters of the algorithm, including: - Parameters.RESPONSE_HEADER => includes the header in the response - Parameters.TIMESTAMP_KEY => the attribute to be used as timestamp - Parameters.MAX_LEN => maximum length of the resulting stream

Returns#

descr: String representing the stream of events

pm4py.algo.querying.llm.abstractions.tempprofile_to_descr module#

You should have received a copy of the GNU Affero General Public License along with this program. If not, see this software project’s root or visit <https://www.gnu.org/licenses/>.

Website: https://processintelligence.solutions Contact: info@processintelligence.solutions

class pm4py.algo.querying.llm.abstractions.tempprofile_to_descr.Parameters(value, names=<not given>, *values, module=None, qualname=None, type=None, start=1, boundary=None)[source]#

Bases: Enum

INCLUDE_HEADER = 'include_header'#

pm4py.algo.querying.llm.abstractions.tempprofile_to_descr.get_model_description()[source]#

pm4py.algo.querying.llm.abstractions.tempprofile_to_descr.get_model_implementation()[source]#

pm4py.algo.querying.llm.abstractions.tempprofile_to_descr.apply(temporal_profile: Dict[Tuple[str, str], Tuple[float, float]], parameters: Dict[Any, Any] | None = None) → str[source]#

Abstracts a temporal profile model to a string.

Parameters#

temporal_profile: Temporal profile
parameters: Parameters of the method, including: - Parameters.INCLUDE_HEADER => includes the header in the response

Returns#

text_abstr: Textual abstraction of the log skeleton