Added history class

Now the environment does not deal with history directly, it delegates it to a
specific class. The analysis also uses history instances instead of either
using the database directly or creating a proxy environment.

This should make it easier to change the implementation in the future.

In fact, the change was motivated by the large size of the csv files in previous
versions. This new implementation only stores results in deltas, and it fills
any necessary values when needed.
history
J. Fernando Sánchez 6 years ago
parent 73c90887e8
commit fc48ed7e09

@ -0,0 +1,8 @@
version: '3'
services:
dev:
build: .
volumes:
- .:/usr/src/app
tty: true
entrypoint: /bin/bash

File diff suppressed because one or more lines are too long

File diff suppressed because one or more lines are too long

File diff suppressed because one or more lines are too long

@ -1,6 +1,6 @@
nxsim nxsim
simpy simpy
networkx networkx>=2.0
numpy numpy
matplotlib matplotlib
pyyaml pyyaml

@ -3,7 +3,7 @@ from setuptools import setup
with open(os.path.join('soil', 'VERSION')) as f: with open(os.path.join('soil', 'VERSION')) as f:
__version__ = f.read().strip() __version__ = f.readlines()[0].strip()
assert __version__ assert __version__

@ -1 +1 @@
0.10.2 0.11

@ -62,7 +62,7 @@ def main():
simulation.run_from_config(args.file, simulation.run_from_config(args.file,
dry_run=args.dry_run, dry_run=args.dry_run,
dump=dump, dump=dump,
parallel=(not args.synchronous), parallel=(not args.synchronous and not args.pdb),
results_dir=args.output) results_dir=args.output)
except Exception as ex: except Exception as ex:
if args.pdb: if args.pdb:

@ -11,9 +11,9 @@ class CounterModel(BaseAgent):
# Outside effects # Outside effects
total = len(list(self.get_all_agents())) total = len(list(self.get_all_agents()))
neighbors = len(list(self.get_neighboring_agents())) neighbors = len(list(self.get_neighboring_agents()))
self.state['times'] = self.state.get('times', 0) + 1 self['times'] = self.get('times', 0) + 1
self.state['neighbors'] = neighbors self['neighbors'] = neighbors
self.state['total'] = total self['total'] = total
class AggregatedCounter(BaseAgent): class AggregatedCounter(BaseAgent):
@ -26,7 +26,7 @@ class AggregatedCounter(BaseAgent):
# Outside effects # Outside effects
total = len(list(self.get_all_agents())) total = len(list(self.get_all_agents()))
neighbors = len(list(self.get_neighboring_agents())) neighbors = len(list(self.get_neighboring_agents()))
self.state['times'] = self.state.get('times', 0) + 1 self['times'] = self.get('times', 0) + 1
self.state['neighbors'] = self.state.get('neighbors', 0) + neighbors self['neighbors'] = self.get('neighbors', 0) + neighbors
self.state['total'] = total = self.state.get('total', 0) + total self['total'] = total = self.get('total', 0) + total
self.debug('Running for step: {}. Total: {}'.format(self.now, total)) self.debug('Running for step: {}. Total: {}'.format(self.now, total))

@ -14,7 +14,7 @@ import json
from functools import wraps from functools import wraps
from .. import utils from .. import utils, history
agent_types = {} agent_types = {}
@ -32,33 +32,67 @@ class BaseAgent(nxsim.BaseAgent, metaclass=MetaAgent):
defaults = {} defaults = {}
def __init__(self, **kwargs): def __init__(self, environment=None, agent_id=None, state=None,
name='network_process', interval=None, **state_params):
# Check for REQUIRED arguments
assert environment is not None, TypeError('__init__ missing 1 required keyword argument: \'environment\'. '
'Cannot be NoneType.')
# Initialize agent parameters
self.id = agent_id
self.name = name
self.state_params = state_params
# Global parameters
self.global_topology = environment.G
self.environment_params = environment.environment_params
# Register agent to environment
self.env = environment
self._neighbors = None self._neighbors = None
self.alive = True self.alive = True
state = deepcopy(self.defaults) real_state = deepcopy(self.defaults)
state.update(kwargs.pop('state', {})) real_state.update(state or {})
kwargs['state'] = state self._state = real_state
super().__init__(**kwargs) self.interval = interval
if not hasattr(self, 'level'): if not hasattr(self, 'level'):
self.level = logging.DEBUG self.level = logging.DEBUG
self.logger = logging.getLogger('{}-Agent-{}'.format(self.env.name, self.id)) self.logger = logging.getLogger('{}-Agent-{}'.format(self.env.name,
self.id))
self.logger.setLevel(self.level) self.logger.setLevel(self.level)
# initialize every time an instance of the agent is created
self.action = self.env.process(self.run())
@property
def state(self):
return self._state
@state.setter
def state(self, value):
for k, v in value.items():
self[k] = v
def __getitem__(self, key): def __getitem__(self, key):
if isinstance(key, tuple): if isinstance(key, tuple):
k, t_step = key key, t_step = key
return self.env[self.id, t_step, k] k = history.Key(key=key, t_step=t_step, agent_id=self.id)
return self.env[k]
return self.state.get(key, None) return self.state.get(key, None)
def __delitem__(self, key): def __delitem__(self, key):
del self.state[key] self.state[key] = None
def __contains__(self, key): def __contains__(self, key):
return key in self.state return key in self.state
def __setitem__(self, key, value): def __setitem__(self, key, value):
self.state[key] = value self.state[key] = value
k = history.Key(t_step=self.now,
agent_id=self.id,
key=key)
self.env[k] = value
def get(self, key, default=None): def get(self, key, default=None):
return self[key] if key in self else default return self[key] if key in self else default
@ -72,7 +106,12 @@ class BaseAgent(nxsim.BaseAgent, metaclass=MetaAgent):
return None return None
def run(self): def run(self):
interval = self.env.interval if self.interval is not None:
interval = self.interval
elif 'interval' in self:
interval = self['interval']
else:
interval = self.env.interval
while self.alive: while self.alive:
res = self.step() res = self.step()
yield res or self.env.timeout(interval) yield res or self.env.timeout(interval)
@ -95,7 +134,7 @@ class BaseAgent(nxsim.BaseAgent, metaclass=MetaAgent):
agents = self.global_topology.nodes() agents = self.global_topology.nodes()
count = 0 count = 0
for agent in agents: for agent in agents:
if state_id and state_id != self.global_topology.node[agent]['agent'].state['id']: if state_id and state_id != self.global_topology.node[agent]['agent']['id']:
continue continue
count += 1 count += 1
return count return count
@ -197,11 +236,13 @@ class FSM(BaseAgent, metaclass=MetaFSM):
def __init__(self, *args, **kwargs): def __init__(self, *args, **kwargs):
super(FSM, self).__init__(*args, **kwargs) super(FSM, self).__init__(*args, **kwargs)
if 'id' not in self.state: if 'id' not in self.state:
self.state['id'] = self.default_state.id if not self.default_state:
raise ValueError('No default state specified for {}'.format(self.id))
self['id'] = self.default_state.id
def step(self): def step(self):
if 'id' in self.state: if 'id' in self.state:
next_state = self.state['id'] next_state = self['id']
elif self.default_state: elif self.default_state:
next_state = self.default_state.id next_state = self.default_state.id
else: else:
@ -215,7 +256,7 @@ class FSM(BaseAgent, metaclass=MetaFSM):
state = state.id state = state.id
if state not in self.states: if state not in self.states:
raise ValueError('{} is not a valid state'.format(state)) raise ValueError('{} is not a valid state'.format(state))
self.state['id'] = state self['id'] = state
return state return state

@ -4,7 +4,7 @@ import glob
import yaml import yaml
from os.path import join from os.path import join
from . import utils from . import utils, history
def read_data(*args, group=False, **kwargs): def read_data(*args, group=False, **kwargs):
@ -15,8 +15,9 @@ def read_data(*args, group=False, **kwargs):
return list(iterable) return list(iterable)
def _read_data(pattern, keys=None, convert_types=False, def _read_data(pattern, *args, from_csv=False, process_args=None, **kwargs):
process=None, from_csv=False, **kwargs): if not process_args:
process_args = {}
for folder in glob.glob(pattern): for folder in glob.glob(pattern):
config_file = glob.glob(join(folder, '*.yml'))[0] config_file = glob.glob(join(folder, '*.yml'))[0]
config = yaml.load(open(config_file)) config = yaml.load(open(config_file))
@ -24,19 +25,20 @@ def _read_data(pattern, keys=None, convert_types=False,
if from_csv: if from_csv:
for trial_data in sorted(glob.glob(join(folder, for trial_data in sorted(glob.glob(join(folder,
'*.environment.csv'))): '*.environment.csv'))):
df = read_csv(trial_data, convert_types=convert_types) df = read_csv(trial_data, **kwargs)
if process:
df = process(df, **kwargs)
yield config_file, df, config yield config_file, df, config
else: else:
for trial_data in sorted(glob.glob(join(folder, '*.db.sqlite'))): for trial_data in sorted(glob.glob(join(folder, '*.db.sqlite'))):
df = read_sql(trial_data, convert_types=convert_types, df = read_sql(trial_data, **kwargs)
keys=keys)
if process:
df = process(df, **kwargs)
yield config_file, df, config yield config_file, df, config
def read_sql(db, *args, **kwargs):
h = history.History(db, backup=False)
df = h.read_sql(*args, **kwargs)
return df
def read_csv(filename, keys=None, convert_types=False, **kwargs): def read_csv(filename, keys=None, convert_types=False, **kwargs):
''' '''
Read a CSV in canonical form: :: Read a CSV in canonical form: ::
@ -49,18 +51,7 @@ def read_csv(filename, keys=None, convert_types=False, **kwargs):
df = convert_types_slow(df) df = convert_types_slow(df)
if keys: if keys:
df = df[df['key'].isin(keys)] df = df[df['key'].isin(keys)]
return df df = process_one(df)
def read_sql(filename, keys=None, convert_types=False, limit=-1):
condition = ''
if keys:
k = map(lambda x: "\'{}\'".format(x), keys)
condition = 'where key in ({})'.format(','.join(k))
query = 'select * from history {} limit {}'.format(condition, limit)
df = pd.read_sql_query(query, 'sqlite:///{}'.format(filename))
if convert_types:
df = convert_types_slow(df)
return df return df
@ -108,8 +99,9 @@ def get_types(df):
return {k:v[0] for k,v in dtypes.iteritems()} return {k:v[0] for k,v in dtypes.iteritems()}
def process_one(df, *keys, columns=['key'], values='value', def process_one(df, *keys, columns=['key', 'agent_id'], values='value',
index=['t_step', 'agent_id'], aggfunc='first', **kwargs): fill=True, index=['t_step',],
aggfunc='first', **kwargs):
''' '''
Process a dataframe in canonical form ``(t_step, agent_id, key, value, value_type)`` into Process a dataframe in canonical form ``(t_step, agent_id, key, value, value_type)`` into
a dataframe with a column per key a dataframe with a column per key
@ -119,35 +111,29 @@ def process_one(df, *keys, columns=['key'], values='value',
if keys: if keys:
df = df[df['key'].isin(keys)] df = df[df['key'].isin(keys)]
dtypes = get_types(df)
df = df.pivot_table(values=values, index=index, columns=columns, df = df.pivot_table(values=values, index=index, columns=columns,
aggfunc=aggfunc, **kwargs) aggfunc=aggfunc, **kwargs)
df = df.fillna(0).astype(dtypes) if fill:
df = fillna(df)
return df return df
def get_count_processed(df, *keys):
if keys:
df = df[list(keys)]
# p = df.groupby(level=0).apply(pd.Series.value_counts)
p = df.unstack().apply(pd.Series.value_counts, axis=1)
return p
def get_count(df, *keys): def get_count(df, *keys):
if keys: if keys:
df = df[df['key'].isin(keys)] df = df[list(keys)]
p = df.groupby(by=['t_step', 'key', 'value']).size().unstack(level=[1,2]).fillna(0) counts = pd.DataFrame()
return p for key in df.columns.levels[0]:
g = df[key].apply(pd.Series.value_counts, axis=1).fillna(0)
for value, series in g.iteritems():
counts[key, value] = series
counts.columns = pd.MultiIndex.from_tuples(counts.columns)
return counts
def get_value(df, *keys, aggfunc='sum'): def get_value(df, *keys, aggfunc='sum'):
if keys: if keys:
df = df[df['key'].isin(keys)] df = df[list(keys)]
p = process_one(df, *keys) return df.groupby(axis=1, level=0).agg(aggfunc, axis=1)
p = p.groupby(level='t_step').agg(aggfunc)
return p
def plot_all(*args, **kwargs): def plot_all(*args, **kwargs):
@ -175,4 +161,6 @@ def group_trials(trials, aggfunc=['mean', 'min', 'max', 'std']):
return pd.concat(trials).groupby(level=0).agg(aggfunc).reorder_levels([2, 0,1] ,axis=1) return pd.concat(trials).groupby(level=0).agg(aggfunc).reorder_levels([2, 0,1] ,axis=1)
def fillna(df):
new_df = df.ffill(axis=0)
return new_df

@ -5,16 +5,26 @@ import csv
import random import random
import simpy import simpy
import tempfile import tempfile
import pandas as pd
from copy import deepcopy from copy import deepcopy
from networkx.readwrite import json_graph from networkx.readwrite import json_graph
import networkx as nx import networkx as nx
import nxsim import nxsim
from . import utils, agents from . import utils, agents, analysis, history
class SoilEnvironment(nxsim.NetworkEnvironment): class SoilEnvironment(nxsim.NetworkEnvironment):
"""
The environment is key in a simulation. It contains the network topology,
a reference to network and environment agents, as well as the environment
params, which are used as shared state between agents.
The environment parameters and the state of every agent can be accessed
both by using the environment as a dictionary or with the environment's
:meth:`soil.environment.SoilEnvironment.get` method.
"""
def __init__(self, name=None, def __init__(self, name=None,
network_agents=None, network_agents=None,
@ -38,19 +48,21 @@ class SoilEnvironment(nxsim.NetworkEnvironment):
self._env_agents = {} self._env_agents = {}
self.dry_run = dry_run self.dry_run = dry_run
self.interval = interval self.interval = interval
self.dir_path = dir_path or tempfile.mkdtemp('soil-env')
self.get_path()
self._history = history.History(name=self.name if not dry_run else None,
dir_path=self.dir_path)
# Add environment agents first, so their events get # Add environment agents first, so their events get
# executed before network agents # executed before network agents
self['SEED'] = seed or time.time()
random.seed(self['SEED'])
self.process(self.save_state())
self.environment_agents = environment_agents or [] self.environment_agents = environment_agents or []
self.network_agents = network_agents or [] self.network_agents = network_agents or []
self.dir_path = dir_path or tempfile.mkdtemp('soil-env')
if self.dry_run: if self.dry_run:
self._db_path = ":memory:" self._db_path = ":memory:"
else: else:
self._db_path = os.path.join(self.get_path(), '{}.db.sqlite'.format(self.name)) self._db_path = os.path.join(self.get_path(), '{}.db.sqlite'.format(self.name))
self.create_db(self._db_path) self.create_db(self._db_path)
self['SEED'] = seed or time.time()
random.seed(self['SEED'])
def create_db(self, db_path=None): def create_db(self, db_path=None):
db_path = db_path or self._db_path db_path = db_path or self._db_path
@ -95,10 +107,8 @@ class SoilEnvironment(nxsim.NetworkEnvironment):
if not network_agents: if not network_agents:
return return
for ix in self.G.nodes(): for ix in self.G.nodes():
i = ix
node = self.G.node[i]
agent, state = agents._agent_from_distribution(network_agents) agent, state = agents._agent_from_distribution(network_agents)
self.set_agent(i, agent_type=agent, state=state) self.set_agent(ix, agent_type=agent, state=state)
def set_agent(self, agent_id, agent_type, state=None): def set_agent(self, agent_id, agent_type, state=None):
node = self.G.nodes[agent_id] node = self.G.nodes[agent_id]
@ -125,16 +135,21 @@ class SoilEnvironment(nxsim.NetworkEnvironment):
return self.G.add_edge(agent1, agent2) return self.G.add_edge(agent1, agent2)
def run(self, *args, **kwargs): def run(self, *args, **kwargs):
self._save_state()
super().run(*args, **kwargs) super().run(*args, **kwargs)
self._history.flush_cache()
def _save_state(self, now=None): def _save_state(self, now=None):
# for agent in self.agents: # for agent in self.agents:
# agent.save_state() # agent.save_state()
utils.logger.debug('Saving state @{}'.format(self.now)) utils.logger.debug('Saving state @{}'.format(self.now))
with self._db: self._history.save_records(self.state_to_tuples(now=now))
self._db.executemany("insert into history(agent_id, t_step, key, value, value_type) values (?, ?, ?, ?, ?)", self.state_to_tuples(now=now))
def save_state(self): def save_state(self):
'''
:DEPRECATED:
Periodically save the state of the environment and the agents.
'''
self._save_state() self._save_state()
while self.peek() != simpy.core.Infinity: while self.peek() != simpy.core.Infinity:
delay = max(self.peek() - self.now, self.interval) delay = max(self.peek() - self.now, self.interval)
@ -149,64 +164,44 @@ class SoilEnvironment(nxsim.NetworkEnvironment):
def __getitem__(self, key): def __getitem__(self, key):
if isinstance(key, tuple): if isinstance(key, tuple):
values = [("agent_id", key[0]), self._history.flush_cache()
("t_step", key[1]), return self._history[key]
("key", key[2]),
("value", None),
("value_type", None)]
fields = list(k for k, v in values if v is None)
conditions = " and ".join("{}='{}'".format(k, v) for k, v in values if v is not None)
query = """SELECT {fields} from history""".format(fields=",".join(fields))
if conditions:
query = """{query} where {conditions}""".format(query=query,
conditions=conditions)
with self._db:
rows = self._db.execute(query).fetchall()
utils.logger.debug(rows)
results = self.rows_to_dict(rows)
return results
return self.environment_params[key] return self.environment_params[key]
def rows_to_dict(self, rows):
if len(rows) < 1:
return None
level = len(rows[0])-2
if level == 0:
if len(rows) != 1:
raise ValueError('Cannot convert {} to dictionaries'.format(rows))
value, value_type = rows[0]
return utils.convert(value, value_type)
results = {}
for row in rows:
item = results
for i in range(level-1):
key = row[i]
if key not in item:
item[key] = {}
item = item[key]
key, value, value_type = row[level-1:]
item[key] = utils.convert(value, value_type)
return results
def __setitem__(self, key, value): def __setitem__(self, key, value):
if isinstance(key, tuple):
k = history.Key(*key)
self._history.save_record(*k,
value=value)
return
self.environment_params[key] = value self.environment_params[key] = value
self._history.save_record(agent_id='env',
t_step=self.now,
key=key,
value=value)
def __contains__(self, key): def __contains__(self, key):
return key in self.environment_params return key in self.environment_params
def get(self, key, default=None): def get(self, key, default=None):
'''
Get the value of an environment attribute in a
given point in the simulation (history).
If key is an attribute name, this method returns
the current value.
To get values at other times, use a
:meth: `soil.history.Key` tuple.
'''
return self[key] if key in self else default return self[key] if key in self else default
def get_path(self, dir_path=None): def get_path(self, dir_path=None):
dir_path = dir_path or self.dir_path dir_path = dir_path or self.dir_path
if not os.path.exists(dir_path): if not os.path.exists(dir_path):
os.makedirs(dir_path) try:
os.makedirs(dir_path)
except FileExistsError:
pass
return dir_path return dir_path
def get_agent(self, agent_id): def get_agent(self, agent_id):
@ -255,17 +250,19 @@ class SoilEnvironment(nxsim.NetworkEnvironment):
if now is None: if now is None:
now = self.now now = self.now
for k, v in self.environment_params.items(): for k, v in self.environment_params.items():
v, v_t = utils.repr(v) yield history.Record(agent_id='env',
yield 'env', now, k, v, v_t t_step=now,
key=k,
value=v)
for agent in self.agents: for agent in self.agents:
for k, v in agent.state.items(): for k, v in agent.state.items():
v, v_t = utils.repr(v) yield history.Record(agent_id=agent.id,
yield agent.id, now, k, v, v_t t_step=now,
key=k,
value=v)
def history_to_tuples(self): def history_to_tuples(self):
with self._db: return self._history.to_tuples()
res = self._db.execute("select agent_id, t_step, key, value, value_type from history ").fetchall()
yield from res
def history_to_graph(self): def history_to_graph(self):
G = nx.Graph(self.G) G = nx.Graph(self.G)
@ -317,14 +314,10 @@ class SoilEnvironment(nxsim.NetworkEnvironment):
def __getstate__(self): def __getstate__(self):
state = self.__dict__.copy() state = self.__dict__.copy()
state['G'] = json_graph.node_link_data(self.G) state['G'] = json_graph.node_link_data(self.G)
state['network_agents'] = agents.serialize_distribution(self.network_agents) state['network_agents'] = agents._serialize_distribution(self.network_agents)
state['environment_agents'] = agents._convert_agent_types(self.environment_agents, state['environment_agents'] = agents._convert_agent_types(self.environment_agents,
to_string=True) to_string=True)
del state['_queue'] del state['_queue']
import inspect
for k, v in state.items():
if inspect.isgeneratorfunction(v):
print(k, v, type(v))
return state return state
def __setstate__(self, state): def __setstate__(self, state):

@ -0,0 +1,231 @@
import time
import os
import pandas as pd
import sqlite3
import copy
from collections import UserDict, Iterable, namedtuple
from . import utils
class History:
"""
Store and retrieve values from a sqlite database.
"""
def __init__(self, db_path=None, name=None, dir_path=None, backup=True):
if db_path is None and name:
db_path = os.path.join(dir_path or os.getcwd(), '{}.db.sqlite'.format(name))
if db_path is None:
db_path = ":memory:"
else:
if backup and os.path.exists(db_path):
newname = db_path.replace('db.sqlite', 'backup{}.sqlite'.format(time.time()))
os.rename(db_path, newname)
self._db_path = db_path
if isinstance(db_path, str):
self._db = sqlite3.connect(db_path)
else:
self._db = db_path
with self._db:
self._db.execute('''CREATE TABLE IF NOT EXISTS history (agent_id text, t_step int, key text, value text text)''')
self._db.execute('''CREATE TABLE IF NOT EXISTS value_types (key text, value_type text)''')
self._db.execute('''CREATE UNIQUE INDEX IF NOT EXISTS idx_history ON history (agent_id, t_step, key);''')
self._dtypes = {}
self._tups = []
def conversors(self, key):
"""Get the serializer and deserializer for a given key."""
if key not in self._dtypes:
self.read_types()
return self._dtypes[key]
@property
def dtypes(self):
return {k:v[0] for k, v in self._dtypes.items()}
def save_tuples(self, tuples):
self.save_records(Record(*tup) for tup in tuples)
def save_records(self, records):
with self._db:
for rec in records:
if not isinstance(rec, Record):
rec = Record(*rec)
if rec.key not in self._dtypes:
name = utils.name(rec.value)
serializer = utils.serializer(name)
deserializer = utils.deserializer(name)
self._dtypes[rec.key] = (name, serializer, deserializer)
self._db.execute("replace into value_types (key, value_type) values (?, ?)", (rec.key, name))
self._db.execute("replace into history(agent_id, t_step, key, value) values (?, ?, ?, ?)", (rec.agent_id, rec.t_step, rec.key, rec.value))
def save_record(self, *args, **kwargs):
self._tups.append(Record(*args, **kwargs))
if len(self._tups) > 100:
self.flush_cache()
def flush_cache(self):
'''
Use a cache to save state changes to avoid opening a session for every change.
The cache will be flushed at the end of the simulation, and when history is accessed.
'''
self.save_records(self._tups)
self._tups = list()
def to_tuples(self):
self.flush_cache()
with self._db:
res = self._db.execute("select agent_id, t_step, key, value from history ").fetchall()
for r in res:
agent_id, t_step, key, value = r
_, _ , des = self.conversors(key)
yield agent_id, t_step, key, des(value)
def read_types(self):
with self._db:
res = self._db.execute("select key, value_type from value_types ").fetchall()
for k, v in res:
serializer = utils.serializer(v)
deserializer = utils.deserializer(v)
self._dtypes[k] = (v, serializer, deserializer)
def __getitem__(self, key):
key = Key(*key)
agent_ids = [key.agent_id] if key.agent_id is not None else []
t_steps = [key.t_step] if key.t_step is not None else []
keys = [key.key] if key.key is not None else []
df = self.read_sql(agent_ids=agent_ids,
t_steps=t_steps,
keys=keys)
r = Records(df, filter=key, dtypes=self._dtypes)
return r.value()
def read_sql(self, keys=None, agent_ids=None, t_steps=None, convert_types=False, limit=-1):
self.read_types()
def escape_and_join(v):
if v is None:
return
return ",".join(map(lambda x: "\'{}\'".format(x), v))
filters = [("key in ({})".format(escape_and_join(keys)), keys),
("agent_id in ({})".format(escape_and_join(agent_ids)), agent_ids)
]
filters = list(k[0] for k in filters if k[1])
last_df = None
if t_steps:
# Look for the last value before the minimum step in the query
min_step = min(t_steps)
last_filters = ['t_step < {}'.format(min_step),]
last_filters = last_filters + filters
condition = ' and '.join(last_filters)
last_query = '''
select h1.*
from history h1
inner join (
select agent_id, key, max(t_step) as t_step
from history
where {condition}
group by agent_id, key
) h2
on h1.agent_id = h2.agent_id and
h1.key = h2.key and
h1.t_step = h2.t_step
'''.format(condition=condition)
last_df = pd.read_sql_query(last_query, self._db)
filters.append("t_step >= '{}' and t_step <= '{}'".format(min_step, max(t_steps)))
condition = ''
if filters:
condition = 'where {} '.format(' and '.join(filters))
query = 'select * from history {} limit {}'.format(condition, limit)
df = pd.read_sql_query(query, self._db)
if last_df is not None:
df = pd.concat([df, last_df])
df_p = df.pivot_table(values='value', index=['t_step'],
columns=['key', 'agent_id'],
aggfunc='first')
for k, v in self._dtypes.items():
if k in df_p:
dtype, _, deserial = v
df_p[k] = df_p[k].fillna(method='ffill').fillna(deserial()).astype(dtype)
if t_steps:
df_p = df_p.reindex(t_steps, method='ffill')
return df_p.ffill()
class Records():
def __init__(self, df, filter=None, dtypes=None):
if not filter:
filter = Key(agent_id=None,
t_step=None,
key=None)
self._df = df
self._filter = filter
self.dtypes = dtypes or {}
super().__init__()
def mask(self, tup):
res = ()
for i, k in zip(tup[:-1], self._filter):
if k is None:
res = res + (i,)
res = res + (tup[-1],)
return res
def filter(self, newKey):
f = list(self._filter)
for ix, i in enumerate(f):
if i is None:
f[ix] = newKey
self._filter = Key(*f)
@property
def resolved(self):
return sum(1 for i in self._filter if i is not None) == 3
def __iter__(self):
for column, series in self._df.iteritems():
key, agent_id = column
for t_step, value in series.iteritems():
r = Record(t_step=t_step,
agent_id=agent_id,
key=key,
value=value)
yield self.mask(r)
def value(self):
if self.resolved:
f = self._filter
try:
i = self._df[f.key][str(f.agent_id)]
ix = i.index.get_loc(f.t_step, method='ffill')
return i.iloc[ix]
except KeyError:
return self.dtypes[f.key][2]()
return self
def __getitem__(self, k):
n = copy.copy(self)
n.filter(k)
return n.value()
def __len__(self):
return len(self._df)
Key = namedtuple('Key', ['agent_id', 't_step', 'key'])
Record = namedtuple('Record', 'agent_id t_step key value')

@ -20,7 +20,7 @@ class SoilSimulation(NetworkSimulation):
""" """
Subclass of nsim.NetworkSimulation with three main differences: Subclass of nsim.NetworkSimulation with three main differences:
1) agent type can be specified by name or by class. 1) agent type can be specified by name or by class.
2) instead of just one type, an network_agents can be used. 2) instead of just one type, a network agents distribution can be used.
The distribution specifies the weight (or probability) of each The distribution specifies the weight (or probability) of each
agent type in the topology. This is an example distribution: :: agent type in the topology. This is an example distribution: ::
@ -95,16 +95,16 @@ class SoilSimulation(NetworkSimulation):
def run_simulation_gen(self, *args, parallel=False, dry_run=False, def run_simulation_gen(self, *args, parallel=False, dry_run=False,
**kwargs): **kwargs):
p = Pool() p = Pool()
with utils.timer('simulation'): with utils.timer('simulation {}'.format(self.name)):
if parallel: if parallel:
func = partial(self.run_trial, dry_run=dry_run, func = partial(self.run_trial, dry_run=dry_run or self.dry_run,
return_env=not parallel, **kwargs) return_env=not parallel, **kwargs)
for i in p.imap_unordered(func, range(self.num_trials)): for i in p.imap_unordered(func, range(self.num_trials)):
yield i yield i
else: else:
for i in range(self.num_trials): for i in range(self.num_trials):
yield self.run_trial(i, dry_run=dry_run, **kwargs) yield self.run_trial(i, dry_run=dry_run or self.dry_run, **kwargs)
if not dry_run or self.dry_run: if not (dry_run or self.dry_run):
logger.info('Dumping results to {}'.format(self.dir_path)) logger.info('Dumping results to {}'.format(self.dir_path))
self.dump_pickle(self.dir_path) self.dump_pickle(self.dir_path)
self.dump_yaml(self.dir_path) self.dump_yaml(self.dir_path)
@ -192,7 +192,7 @@ class SoilSimulation(NetworkSimulation):
return state return state
def from_config(config, G=None): def from_config(config):
config = list(utils.load_config(config)) config = list(utils.load_config(config))
if len(config) > 1: if len(config) > 1:
raise AttributeError('Provide only one configuration') raise AttributeError('Provide only one configuration')
@ -201,9 +201,10 @@ def from_config(config, G=None):
return sim return sim
def run_from_config(*configs, results_dir='soil_output', dump=None, timestamp=False, **kwargs): def run_from_config(*configs, results_dir='soil_output', dry_run=False, dump=None, timestamp=False, **kwargs):
for config_def in configs: for config_def in configs:
for config, cpath in utils.load_config(config_def): # logger.info("Found {} config(s)".format(len(ls)))
for config, _ in utils.load_config(config_def):
name = config.get('name', 'unnamed') name = config.get('name', 'unnamed')
logger.info("Using config(s): {name}".format(name=name)) logger.info("Using config(s): {name}".format(name=name))
@ -215,4 +216,4 @@ def run_from_config(*configs, results_dir='soil_output', dump=None, timestamp=Fa
dir_path = os.path.join(results_dir, sim_folder) dir_path = os.path.join(results_dir, sim_folder)
sim = SoilSimulation(dir_path=dir_path, dump=dump, **config) sim = SoilSimulation(dir_path=dir_path, dump=dump, **config)
logger.info('Dumping results to {} : {}'.format(sim.dir_path, sim.dump)) logger.info('Dumping results to {} : {}'.format(sim.dir_path, sim.dump))
results = sim.run_simulation(**kwargs) sim.run_simulation(**kwargs)

@ -1,6 +1,7 @@
import os import os
import yaml import yaml
import logging import logging
import importlib
from time import time from time import time
from glob import glob from glob import glob
from random import random from random import random
@ -72,13 +73,22 @@ def timer(name='task', pre="", function=logger.info, to_object=None):
def repr(v): def repr(v):
if isinstance(v, bool): func = serializer(v)
v = "true" if v else "" tname = name(v)
return v, bool.__name__ return func(v), tname
return v, type(v).__name__
def convert(value, type_):
import importlib def name(v):
return type(v).__name__
def serializer(type_):
if type_ == 'bool':
return lambda x: "true" if x else ""
return lambda x: x
def deserializer(type_):
try: try:
# Check if it's a builtin type # Check if it's a builtin type
module = importlib.import_module('builtins') module = importlib.import_module('builtins')
@ -88,4 +98,8 @@ def convert(value, type_):
module, type_ = type_.rsplit(".", 1) module, type_ = type_.rsplit(".", 1)
module = importlib.import_module(module) module = importlib.import_module(module)
cls = getattr(module, type_) cls = getattr(module, type_)
return cls(value) return cls
def convert(value, type_):
return deserializer(type_)(value)

@ -0,0 +1,16 @@
agent_id,t_step,key,value,value_type
a0,0,hello,w,str
a0,1,hello,o,str
a0,2,hello,r,str
a0,3,hello,l,str
a0,4,hello,d,str
a0,5,hello,!,str
env,1,started,,bool
env,2,started,True,bool
env,7,started,,bool
a0,0,hello,w,str
a0,1,hello,o,str
a0,2,hello,r,str
a0,3,hello,l,str
a0,4,hello,d,str
a0,5,hello,!,str
1 agent_id t_step key value value_type
2 a0 0 hello w str
3 a0 1 hello o str
4 a0 2 hello r str
5 a0 3 hello l str
6 a0 4 hello d str
7 a0 5 hello ! str
8 env 1 started bool
9 env 2 started True bool
10 env 7 started bool
11 a0 0 hello w str
12 a0 1 hello o str
13 a0 2 hello r str
14 a0 3 hello l str
15 a0 4 hello d str
16 a0 5 hello ! str

@ -0,0 +1,90 @@
from unittest import TestCase
import os
import pandas as pd
import yaml
from functools import partial
from os.path import join
from soil import simulation, analysis, agents
ROOT = os.path.abspath(os.path.dirname(__file__))
class Ping(agents.FSM):
defaults = {
'count': 0,
}
@agents.default_state
@agents.state
def even(self):
self['count'] += 1
return self.odd
@agents.state
def odd(self):
self['count'] += 1
return self.even
class TestAnalysis(TestCase):
# Code to generate a simple sqlite history
def setUp(self):
"""
The initial states should be applied to the agent and the
agent should be able to update its state."""
config = {
'name': 'analysis',
'dry_run': True,
'seed': 'seed',
'network_params': {
'generator': 'complete_graph',
'n': 2
},
'agent_type': Ping,
'states': [{'interval': 1}, {'interval': 2}],
'max_time': 30,
'num_trials': 1,
'environment_params': {
}
}
s = simulation.from_config(config)
self.env = s.run_simulation()[0]
def test_saved(self):
env = self.env
assert env.get_agent(0)['count', 0] == 1
assert env.get_agent(0)['count', 29] == 30
assert env.get_agent(1)['count', 0] == 1
assert env.get_agent(1)['count', 29] == 15
assert env['env', 29, None]['SEED'] == env['env', 29, 'SEED']
def test_count(self):
env = self.env
df = analysis.read_sql(env._history._db)
res = analysis.get_count(df, 'SEED', 'id')
assert res['SEED']['seedanalysis_trial_0'].iloc[0] == 1
assert res['SEED']['seedanalysis_trial_0'].iloc[-1] == 1
assert res['id']['odd'].iloc[0] == 2
assert res['id']['even'].iloc[0] == 0
assert res['id']['odd'].iloc[-1] == 1
assert res['id']['even'].iloc[-1] == 1
def test_value(self):
env = self.env
df = analysis.read_sql(env._history._db)
res_sum = analysis.get_value(df, 'count')
assert res_sum['count'].iloc[0] == 2
import numpy as np
res_mean = analysis.get_value(df, 'count', aggfunc=np.mean)
assert res_mean['count'].iloc[0] == 1
res_total = analysis.get_value(df)
res_total['SEED'].iloc[0] == 'seedanalysis_trial_0'

@ -0,0 +1,90 @@
from unittest import TestCase
import os
import pandas as pd
from soil import history, analysis
ROOT = os.path.abspath(os.path.dirname(__file__))
class TestHistory(TestCase):
def test_history(self):
"""
"""
tuples = (
('a_0', 0, 'id', 'h', ),
('a_0', 1, 'id', 'e', ),
('a_0', 2, 'id', 'l', ),
('a_0', 3, 'id', 'l', ),
('a_0', 4, 'id', 'o', ),
('a_1', 0, 'id', 'v', ),
('a_1', 1, 'id', 'a', ),
('a_1', 2, 'id', 'l', ),
('a_1', 3, 'id', 'u', ),
('a_1', 4, 'id', 'e', ),
('env', 1, 'prob', 1),
('env', 3, 'prob', 2),
('env', 5, 'prob', 3),
('a_2', 7, 'finished', True),
)
h = history.History()
h.save_tuples(tuples)
# assert h['env', 0, 'prob'] == 0
for i in range(1, 7):
assert h['env', i, 'prob'] == ((i-1)//2)+1
for i, k in zip(range(5), 'hello'):
assert h['a_0', i, 'id'] == k
for record, value in zip(h['a_0', None, 'id'], 'hello'):
t_step, val = record
assert val == value
for i, k in zip(range(5), 'value'):
assert h['a_1', i, 'id'] == k
for i in range(5, 8):
assert h['a_1', i, 'id'] == 'e'
for i in range(7):
assert h['a_2', i, 'finished'] == False
assert h['a_2', 7, 'finished']
def test_history_gen(self):
"""
"""
tuples = (
('a_1', 0, 'id', 'v', ),
('a_1', 1, 'id', 'a', ),
('a_1', 2, 'id', 'l', ),
('a_1', 3, 'id', 'u', ),
('a_1', 4, 'id', 'e', ),
('env', 1, 'prob', 1),
('env', 2, 'prob', 2),
('env', 3, 'prob', 3),
('a_2', 7, 'finished', True),
)
h = history.History()
h.save_tuples(tuples)
for t_step, key, value in h['env', None, None]:
assert t_step == value
assert key == 'prob'
records = list(h[None, 7, None])
assert len(records) == 3
for i in records:
agent_id, key, value = i
if agent_id == 'a_1':
assert key == 'id'
assert value == 'e'
elif agent_id == 'a_2':
assert key == 'finished'
assert value == True
else:
assert key == 'prob'
assert value == 3
records = h['a_1', 7, None]
assert records['id'] == 'e'

@ -22,6 +22,7 @@ class TestMain(TestCase):
Raise an exception otherwise. Raise an exception otherwise.
""" """
config = { config = {
'dry_run': True,
'network_params': { 'network_params': {
'path': join(ROOT, 'test.gexf') 'path': join(ROOT, 'test.gexf')
} }
@ -31,6 +32,7 @@ class TestMain(TestCase):
assert len(G) == 2 assert len(G) == 2
with self.assertRaises(AttributeError): with self.assertRaises(AttributeError):
config = { config = {
'dry_run': True,
'network_params': { 'network_params': {
'path': join(ROOT, 'unknown.extension') 'path': join(ROOT, 'unknown.extension')
} }
@ -44,6 +46,7 @@ class TestMain(TestCase):
should be used to generate a network should be used to generate a network
""" """
config = { config = {
'dry_run': True,
'network_params': { 'network_params': {
'generator': 'barabasi_albert_graph' 'generator': 'barabasi_albert_graph'
} }
@ -58,6 +61,7 @@ class TestMain(TestCase):
def test_empty_simulation(self): def test_empty_simulation(self):
"""A simulation with a base behaviour should do nothing""" """A simulation with a base behaviour should do nothing"""
config = { config = {
'dry_run': True,
'network_params': { 'network_params': {
'path': join(ROOT, 'test.gexf') 'path': join(ROOT, 'test.gexf')
}, },
@ -74,11 +78,12 @@ class TestMain(TestCase):
agent should be able to update its state.""" agent should be able to update its state."""
config = { config = {
'name': 'CounterAgent', 'name': 'CounterAgent',
'dry_run': True,
'network_params': { 'network_params': {
'path': join(ROOT, 'test.gexf') 'path': join(ROOT, 'test.gexf')
}, },
'agent_type': 'CounterModel', 'agent_type': 'CounterModel',
'states': [{'neighbors': 10}, {'total': 12}], 'states': [{'times': 10}, {'times': 20}],
'max_time': 2, 'max_time': 2,
'num_trials': 1, 'num_trials': 1,
'environment_params': { 'environment_params': {
@ -86,10 +91,10 @@ class TestMain(TestCase):
} }
s = simulation.from_config(config) s = simulation.from_config(config)
env = s.run_simulation(dry_run=True)[0] env = s.run_simulation(dry_run=True)[0]
assert env.get_agent(0)['neighbors', 0] == 10 assert env.get_agent(0)['times', 0] == 11
assert env.get_agent(0)['neighbors', 1] == 1 assert env.get_agent(0)['times', 1] == 12
assert env.get_agent(1)['total', 0] == 12 assert env.get_agent(1)['times', 0] == 21
assert env.get_agent(1)['neighbors', 1] == 1 assert env.get_agent(1)['times', 1] == 22
def test_counter_agent_history(self): def test_counter_agent_history(self):
""" """
@ -97,6 +102,7 @@ class TestMain(TestCase):
""" """
config = { config = {
'name': 'CounterAgent', 'name': 'CounterAgent',
'dry_run': True,
'network_params': { 'network_params': {
'path': join(ROOT, 'test.gexf') 'path': join(ROOT, 'test.gexf')
}, },
@ -114,11 +120,10 @@ class TestMain(TestCase):
env = s.run_simulation(dry_run=True)[0] env = s.run_simulation(dry_run=True)[0]
for agent in env.network_agents: for agent in env.network_agents:
last = 0 last = 0
assert len(agent[None, None]) == 11 assert len(agent[None, None]) == 10
for step, total in agent['total', None].items(): for step, total in sorted(agent['total', None]):
if step > 0: assert total == last + 2
assert total == last + 2 last = total
last = total
def test_custom_agent(self): def test_custom_agent(self):
"""Allow for search of neighbors with a certain state_id""" """Allow for search of neighbors with a certain state_id"""
@ -127,6 +132,7 @@ class TestMain(TestCase):
self.state['neighbors'] = self.count_agents(state_id=0, self.state['neighbors'] = self.count_agents(state_id=0,
limit_neighbors=True) limit_neighbors=True)
config = { config = {
'dry_run': True,
'network_params': { 'network_params': {
'path': join(ROOT, 'test.gexf') 'path': join(ROOT, 'test.gexf')
}, },
@ -150,7 +156,8 @@ class TestMain(TestCase):
config['network_params']['path'] = join(EXAMPLES, config['network_params']['path'] = join(EXAMPLES,
config['network_params']['path']) config['network_params']['path'])
s = simulation.from_config(config) s = simulation.from_config(config)
env = s.run_simulation(dry_run=True)[0] s.dry_run = True
env = s.run_simulation()[0]
for a in env.network_agents: for a in env.network_agents:
skill_level = a.state['skill_level'] skill_level = a.state['skill_level']
if a.id == 'Torvalds': if a.id == 'Torvalds':
@ -174,14 +181,15 @@ class TestMain(TestCase):
with utils.timer('loading'): with utils.timer('loading'):
config = utils.load_file(join(EXAMPLES, 'complete.yml'))[0] config = utils.load_file(join(EXAMPLES, 'complete.yml'))[0]
s = simulation.from_config(config) s = simulation.from_config(config)
s.dry_run = True
with utils.timer('serializing'): with utils.timer('serializing'):
serial = s.to_yaml() serial = s.to_yaml()
with utils.timer('recovering'): with utils.timer('recovering'):
recovered = yaml.load(serial) recovered = yaml.load(serial)
with utils.timer('deleting'): with utils.timer('deleting'):
del recovered['topology'] del recovered['topology']
del recovered['dry_run']
del recovered['load_module'] del recovered['load_module']
del recovered['dry_run']
assert config == recovered assert config == recovered
def test_configuration_changes(self): def test_configuration_changes(self):
@ -191,6 +199,7 @@ class TestMain(TestCase):
""" """
config = utils.load_file('examples/complete.yml')[0] config = utils.load_file('examples/complete.yml')[0]
s = simulation.from_config(config) s = simulation.from_config(config)
s.dry_run = True
for i in range(5): for i in range(5):
s.run_simulation(dry_run=True) s.run_simulation(dry_run=True)
nconfig = s.to_dict() nconfig = s.to_dict()
@ -206,17 +215,14 @@ class TestMain(TestCase):
pass pass
def test_row_conversion(self): def test_row_conversion(self):
sim = simulation.SoilSimulation()
env = environment.SoilEnvironment(dry_run=True) env = environment.SoilEnvironment(dry_run=True)
env['test'] = 'test_value' env['test'] = 'test_value'
env._save_state(now=0)
res = list(env.history_to_tuples()) res = list(env.history_to_tuples())
assert len(res) == len(env.environment_params) assert len(res) == len(env.environment_params)
assert ('env', 0, 'test', 'test_value', 'str') in res
env._now = 1
env['test'] = 'second_value' env['test'] = 'second_value'
env._save_state(now=1)
res = list(env.history_to_tuples()) res = list(env.history_to_tuples())
assert env['env', 0, 'test' ] == 'test_value' assert env['env', 0, 'test' ] == 'test_value'

Loading…
Cancel
Save