main_heuristic.py
from datetime import datetime
import pandas as pd
import numpy as np
import copy
import sys
import csv
import warnings
import CPH
warnings.simplefilter(action='ignore', category=FutureWarning)
'''
Code consisting of main run file and two functions:
- run_heuristic: 1. open tree set and make CPH.PhT environment for each tree
2. run cherry picking heuristic (CPH)
3. return results
- run_main: run CPH with four "PickNextCherry" methods:
1. ML
2. TrivialML
3. Rand
4. TrivialRand
RUN in terminal:
python main_heuristic.py <instance num.> <ML model name> <leaf number> <bool (0/1) for exact input> <option>
option:
if exact input = 0:
option = reticulation number
else:
option = forest size
EXAMPLE:
python main_heuristic.py 0 N10_maxL100_random_balanced 20 0 50
'''
def run_heuristic(tree_set=None, tree_set_newick=None, inst_num=0, repeats=1, time_limit=None,
progress=False, reduce_trivial=False, pick_ml=False, pick_ml_triv=False,
pick_random=False, model_name=None, relabel=False, relabel_cher_triv=False, problem_type="",
full_leaf_set=True, ml_thresh=None):
# READ TREE SET
now = datetime.now().time()
if progress:
print(f"Instance {inst_num} {problem_type}: Start at {now}")
if tree_set is None and tree_set_newick is not None:
# Empty set of inputs
inputs = []
# Read each line of the input file with name set by "option_file_argument"
f = open(tree_set_newick, "rt")
reader = csv.reader(f, delimiter='~', quotechar='|')
for row in reader:
inputs.append(str(row[0]))
f.close()
# Make the set of inputs usable for all algorithms: use the CPH class
tree_set = CPH.Input_Set(newick_strings=inputs, instance=inst_num, full_leaf_set=full_leaf_set)
# RUN HEURISTIC CHERRY PICKING SEQUENCE
# Run the heuristic to find a cherry-picking sequence `seq' for the set of input trees.
# Arguments are set as given by the terminal arguments
seq_dist, seq, df_pred = tree_set.CPSBound(repeats=repeats,
progress=progress,
time_limit=time_limit,
reduce_trivial=reduce_trivial,
pick_ml=pick_ml,
pick_ml_triv=pick_ml_triv,
pick_random=pick_random,
relabel=relabel,
relabel_cher_triv=relabel_cher_triv,
model_name=model_name,
ml_thresh=ml_thresh,
problem_type=problem_type)
# Output the computation time for the heuristic
now = datetime.now().time()
if progress:
print(f"Instance {inst_num} {problem_type}: Finish at {now}")
print(f"Instance {inst_num} {problem_type}: Computation time heuristic: {tree_set.CPS_Compute_Time}")
print(f"Instance {inst_num} {problem_type}: Reticulation number = {min(tree_set.RetPerTrial.values())}")
if pick_ml:
return tree_set.RetPerTrial, tree_set.DurationPerTrial, seq, df_pred
else:
return tree_set.RetPerTrial, tree_set.DurationPerTrial, seq
def run_main(i, l, exact, ret=None, forest_size=None,
repeats=1000, time_limit=None, ml_name=None, full_leaf_set=True, ml_thresh=None, progress=False):
if exact:
test_info = f"L{l}_R{ret}_exact_all"
file_info = f"L{l}_R{ret}_exact"
else:
test_info = f"L{l}_T{forest_size}_all"
file_info = f"L{l}_T{forest_size}"
# ML MODEL
model_name = f"LearningCherries/RFModels/rf_cherries_{ml_name}.joblib"
# save results
score = pd.DataFrame(
index=pd.MultiIndex.from_product([[i], ["RetNum", "Time"], np.arange(repeats)]),
columns=["ML", "TrivialML", "Rand", "TrivialRand", "UB"], dtype=float)
df_seq = pd.DataFrame()
env_info_file = f"Data/Test/inst_results/tree_data_{file_info}_{i}.pickle"
# INSTANCE
tree_set_newick = f"Data/Test/TreeSetsNewick/tree_set_newick_{file_info}_{i}_LGT.txt"
# ML HEURISTIC
ret_score, time_score, seq_ml, df_pred = run_heuristic(
tree_set_newick=tree_set_newick,
inst_num=i,
repeats=1,
time_limit=time_limit,
pick_ml=True,
relabel=True,
model_name=model_name,
problem_type="ML",
full_leaf_set=full_leaf_set,
ml_thresh=ml_thresh,
progress=progress)
score.loc[i, "RetNum", 0]["ML"] = copy.copy(ret_score[0])
score.loc[i, "Time", 0]["ML"] = copy.copy(time_score[0])
ml_time = score.loc[i, "Time", 0]["ML"]
ml_ret = int(score.loc[i, "RetNum"]["ML"][0])
df_seq = pd.concat([df_seq, pd.Series(seq_ml)], axis=1)
# ML Trivial HEURISTIC
ret_score, time_score, seq_ml_triv = run_heuristic(
tree_set_newick=tree_set_newick,
inst_num=i,
repeats=1,
time_limit=time_limit,
pick_ml_triv=True,
relabel=True,
model_name=model_name,
problem_type="TrivialML",
full_leaf_set=full_leaf_set,
ml_thresh=ml_thresh,
progress=progress)
score.loc[i, "RetNum", 0]["TrivialML"] = copy.copy(ret_score[0])
score.loc[i, "Time", 0]["TrivialML"] = copy.copy(time_score[0])
ml_triv_ret = int(score.loc[i, "RetNum"]["TrivialML"][0])
df_seq = pd.concat([df_seq, pd.Series(seq_ml_triv)], axis=1)
# RANDOM HEURISTIC
ret_score, time_score, seq_ra = run_heuristic(
tree_set_newick=tree_set_newick,
inst_num=i,
repeats=repeats,
time_limit=ml_time,
problem_type="Rand",
pick_random=True,
relabel=False,
full_leaf_set=full_leaf_set,
progress=progress)
for r, ret in ret_score.items():
score.loc[i, "RetNum", r]["Rand"] = copy.copy(ret)
score.loc[i, "Time", r]["Rand"] = copy.copy(time_score[r])
ra_ret = int(min(score.loc[i, "RetNum"]["Rand"]))
df_seq = pd.concat([df_seq, pd.Series(seq_ra)], axis=1)
# TRIVIAL RANDOM
ret_score, time_score, seq_tr = run_heuristic(
tree_set_newick=tree_set_newick,
inst_num=i,
repeats=repeats,
time_limit=ml_time,
reduce_trivial=True,
relabel=True,
problem_type="TrivialRand",
full_leaf_set=full_leaf_set,
progress=progress)
for r, ret in ret_score.items():
score.loc[i, "RetNum", r]["TrivialRand"] = copy.copy(ret)
score.loc[i, "Time", r]["TrivialRand"] = copy.copy(time_score[r])
tr_ret = int(min(score.loc[i, "RetNum"]["TrivialRand"]))
df_seq = pd.concat([df_seq, pd.Series(seq_tr)], axis=1)
# upper bound of ret
env_info = pd.read_pickle(env_info_file)
ub_ret = int(env_info["metadata"]["rets"])
score.loc[i, "RetNum"]["UB"] = np.ones(repeats)*ub_ret
# print results
if progress:
print()
print("FINAL RESULTS\n"
f"Instance = {i} \n"
f"RETICULATIONS\n"
f"ML = {ml_ret}\n"
f"TrivialML = {ml_triv_ret}\n"
f"Rand = {ra_ret}\n"
f"TrivialRand = {tr_ret}\n"
f"Reference = {ub_ret}\n"
f"ML time = {np.round(ml_time)}s")
else:
print(i, ml_ret, ml_triv_ret, ra_ret, tr_ret, ub_ret, ml_time)
# save dataframes
# scores
score.dropna(axis=0, how="all").to_pickle(f"Data/Results/inst_results/heuristic_scores_lgt_ML[{ml_name}]_"
f"TEST[{test_info}]_"
f"{i}.pickle")
# ml predictions
df_pred.to_pickle(f"Data/Results/inst_results/cherry_prediction_info_ML[{ml_name}]_"
f"TEST[{test_info}]_"
f"{i}.pickle")
# best sequences
df_seq.columns = score.columns[:-1]
df_seq.index = pd.MultiIndex.from_product([[i], df_seq.index])
df_seq.to_pickle(f"Data/Results/inst_results/cherry_seq_ML[{ml_name}]_"
f"TEST[{test_info}]_"
f"{i}.pickle")
if __name__ == "__main__":
i = int(sys.argv[1])
ml_name = str(sys.argv[2])
l = int(sys.argv[3])
exact_input = int(sys.argv[4])
if exact_input:
exact = True
ret = int(sys.argv[5])
forest_size = None
else:
exact = False
ret = None
forest_size = int(sys.argv[5])
if len(sys.argv) == 7:
ml_thresh = int(sys.argv[6])
else:
ml_thresh = None
run_main(i, l, exact, ret, forest_size, ml_name=ml_name, full_leaf_set=True, ml_thresh=ml_thresh, progress=True)
