source: framspy/FramsticksCLI.py @ 1024

Last change on this file since 1024 was 1024, checked in by Maciej Komosinski, 4 months ago

evaluate() now handles a list of genotypes, not just a single genotype

File size: 16.0 KB
Line 
1from subprocess import Popen, PIPE, check_output
2from enum import Enum
3from typing import List  # to be able to specify a type hint of list(something)
4from itertools import count  # for tracking multiple instances
5import json
6import sys, os
7import argparse
8import numpy as np
9
10
11class FramsticksCLI:
12        """Runs Framsticks CLI (command-line) executable and communicates with it using standard input and output.
13        You can perform basic operations like mutation, crossover, and evaluation of genotypes.
14        This way you can perform evolution controlled by python as well as access and manipulate genotypes.
15        You can even design and use in evolution your own genetic representation implemented entirely in python.
16
17        You need to provide one or two parameters when you run this class: the path to Framsticks CLI
18        and the name of the Framsticks CLI executable (if it is non-standard). See::
19                FramsticksCLI.py -h"""
20
21        PRINT_FRAMSTICKS_OUTPUT: bool = False  # set to True for debugging
22        DETERMINISTIC: bool = False  # set to True to have the same results on each run
23
24        GENO_SAVE_FILE_FORMAT = Enum('GENO_SAVE_FILE_FORMAT', 'NATIVEFRAMS RAWGENO')  # how to save genotypes
25        OUTPUT_DIR = "scripts_output"
26        STDOUT_ENDOPER_MARKER = "FileObject.write"  # we look for this message on Framsticks CLI stdout to detect when Framsticks created a file with the result we expect
27
28        FILE_PREFIX = 'framspy_'
29
30        RANDOMIZE_CMD = "Math.randomize();"
31        SETEXPEDEF_CMD = "Simulator.expdef=\"standard-eval\";"
32        GETSIMPLEST_CMD = "getsimplest"
33        GETSIMPLEST_FILE = "simplest.gen"
34        EVALUATE_CMD = "evaluate eval-allcriteria.sim"
35        EVALUATE_FILE = "genos_eval.json"
36        CROSSOVER_CMD = "crossover"
37        CROSSOVER_FILE = "child.gen"
38        DISSIMIL_CMD = "dissimil"
39        DISSIMIL_FILE = "dissimilarity_matrix.tsv"  # tab-separated values
40        ISVALID_CMD = "arevalid"
41        ISVALID_FILE = "validity.txt"
42        MUTATE_CMD = "mutate"
43        MUTATE_FILE = "mutant.gen"
44
45        CLI_INPUT_FILE = "genotypes.gen"
46
47        _next_instance_id = count(0)  # "static" counter incremented when a new instance is created. Used to ensure unique filenames for each instance.
48
49
50        def __init__(self, framspath, framsexe, pid=""):
51                self.pid = pid if pid is not None else ""
52                self.id = next(FramsticksCLI._next_instance_id)
53                self.frams_path = framspath
54                self.frams_exe = framsexe if framsexe is not None else 'frams.exe' if os.name == "nt" else 'frams.linux'
55                self.writing_path = None
56                mainpath = os.path.join(self.frams_path, self.frams_exe)
57                exe_call = [mainpath, '-Q', '-s', '-c', '-icliutils.ini']  # -c will be ignored in Windows Framsticks (this option is meaningless because the Windows version does not support color console, so no need to deactivate this feature using -c)
58                exe_call_to_get_version = [mainpath, '-V']
59                exe_call_to_get_path = [mainpath, '-?']
60                try:
61                        print("\n".join(self.__readAllOutput(exe_call_to_get_version)))
62                        help = self.__readAllOutput(exe_call_to_get_path)
63                        for helpline in help:
64                                if 'dDIRECTORY' in helpline:
65                                        self.writing_path = helpline.split("'")[1]
66                except FileNotFoundError:
67                        print("Could not find Framsticks executable ('%s') in the given location ('%s')." % (self.frams_exe, self.frams_path))
68                        sys.exit(1)
69                print("Temporary files with results will be saved in detected writable working directory '%s'" % self.writing_path)
70                self.__spawnFramsticksCLI(exe_call)
71
72
73        def __readAllOutput(self, command):
74                frams_process = Popen(command, stdout=PIPE, stderr=PIPE, stdin=PIPE)
75                return [line.decode('utf-8').rstrip() for line in iter(frams_process.stdout.readlines())]
76
77
78        def __spawnFramsticksCLI(self, args):
79                # the child app (Framsticks CLI) should not buffer outputs and we need to immediately read its stdout, hence we use pexpect/wexpect
80                print('Spawning Framsticks CLI for continuous stdin/stdout communication... ', end='')
81                if os.name == "nt":  # Windows:
82                        import wexpect  # https://pypi.org/project/wexpect/
83                        # https://github.com/raczben/wexpect/tree/master/examples
84                        self.child = wexpect.spawn(' '.join(args))
85                else:
86                        import pexpect  # https://pexpect.readthedocs.io/en/stable/
87                        self.child = pexpect.spawn(' '.join(args))
88                self.child.setecho(False)  # ask the communication to not copy to stdout what we write to stdin
89                print('OK.')
90
91                self.__readFromFramsCLIUntil("UserScripts.autoload")
92                print('Performing a basic test 1/2... ', end='')
93                assert self.getSimplest("1") == "X"
94                print('OK.')
95                print('Performing a basic test 2/2... ', end='')
96                assert self.isValid(["X[0:0],", "X[0:0]", "X[1:0]"]) == [False, True, False]
97                print('OK.')
98                if not self.DETERMINISTIC:
99                        self.sendDirectCommand(self.RANDOMIZE_CMD)
100                self.sendDirectCommand(self.SETEXPEDEF_CMD)
101
102
103        def closeFramsticksCLI(self):
104                # End gracefully by sending end-of-file character: ^Z or ^D
105                # Without the -Q argument ("quiet mode"), Framsticks CLI would print "Shell closed." for goodbye.
106                self.child.sendline(chr(26 if os.name == "nt" else 4))
107
108
109        def __getPrefixedFilename(self, filename: str) -> str:
110                # Returns filename with unique instance id appended so there is no clash when many instances of this class use the same Framsticks CLI executable
111                return FramsticksCLI.FILE_PREFIX + self.pid + str(chr(ord('A') + self.id)) + '_' + filename
112
113
114        def __saveGenotypeToFile(self, genotype, name, mode, saveformat):
115                relname = self.__getPrefixedFilename(name)
116                absname = os.path.join(self.writing_path, relname)
117                if mode == 'd':  # special mode, 'delete'
118                        if os.path.exists(absname):
119                                os.remove(absname)
120                else:
121                        outfile = open(absname, mode)
122                        if saveformat == self.GENO_SAVE_FILE_FORMAT["RAWGENO"]:
123                                outfile.write(genotype)
124                        else:
125                                outfile.write("org:\n")
126                                outfile.write("genotype:~\n")
127                                outfile.write(genotype + "~\n\n")  # TODO proper quoting of special characters in genotype...
128                        outfile.close()
129                return relname, absname
130
131
132        def __readFromFramsCLIUntil(self, until_marker: str) -> str:
133                output = ""
134                while True:
135                        self.child.expect('\r\n' if os.name == "nt" else '\n')
136                        msg = str(self.child.before)
137                        if self.PRINT_FRAMSTICKS_OUTPUT or msg.startswith("[ERROR]") or msg.startswith("[CRITICAL]"):
138                                print(msg)
139                        if until_marker in msg:
140                                break
141                        else:
142                                output += msg + '\n'
143                return output
144
145
146        def __runCommand(self, command, genotypes, result_file_name, saveformat) -> List[str]:
147                filenames_rel = []  # list of file names with input data for the command
148                filenames_abs = []  # same list but absolute paths actually used
149                if saveformat == self.GENO_SAVE_FILE_FORMAT["RAWGENO"]:
150                        for i in range(len(genotypes)):
151                                # plain text format = must have a separate file for each genotype
152                                rel, abs = self.__saveGenotypeToFile(genotypes[i], "genotype" + str(i) + ".gen", "w", self.GENO_SAVE_FILE_FORMAT["RAWGENO"])
153                                filenames_rel.append(rel)
154                                filenames_abs.append(abs)
155                elif saveformat == self.GENO_SAVE_FILE_FORMAT["NATIVEFRAMS"]:
156                        self.__saveGenotypeToFile(None, self.CLI_INPUT_FILE, 'd', None)  # 'd'elete: ensure there is nothing left from the last run of the program because we "a"ppend to file in the loop below
157                        for i in range(len(genotypes)):
158                                rel, abs = self.__saveGenotypeToFile(genotypes[i], self.CLI_INPUT_FILE, "a", self.GENO_SAVE_FILE_FORMAT["NATIVEFRAMS"])
159                        #  since we use the same file in the loop above, add this file only once (i.e., outside of the loop)
160                        filenames_rel.append(rel)
161                        filenames_abs.append(abs)
162
163                result_file_name = self.__getPrefixedFilename(result_file_name)
164                cmd = command + " " + " ".join(filenames_rel) + " " + result_file_name
165                self.child.sendline(cmd)
166                self.__readFromFramsCLIUntil(self.STDOUT_ENDOPER_MARKER)
167                filenames_abs.append(os.path.join(self.writing_path, self.OUTPUT_DIR, result_file_name))
168                return filenames_abs  # last element is a path to the file containing results
169
170
171        def __cleanUpCommandResults(self, filenames):
172                """Deletes files with results just created by the command."""
173                for name in filenames:
174                        os.remove(name)
175
176
177        sendDirectCommand_counter = count(0)  # an internal counter for the sendDirectCommand() method; should be static within that method but python does not allow
178
179
180        def sendDirectCommand(self, command: str) -> str:
181                """Sends any command to Framsticks CLI. Use when you know Framsticks and its scripting language, Framscript.
182
183                Returns:
184                        The output of the command, likely with extra \\n because for each entered command, Framsticks CLI responds with a (muted in Quiet mode) prompt and a \\n.
185                """
186                self.child.sendline(command.strip())
187                next(FramsticksCLI.sendDirectCommand_counter)
188                STDOUT_ENDOPER_MARKER = "uniqe-marker-" + str(FramsticksCLI.sendDirectCommand_counter)
189                self.child.sendline("Simulator.print(\"%s\");" % STDOUT_ENDOPER_MARKER)
190                return self.__readFromFramsCLIUntil(STDOUT_ENDOPER_MARKER)
191
192
193        def getSimplest(self, genetic_format) -> str:
194                files = self.__runCommand(self.GETSIMPLEST_CMD + " " + genetic_format + " ", [], self.GETSIMPLEST_FILE, self.GENO_SAVE_FILE_FORMAT["RAWGENO"])
195                with open(files[-1]) as f:
196                        genotype = "".join(f.readlines())
197                self.__cleanUpCommandResults(files)
198                return genotype
199
200
201        def evaluate(self, genotype_list: List[str]):
202                """
203                Returns:
204                        List of dictionaries containing the performance of genotypes evaluated with self.EVALUATE_COMMAND.
205                        Note that for whatever reason (e.g. incorrect genotype), the dictionaries you will get may be empty or
206                        partially empty and may not have the fields you expected, so handle such cases properly.
207                """
208                assert isinstance(genotype_list, list)  # because in python str has similar capabilities as list and here it would pretend to work too, so to avoid any ambiguity
209                files = self.__runCommand(self.EVALUATE_CMD, genotype_list, self.EVALUATE_FILE, self.GENO_SAVE_FILE_FORMAT["NATIVEFRAMS"])
210                with open(files[-1]) as f:
211                        data = json.load(f)
212                if len(data) > 0:
213                        self.__cleanUpCommandResults(files)
214                        return data
215                else:
216                        print("Evaluating genotype: no performance data was returned in", self.EVALUATE_FILE)  # we do not delete files here
217                        return None
218
219
220        def mutate(self, genotype: str) -> str:
221                """
222                Returns:
223                        The genotype of the mutated individual. Empty string if the mutation failed.
224                """
225                files = self.__runCommand(self.MUTATE_CMD, [genotype], self.MUTATE_FILE, self.GENO_SAVE_FILE_FORMAT["RAWGENO"])
226                with open(files[-1]) as f:
227                        newgenotype = "".join(f.readlines())
228                self.__cleanUpCommandResults(files)
229                return newgenotype
230
231
232        def crossOver(self, genotype_parent1: str, genotype_parent2: str) -> str:
233                """
234                Returns:
235                        The genotype of the offspring. Empty string if the crossing over failed.
236                """
237                files = self.__runCommand(self.CROSSOVER_CMD, [genotype_parent1, genotype_parent2], self.CROSSOVER_FILE, self.GENO_SAVE_FILE_FORMAT["RAWGENO"])
238                with open(files[-1]) as f:
239                        child_genotype = "".join(f.readlines())
240                self.__cleanUpCommandResults(files)
241                return child_genotype
242
243
244        def dissimilarity(self, genotype_list: List[str]) -> np.ndarray:
245                """
246                Returns:
247                        A square array with dissimilarities of each pair of genotypes.
248                """
249                assert isinstance(genotype_list, list)  # because in python str has similar capabilities as list and here it would pretend to work too, so to avoid any ambiguity
250                files = self.__runCommand(self.DISSIMIL_CMD, genotype_list, self.DISSIMIL_FILE, self.GENO_SAVE_FILE_FORMAT["NATIVEFRAMS"])
251                with open(files[-1]) as f:
252                        dissimilarity_matrix = np.genfromtxt(f, dtype=np.float64, comments='#', encoding=None, delimiter='\t')
253                # We would like to skip column #1 while reading and read everything else, but... https://stackoverflow.com/questions/36091686/exclude-columns-from-genfromtxt-with-numpy
254                # This would be too complicated, so strings (names) in column #1 become NaN as floats (unless they accidentally are valid numbers) - not great, not terrible
255                square_matrix = dissimilarity_matrix[:, 2:]  # get rid of two first columns (fitness and name)
256                EXPECTED_SHAPE = (len(genotype_list), len(genotype_list))
257                # print(square_matrix)
258                assert square_matrix.shape == EXPECTED_SHAPE, f"Not a correct dissimilarity matrix, expected {EXPECTED_SHAPE} "
259                for i in range(len(square_matrix)):
260                        assert square_matrix[i][i] == 0, "Not a correct dissimilarity matrix, diagonal expected to be 0"
261                assert (square_matrix == square_matrix.T).all(), "Probably not a correct dissimilarity matrix, expecting symmetry, verify this"  # could introduce tolerance in comparison (e.g. class field DISSIMIL_DIFF_TOLERANCE=10^-5) so that miniscule differences do not fail here
262                self.__cleanUpCommandResults(files)
263                return square_matrix
264
265
266        def isValid(self, genotype_list: List[str]) -> List[bool]:
267                assert isinstance(genotype_list, list)  # because in python str has similar capabilities as list and here it would pretend to work too, so to avoid any ambiguity
268                files = self.__runCommand(self.ISVALID_CMD, genotype_list, self.ISVALID_FILE, self.GENO_SAVE_FILE_FORMAT["NATIVEFRAMS"])
269                valid = []
270                with open(files[-1]) as f:
271                        for line in f:
272                                valid.append(line.strip() == "1")
273                self.__cleanUpCommandResults(files)
274                assert len(genotype_list) == len(valid), "Submitted %d genotypes, received %d validity values" % (len(genotype_list), len(valid))
275                return valid
276
277
278def parseArguments():
279        parser = argparse.ArgumentParser(description='Run this program with "python -u %s" if you want to disable buffering of its output.' % sys.argv[0])
280        parser.add_argument('-path', type=ensureDir, required=True, help='Path to Framsticks CLI without trailing slash.')
281        parser.add_argument('-exe', required=False, help='Executable name. If not given, "frams.exe" or "frams.linux" is assumed depending on the platform.')
282        parser.add_argument('-genformat', required=False, help='Genetic format for the demo run, for example 4, 9, or S. If not given, f1 is assumed.')
283        parser.add_argument('-pid', required=False, help='Unique ID of this process. Only relevant when you run multiple instances of this class simultaneously but as separate processes, and they use the same Framsticks CLI executable. This value will be appended to the names of created files to avoid conflicts.')
284        return parser.parse_args()
285
286
287def ensureDir(string):
288        if os.path.isdir(string):
289                return string
290        else:
291                raise NotADirectoryError(string)
292
293
294if __name__ == "__main__":
295        # A demo run.
296
297        # TODO ideas:
298        # - check_validity with three levels (invalid, corrected, valid)
299        # - "vectorize" some operations (isvalid, evaluate) so that a number of genotypes is handled in one call
300        # - use threads for non-blocking reading from frams' stdout and thus not relying on specific strings printed by frams
301        # - a pool of binaries run at the same time, balance load - in particular evaluation
302        # - if we read genotypes in "org:" format anywhere: import https://pypi.org/project/framsreader/0.1.2/ and use it if successful,
303        #    if not then print a message "framsreader not available, using simple internal method to save a genotype" and proceed as it is now.
304        #    So far we don't read, but we should use the proper writer to handle all special cases like quoting etc.
305
306        parsed_args = parseArguments()
307        framsCLI = FramsticksCLI(parsed_args.path, parsed_args.exe, parsed_args.pid)
308
309        print("Sending a direct command to Framsticks CLI that calculates \"4\"+2 yields", repr(framsCLI.sendDirectCommand("Simulator.print(\"4\"+2);")))
310
311        simplest = framsCLI.getSimplest('1' if parsed_args.genformat is None else parsed_args.genformat)
312        print("\tSimplest genotype:", simplest)
313        parent1 = framsCLI.mutate(simplest)
314        parent2 = parent1
315        MUTATE_COUNT = 10
316        for x in range(MUTATE_COUNT):  # example of a chain of 20 mutations
317                parent2 = framsCLI.mutate(parent2)
318        print("\tParent1 (mutated simplest):", parent1)
319        print("\tParent2 (Parent1 mutated %d times):" % MUTATE_COUNT, parent2)
320        offspring = framsCLI.crossOver(parent1, parent2)
321        print("\tCrossover (Offspring):", offspring)
322        print('\tDissimilarity of Parent1 and Offspring:', framsCLI.dissimilarity([parent1, offspring])[0, 1])
323        print('\tPerformance of Offspring:', framsCLI.evaluate([offspring]))
324        print('\tValidity of Parent1, Parent 2, and Offspring:', framsCLI.isValid([parent1, parent2, offspring]))
325
326        framsCLI.closeFramsticksCLI()
Note: See TracBrowser for help on using the repository browser.