Пример #1
0
def eval_generated_data(num=4000):
    evaluator = RhymeEvaluator()

    planner = Planner()
    predictor = Seq2SeqPredictor()

    poems = []
    sentences = []
    i = 1
    with open("./data/samples/default.txt") as f:
        for text in f.readlines():

            sentences.append(text.strip()[:-1])
            if i % 4 == 0:
                poems.append(sentences)
                sentences = []
            i += 1
    # for _ in range(num):
    #     keywords = planner.plan(u'')
    #     assert 4 == len(keywords)
    #
    #     sentences = predictor.predict(keywords)
    #     poems.append(sentences)

    print("Testing {} quatrains generated by model.".format(num))
    eval_poems(evaluator, poems)
Пример #2
0
def main(args, cangtou=False):

    #planner = Planner()
    with Seq2SeqPredictor() as predictor:
        # Run loop
        terminate = False
        while not terminate:
            try:
                input = raw_input('Input Text:\n').decode('utf-8').strip()

                if not input:
                    print 'Input cannot be empty!'
                elif input.lower() in ['quit', 'exit']:
                    terminate = True
                else:
                    lines = predictor.predict(input)
                    lines = lines.replace('UNK', '江')
                    lines = lines[:-1] + rhyme_boosting(
                        lines[-1], get_rhyme(input[-1]))
                    print lines

            except EOFError:
                terminate = True
            except KeyboardInterrupt:
                terminate = True
    print '\nTerminated.'
def main():
    with Seq2SeqPredictor() as predictor:
        # Run loop
        terminate = False
        while not terminate:
            try:
                input = raw_input('Input Text, use | to separate keywords: \n'
                                  ).decode('utf-8').strip()

                if not input:
                    print 'Input cannot be empty!'
                elif input.lower() in ['quit', 'exit']:
                    terminate = True
                else:
                    # Generate poems
                    keywords = input.split('|')
                    lines = predictor.predict(keywords)

                    # Print keywords and poems
                    print 'Keyword:\tSentence:'
                    for line_number in xrange(len(lines)):
                        punctuation = u','
                        if line_number == len(lines) - 1:
                            punctuation = u'。'
                        print u'{keyword}\t{line}{punctuation}'.format(
                            keyword=keywords[line_number],
                            line=lines[line_number],
                            punctuation=punctuation)

            except EOFError:
                terminate = True
            except KeyboardInterrupt:
                terminate = True

    print '\nTerminated.'
Пример #4
0
def generate_rnn_samples(sampled_poems):
    planner = Planner()
    with Seq2SeqPredictor() as predictor:
        with open(rnn_samples_path, 'w+') as fout:
            for poem_idx, poem in enumerate(sampled_poems):
                input = string.join(poem).strip()
                keywords = planner.plan(input)

                print 'Predicting poem {}.'.format(poem_idx)
                lines = predictor.predict(keywords)

                for idx, sentence in enumerate(lines):
                    punctuation = u'\uff0c' if idx % 2 == 0 else u'\u3002'
                    line = (sentence + punctuation + '\n').encode('utf-8')
                    fout.write(line)
Пример #5
0
def eval_generated_data(num=100):
    evaluator = RhymeEvaluator()

    planner = Planner()
    predictor = Seq2SeqPredictor()

    poems = []
    for _ in range(num):
        keywords = planner.plan(u'')
        assert 4 == len(keywords)

        sentences = predictor.predictor(keywords)
        poems.append(sentences)

    print("Testing {} quatrains generated by model.".format(num))
    eval_poems(evaluator, poems)
Пример #6
0
def main(args, cangtou=False):
    planner = Planner()
    with Seq2SeqPredictor() as predictor:
        # Run loop
        terminate = False
        Judge = MatchUtil()
        while not terminate:
            try:
                input = args.Input.decode('utf-8').strip()
                if not input:
                    print 'Input cannot be empty!'
                elif input.lower() in ['quit', 'exit']:
                    terminate = True
                else:
                    if cangtou:
                        keywords = get_cangtou_keywords(input)
                    else:
                        # Generate keywords
                        keywords = planner.plan(input)

                    # Generate poem
                    lines = predictor.predict(keywords)

                    # whether the couplet is in accordance with the rules
                    result = Judge.eval_rhyme(lines)

                    if result == True:
                        # Print keywords and poem
                        print 'Keyword:\t\tPoem:'
                        for line_number in xrange(2):
                            punctuation = u',' if line_number % 2 == 0 else u'。'
                            print u'{keyword}\t\t{line}{punctuation}'.format(
                                keyword=keywords[line_number],
                                line=lines[line_number],
                                punctuation=punctuation)
                            terminate = True

            except EOFError:
                terminate = True
            except KeyboardInterrupt:
                terminate = True
    print '\nTerminated.'
Пример #7
0
def main(args, cangtou=False):

    planner = Planner()
    with Seq2SeqPredictor() as predictor:
        # Run loop
        terminate = False
        while not terminate:
            try:
                input = raw_input('Input Text:\n').decode('utf-8').strip()

                if not input:
                    print 'Input cannot be empty!'
                elif input.lower() in ['quit', 'exit']:
                    terminate = True
                else:
                    if cangtou:
                        keywords = get_cangtou_keywords(input)
                    else:
                        # Generate keywords
                        keywords = planner.plan(input)

                    # Generate poem
                    if args.nocouplet:
                        lines = predictor.predict(keywords)
                    else:
                        lines = predictor.predict_with_couplet(keywords)

                    # Print keywords and poem
                    print 'Keyword:\t\tPoem:'
                    for line_number in xrange(4):
                        punctuation = u',' if line_number % 2 == 0 else u'。'
                        print u'{keyword}\t\t{line}{punctuation}'.format(
                            keyword=keywords[line_number],
                            line=lines[line_number],
                            punctuation=punctuation)

            except EOFError:
                terminate = True
            except KeyboardInterrupt:
                terminate = True
    print '\nTerminated.'
Пример #8
0
def main(cangtou=False):
    planner = Planner()
    with Seq2SeqPredictor() as predictor:
        # Run loop
        terminate = False
        while not terminate:
            try:
                import ipdb
                ipdb.set_trace()
                input = input('Input Text:\n').decode('utf-8').strip()

                if not input:
                    print('Input cannot be empty!')
                elif input.lower() in ['quit', 'exit']:
                    terminate = True
                else:
                    if cangtou:
                        keywords = get_cangtou_keywords(input)
                    else:
                        # Generate keywords
                        keywords = planner.plan(input)

                    # Generate poem
                    lines = predictor.predict(keywords)

                    # Print keywords and poem
                    print('Keyword:\t\tPoem:')
                    for line_number in range(4):
                        punctuation = ',' if line_number % 2 == 0 else '。'
                        print('{keyword}\t\t{line}{punctuation}'.format(
                            keyword=keywords[line_number],
                            line=lines[line_number],
                            punctuation=punctuation))

            except EOFError:
                terminate = True
            except KeyboardInterrupt:
                terminate = True
    print('\nTerminated.')
Пример #9
0
def main(cangtou=False):
    planner = Planner()
    with Seq2SeqPredictor() as predictor:
        # Run loop
        terminate = False
        while not terminate:
            try:
                inputs =input('Input Text:\n').strip()

                if not inputs:
                    print( 'Input cannot be empty!')
                elif inputs.lower() in ['quit', 'exit']:
                    terminate = True
                else:
                    if cangtou:
                        keywords = get_cangtou_keywords(inputs)
                    else:
                        # Generate keywords
                        #将输入的句子切词并按照textrank 值进行降序排列,并且选择前四个词作为keyword
                        keywords = planner.plan(inputs)
                        print(keywords)
                    # Generate poem
                    lines = predictor.predict(keywords)
                    # Print keywords and poem
                    print( 'Keyword:\t\tPoem:')
                    for line_number in range(4):
                        punctuation = u',' if line_number % 2 == 0 else u'。'
                        print (u'{keyword}\t\t{line}{punctuation}'.format(
                            keyword=keywords[line_number],
                            line=lines[line_number],
                            punctuation=punctuation
                        ))

            except EOFError:
                terminate = True
            except KeyboardInterrupt:
                terminate = True
    print ('\nTerminated.')
Пример #10
0
 def __init__(self):
     self.planner = Planner()
     self.predictor = Seq2SeqPredictor()
     self.Judge = MatchUtil()
Пример #11
0
def main(args):

    #planner = Planner()
    with Seq2SeqPredictor() as predictor:

        def client_thread(conn, ip, port, MAX_BUFFER_SIZE=4096):

            # the input is in bytes, so decode it
            input_from_client_bytes = conn.recv(MAX_BUFFER_SIZE)

            # MAX_BUFFER_SIZE is how big the message can be
            # this is test if it's sufficiently big
            import sys
            siz = sys.getsizeof(input_from_client_bytes)
            if siz >= MAX_BUFFER_SIZE:
                print(
                    "The length of input is probably too long: {}".format(siz))

            # decode input and strip the end of line
            input_from_client = input_from_client_bytes.decode("utf8").rstrip()
            ryhme_sentence, input_sentence = input_from_client[:len(
                input_from_client
            ) / 2], input_from_client[len(input_from_client) / 2:]

            res = predictor.predict(input_sentence)
            res = res.replace('UNK', '江')
            if args.fixrhyme:
                res = res[:-1] + \
                    rhyme_boosting(res[-1], ryhme_sentence[-1])
            print("Result of processing {} is: {}".format(input_sentence, res))

            vysl = res.encode("utf8")  # encode the result string
            conn.sendall(vysl)  # send it to client
            conn.close()  # close connection
            print('Connection ' + ip + ':' + port + " ended")

        def start_server():

            import socket
            soc = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
            # this is for easy starting/killing the app
            soc.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
            print('Socket created')

            try:
                soc.bind(("127.0.0.1", 12345))
                print('Socket bind complete')
            except socket.error as msg:
                import sys
                print('Bind failed. Error : ' + str(sys.exc_info()))
                sys.exit()

            # Start listening on socket
            soc.listen(10)
            print('Socket now listening')

            # for handling task in separate jobs we need threading
            from threading import Thread

            # this will make an infinite loop needed for
            # not reseting server for every client
            while True:
                conn, addr = soc.accept()
                ip, port = str(addr[0]), str(addr[1])
                print('Accepting connection from ' + ip + ':' + port)
                try:
                    Thread(target=client_thread, args=(conn, ip, port)).start()
                except:
                    print("Terible error!")
                    import traceback
                    traceback.print_exc()
            soc.close()

        start_server()
Пример #12
0
def main():

    #planner = Planner()
    with Seq2SeqPredictor() as predictor:
        def client_thread(conn, ip, port, MAX_BUFFER_SIZE = 4096):

            # the input is in bytes, so decode it
            input_from_client_bytes = conn.recv(MAX_BUFFER_SIZE)

            # MAX_BUFFER_SIZE is how big the message can be
            # this is test if it's sufficiently big
            import sys
            siz = sys.getsizeof(input_from_client_bytes)
            if  siz >= MAX_BUFFER_SIZE:
                print("The length of input is probably too long: {}".format(siz))

            # decode input and strip the end of line
            input_from_client = input_from_client_bytes.decode("utf8").rstrip()

            res = predictor.predict(input_from_client)
            print("Result of processing {} is: {}".format(input_from_client, res))

            vysl = res.encode("utf8")  # encode the result string
            conn.sendall(vysl)  # send it to client
            conn.close()  # close connection
            print('Connection ' + ip + ':' + port + " ended")

        def start_server():

            import socket
            soc = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
            # this is for easy starting/killing the app
            soc.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
            print('Socket created')

            try:
                soc.bind(("127.0.0.1", 12345))
                print('Socket bind complete')
            except socket.error as msg:
                import sys
                print('Bind failed. Error : ' + str(sys.exc_info()))
                sys.exit()

            #Start listening on socket
            soc.listen(10)
            print('Socket now listening')

            # for handling task in separate jobs we need threading
            from threading import Thread

            # this will make an infinite loop needed for
            # not reseting server for every client
            while True:
                conn, addr = soc.accept()
                ip, port = str(addr[0]), str(addr[1])
                print('Accepting connection from ' + ip + ':' + port)
                try:
                    Thread(target=client_thread, args=(conn, ip, port)).start()
                except:
                    print("Terible error!")
                    import traceback
                    traceback.print_exc()
            soc.close()
        start_server()
        # Run loop
        terminate = False
        while not terminate:
            try:
                input = raw_input('Input Text:\n').decode('utf-8').strip()

                if not input:
                    print 'Input cannot be empty!'
                elif input.lower() in ['quit', 'exit']:
                    terminate = True
                else:
                    lines = predictor.predict(input)
                    print lines

            except EOFError:
                terminate = True
            except KeyboardInterrupt:
                terminate = True
    print '\nTerminated.'
import codecs
from predict import Seq2SeqPredictor

if __name__ == '__main__':
    if len(sys.argv) < 4:
        print "Usage: python", sys.argv[0], "<input> <count> <output>"
        exit(-1)
    else:
        fn_in = sys.argv[1]
        num = int(sys.argv[2])
        fn_out = sys.argv[3]

    fr = codecs.open(fn_in, 'r', 'utf-8')
    fw = codecs.open(fn_out, 'w', 'utf-8')

    with Seq2SeqPredictor() as predictor:
        for line in fr:
            line = line.strip()

            # topic examples
            # 时尚教主_喜悦   潮人 气派 时尚教主 高兴 为之一喜 欢呼雀跃
            # 时尚教主_喜悦   潮人 兴高彩烈 教主 可贺 惊喜万分 手舞足蹈

            [label, topic] = line.split('\t')
            keywords = topic.split('|')

            # predict poetry lines, totally num poems
            for k in range(0, num):

                # output label and keywords first
                fw.write(label + '\n')