Python config.Config方法代码示例

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    """Procedure to build data

    You MUST RUN this procedure to preprocess datasets. 


    Args:
        config: (instance of Config) has attributes like hyper-params...

    """
    # get config
    config = Config(load=False)

    # Process pre trained word vectors
    process_wordvectors(config.filename_wordvectors, config.filename_words, \
                        config.filename_embeddings)

    # Process relation2id
    process_relation2id(config.filename_relation_origin, config.filename_relation)

    # Process train and test datasets
    check_entity_in_sentence(config.filename_train_origin, config.filename_train, \
                            config.filename_train_wrong)
    check_entity_in_sentence(config.filename_test_origin, config.filename_test, \
                            config.filename_test_wrong) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()

    # build model
    model = PCNNModel(config)
    model.build()
    model.restore_session(config.restore_model)

    # create dataset
    test  = getDataset(config.filename_test, config.processing_word,
                         config.processing_tag, config.max_iter)

    # evaluate and interact
    model.evaluate(test)
    interactive_shell(model) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()

    # build model
    model = PCNNModel(config)
    model.build()
    # model.restore_session("results/crf/model.weights/") # optional, restore weights
    # model.reinitialize_weights("proj")

    # create datasets
    dev   = getDataset(config.filename_dev, config.processing_word,
                         config.processing_relation, config.max_iter)
    train = getDataset(config.filename_train, config.processing_word,
                         config.processing_relation, config.max_iter)

    # train model
    model.train(train, dev) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main(argv=None):
    # Configurations
    config = Config()
    config.DATA_DIR = ['/data/']
    config.LOG_DIR = './log/model'
    config.MODE = 'training'
    config.STEPS_PER_EPOCH_VAL = 180
    config.display()

    # Get images and labels.
    dataset_train = Dataset(config, 'train')
    # Build a Graph
    model = Model(config)

    # Train the model
    model.compile()
    model.train(dataset_train, None) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()
    if config.use_elmo: config.processing_word = None

    #build model
    model = NERModel(config)

    # create datasets
    dev = CoNLLDataset(config.filename_dev, config.processing_word,
                         config.processing_tag, config.max_iter, config.use_crf)
    train = CoNLLDataset(config.filename_train, config.processing_word,
                         config.processing_tag, config.max_iter, config.use_crf)

    learn = NERLearner(config, model)
    learn.fit(train, dev) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def predict():
    config=Config()
    threshold=(config.sequence_length/2)+1
    config.batch_size=1
    model = BertModel(config)
    gpu_config = tf.ConfigProto()
    gpu_config.gpu_options.allow_growth = True
    saver = tf.train.Saver()
    ckpt_dir = config.ckpt_dir
    print("ckpt_dir:",ckpt_dir)
    with tf.Session(config=gpu_config) as sess:
        sess.run(tf.global_variables_initializer())
        saver.restore(sess, tf.train.latest_checkpoint(ckpt_dir))
        for i in range(100):
            # 2.feed data
            input_x = np.random.randn(config.batch_size, config.sequence_length)  # [None, self.sequence_length]
            input_x[input_x >= 0] = 1
            input_x[input_x < 0] = 0
            target_label = generate_label(input_x,threshold)
            input_sum=np.sum(input_x)
            # 3.run session to train the model, print some logs.
            logit,prediction = sess.run([model.logits, model.predictions],feed_dict={model.input_x: input_x ,model.dropout_keep_prob: config.dropout_keep_prob})
            print("target_label:", target_label,";input_sum:",input_sum,"threshold:",threshold,";prediction:",prediction);
            print("input_x:",input_x,";logit:",logit) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def predict():
    config=Config()
    threshold=(config.sequence_length/2)+1
    config.batch_size=1
    model = BertCNNModel(config)
    gpu_config = tf.ConfigProto()
    gpu_config.gpu_options.allow_growth = True
    saver = tf.train.Saver()
    ckpt_dir = config.ckpt_dir
    print("ckpt_dir:",ckpt_dir)
    with tf.Session(config=gpu_config) as sess:
        sess.run(tf.global_variables_initializer())
        saver.restore(sess, tf.train.latest_checkpoint(ckpt_dir))
        for i in range(100):
            # 2.feed data
            input_x = np.random.randn(config.batch_size, config.sequence_length)  # [None, self.sequence_length]
            input_x[input_x >= 0] = 1
            input_x[input_x < 0] = 0
            target_label = generate_label(input_x,threshold)
            input_sum=np.sum(input_x)
            # 3.run session to train the model, print some logs.
            logit,prediction = sess.run([model.logits, model.predictions],feed_dict={model.input_x: input_x ,model.dropout_keep_prob: config.dropout_keep_prob})
            print("target_label:", target_label,";input_sum:",input_sum,"threshold:",threshold,";prediction:",prediction);
            print("input_x:",input_x,";logit:",logit) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()

    # build model
    model = ASPECTModel(config)
    model.build()
    model.restore_session(config.dir_model)

    # create dataset
    test  = CoNLLDataset(config.filename_test, config.processing_word,
                         config.processing_tag, config.max_iter)

    # evaluate and interact
    model.evaluate(test)
    interactive_shell(model) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()

    # build model
    model = ASPECTModel(config)
    model.build()
    # model.restore_session("results/crf/model.weights/") # optional, restore weights
    # model.reinitialize_weights("proj")

    # create datasets
    dev   = CoNLLDataset(config.filename_dev, config.processing_word,
                         config.processing_tag, config.max_iter)
    train = CoNLLDataset(config.filename_train, config.processing_word,
                         config.processing_tag, config.max_iter)

    # train model
    model.train(train, dev) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()

    model = BLSTMCRF(config) #Word_BLSTM(config)

    model.build()
    model.compile(optimizer=model.get_optimizer(), loss=model.get_loss()) #, metrics=['acc']

    model.load_weights('./saves/test20.h5') #./saves/blstmCrf_15.h5

    test = CoNLLDataset(config.filename_test, config.processing_word,
                        config.processing_tag, config.max_iter)

    batch_size = config.batch_size
    nbatches_test, test_generator = model.batch_iter(test, batch_size, return_lengths=True)

    model.run_evaluate(test_generator, nbatches_test)
    # test predictions
    words = "Fa Mulan is from Dynasty Trading Limited"
    words = words.split(" ")
    pred = model.predict_words(words)
    print(words)
    print(pred) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()

    # build model
    model = NERModel(config)
    model.build()
    model.restore_session(config.dir_model)

    # create dataset
    test  = CoNLLDataset(config.filename_test, config.processing_word,
                         config.processing_tag, config.max_iter)

    # evaluate and interact
    model.evaluate(test)
    interactive_shell(model) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()

    # build model
    model = NERModel(config)
    model.build()
    # model.restore_session("results/crf/model.weights/") # optional, restore weights
    # model.reinitialize_weights("proj")

    # create datasets
    dev   = CoNLLDataset(config.filename_dev, config.processing_word,
                         config.processing_tag, config.max_iter)
    train = CoNLLDataset(config.filename_train, config.processing_word,
                         config.processing_tag, config.max_iter)

    # train model
    model.train(train, dev) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def main():
    # create instance of config
    config = Config()
    if config.use_elmo: config.processing_word = None

    #build model
    model = NERModel(config)

    learn = NERLearner(config, model)
    learn.load()

    if len(sys.argv) == 1:
        print("No arguments given. Running full test")
        sys.argv.append("eval")
        sys.argv.append("pred")

    if sys.argv[1] == "eval":
        # create datasets
        test = CoNLLDataset(config.filename_test, config.processing_word,
                             config.processing_tag, config.max_iter)
        learn.evaluate(test)

    if sys.argv[1] == "pred" or sys.argv[2] == "pred":
        try:
            sent = (sys.argv[2] if sys.argv[1] == "pred" else sys.argv[3])
        except IndexError:
            sent = ["Peter", "Johnson", "lives", "in", "Los", "Angeles"]

        print("Predicting sentence: ", sent)
        pred = learn.predict(sent)
        print(pred) 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def train():
    # 1.init config and model
    config=Config()
    threshold=(config.sequence_length/2)+1
    model = BertModel(config)
    gpu_config = tf.ConfigProto()
    gpu_config.gpu_options.allow_growth = True
    saver = tf.train.Saver()
    save_path = config.ckpt_dir + "model.ckpt"
    #if not os.path.exists(config.ckpt_dir):
    #    os.makedirs(config.ckpt_dir)
    with tf.Session(config=gpu_config) as sess:
        sess.run(tf.global_variables_initializer())
        if os.path.exists(config.ckpt_dir): #
            saver.restore(sess, tf.train.latest_checkpoint(save_path))
        for i in range(10000):
            # 2.feed data
            input_x = np.random.randn(config.batch_size, config.sequence_length)  # [None, self.sequence_length]
            input_x[input_x >= 0] = 1
            input_x[input_x < 0] = 0
            input_y = generate_label(input_x,threshold)
            p_mask_lm=[i for i in range(batch_size)]
            # 3.run session to train the model, print some logs.
            loss, _ = sess.run([model.loss_val,  model.train_op],feed_dict={model.x_mask_lm: input_x, model.y_mask_lm: input_y,model.p_mask_lm:p_mask_lm,
                                                                            model.dropout_keep_prob: config.dropout_keep_prob})
            print(i, "loss:", loss, "-------------------------------------------------------")
            if i==300:
                print("label[0]:", input_y[0]);print("input_x:",input_x)
            if i % 500 == 0:
                saver.save(sess, save_path, global_step=i)

# use saved checkpoint from model to make prediction, and print it, to see whether it is able to do toy task successfully. 

# 需要导入模块: from model import config [as 别名]
# 或者: from model.config import Config [as 别名]
def train():
    # 1.init config and model
    config=Config()
    threshold=(config.sequence_length/2)+1
    model = BertCNNModel(config)
    gpu_config = tf.ConfigProto()
    gpu_config.gpu_options.allow_growth = True
    saver = tf.train.Saver()
    save_path = config.ckpt_dir + "model.ckpt"
    #if not os.path.exists(config.ckpt_dir):
    #    os.makedirs(config.ckpt_dir)
    with tf.Session(config=gpu_config) as sess:
        sess.run(tf.global_variables_initializer())
        if os.path.exists(config.ckpt_dir): #
            saver.restore(sess, tf.train.latest_checkpoint(save_path))
        for i in range(10000):
            # 2.feed data
            input_x = np.random.randn(config.batch_size, config.sequence_length)  # [None, self.sequence_length]
            input_x[input_x >= 0] = 1
            input_x[input_x < 0] = 0
            input_y = generate_label(input_x,threshold)
            p_mask_lm=[i for i in range(batch_size)]
            # 3.run session to train the model, print some logs.
            loss, _ = sess.run([model.loss_val,  model.train_op],feed_dict={model.x_mask_lm: input_x, model.y_mask_lm: input_y,model.p_mask_lm:p_mask_lm,
                                                                            model.dropout_keep_prob: config.dropout_keep_prob})
            print(i, "loss:", loss, "-------------------------------------------------------")
            if i==300:
                print("label[0]:", input_y[0]);print("input_x:",input_x)
            if i % 500 == 0:
                saver.save(sess, save_path, global_step=i)

# use saved checkpoint from model to make prediction, and print it, to see whether it is able to do toy task successfully.