prrao87

def transform_stance(X1):
    # Input transform for classification task-head
    n_batch = len(X1)
    xmb = np.zeros((n_batch, 1, n_ctx, 2), dtype=np.int32)
    mmb = np.zeros((n_batch, 1, n_ctx), dtype=np.float32)
    start = encoder['_start_']
    for i, x1 in enumerate(X1):
        x12 = [start] + x1[:max_len] + [clf_token]
        l12 = len(x12)
        xmb[i, 0, :l12, 0] = x12

prrao87

def _stance(path, topic=None):
    def clean_ascii(text):
        # function to remove non-ASCII chars from data
        return ''.join(i for i in text if ord(i) < 128)
    orig = pd.read_csv(path, delimiter='\t', header=0, encoding = "latin-1")
    orig['Tweet'] = orig['Tweet'].apply(clean_ascii)
    df = orig
    # Get only those tweets that pertain to a single topic in the training data
    if topic is not None:
        df = df.loc[df['Target'] == topic]

prrao87

def stance(data_dir, topic=None):
    path = Path(data_dir)
    trainfile = 'semeval2016-task6-trainingdata.txt'
    testfile = 'SemEval2016-Task6-subtaskA-testdata.txt'

    X, Y = _stance(path/trainfile, topic=topic)
    teX, _ = _stance(path/testfile, topic=topic)
    tr_text, va_text, tr_sent, va_sent = train_test_split(X, Y, test_size=0.2, random_state=seed)
    trX = []
    trY = []

prrao87

# Load data
import pytreebank
import sys
import os

out_path = os.path.join(sys.path[0], 'sst_{}.txt')
dataset = pytreebank.load_sst('./raw_data')

# Store train, dev and test in separate files
for category in ['train', 'test', 'dev']:

prrao87

import pandas as pd
from sklearn.metrics import f1_score, accuracy_score


class Base:
    """Base class that houses common utilities for reading in test data
    and calculating model accuracy and F1 scores.
    """
    def __init__(self) -> None:
        pass

prrao87

class ExampleSentiment(Base):
    """Predict sentiment scores using using X classifier"""
    def __init__(self, model_file: str=None) -> None:
        super().__init__()   # Inherit methods from Base class

    def score(self, text: str) -> int:
        """Return a sentiment score on sample text, an integer in the range [1, 2, 3, 4, 5]"""
        # Apply some sentiment scoring technique here
        
    def predict(self, train_file: None, test_file: str, lower_case: bool) -> pd.DataFrame:

prrao87

class TextBlobSentiment(Base):
    """Predict fine-grained sentiment classes using TextBlob."""
    def __init__(self, model_file: str=None) -> None:
        super().__init__()

    def score(self, text: str) -> float:
        # pip install textblob
        from textblob import TextBlob
        return TextBlob(text).sentiment.polarity

prrao87

class VaderSentiment(Base):
    """Predict fine-grained sentiment classes using Vader."""
    def __init__(self, model_file: str=None) -> None:
        super().__init__()
        from nltk.sentiment.vader import SentimentIntensityAnalyzer
        self.vader = SentimentIntensityAnalyzer()

    def score(self, text: str) -> float:
        return self.vader.polarity_scores(text)['compound']

prrao87

class LogisticRegressionSentiment(Base):
    """Predict fine-grained sentiment scores using a sklearn Logistic Regression pipeline."""
    def __init__(self, model_file: str=None) -> None:
        super().__init__()
        from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer
        from sklearn.linear_model import LogisticRegression
        from sklearn.pipeline import Pipeline
        self.pipeline = Pipeline(
            [
                ('vect', CountVectorizer()),

prrao87

class SVMSentiment(Base):
    """Predict fine-grained sentiment scores using a sklearn 
    linear Support Vector Machine (SVM) pipeline."""
    def __init__(self, model_file: str=None) -> None:
        super().__init__()
        from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer
        from sklearn.linear_model import SGDClassifier
        from sklearn.pipeline import Pipeline
        self.pipeline = Pipeline(
            [