van-william

import json
from tulip_api import TulipAPI,TulipTable, CachedTulipTable
import pandas as pd
import numpy as np
from datetime import datetime
import os

def lambda_handler(event, context):
    # instance url, api key, and api secret are stored as environment variables
    instance_url = os.getenv('instance')

van-william

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd

fig, ax = plt.subplots(figsize=(15, 5))
sns.lineplot(ax=ax, data=df_chapters, hue='book', linewidth=3, x='chapter', y='weighted_score')
plt.xlabel("Chapter #")
plt.ylabel("Avg. Sentiment")
plt.title("Gospels Analysis - Avg. Sentiment by Chapter\n(Higher Values are Positive Sentiment; Lower Values are Negative Sentiment)")
plt.show(fig)

van-william

import pandas as pd
import numpy as np
from transformers import pipeline

def sentiment_score_weight(sentiment):
    label = sentiment['label']
    score = float(sentiment['score'])
    label_value = 1 if label == 'POSITIVE' else -1
    weighted_score = label_value*score
    return weighted_score

van-william

def book_split(text):
    book = text.split(' ')
    return ' '.join(book[0:len(book)-1])


def chapter_split(text):
    reference_list = text.split(' ')
    chapter_verse = reference_list[-1].split(':')
    return int(chapter_verse[0])

van-william

bible_url = 'https://bereanbible.com/bsb.txt'
with open('./data/bereanbible.txt', 'r') as f:
    text = f.read()
response = requests.get(bible_url)
raw_bible = response.text
bible_list = raw_bible.splitlines()
bible_array = np.array([item.split('\t') for item in bible_list])

def printable_string(text):
    filtered_string = ''.join(s for s in text if s in string.printable)

van-william

def sentiment_pipeline(text, model):
  sentiment = model(text)
  return pd.Series([sentiment[0]['label'], sentiment[0]['score']])

model = pipeline("sentiment-analysis")
df[['sentiment_label','sentiment_score']]=df['prompt'].apply(lambda x: sentiment_pipeline(text=x,model=model))

van-william

from sklearn.model_selection import train_test_split

def split_data(df: pd.DataFrame, parameters: Dict) -> Tuple:
    """Splits data into features and targets training and test sets.

    Args:
        data: Data containing features and target.
        parameters: Parameters defined in parameters/data_science.yml.
    Returns:
        Split data.

van-william

%%writefile exploratory_data_analysis.py

import pandas as pd
import numpy as np
import seaborn as sns

## Python Code Here

van-william

SELECT
    geolocation.lat,
    geolocation.lon,
    geolocation.alt,
    system.timestamp as "event_time"
    
INTO
    [sql]
FROM
    [iot]

van-william

SELECT
    TRY_CAST(decoded.payload.TempC_SHT AS float) AS InternalTemp_C,
    (TRY_CAST(decoded.payload.TempC_SHT AS float) * 9/5) + 32 AS InternalTemp_F,
    TRY_CAST(decoded.payload.Hum_SHT AS float) AS InternalHum,
    TRY_CAST(decoded.payload.TempC_DS AS float) AS ExternalTemp_C,
    (TRY_CAST(decoded.payload.TempC_DS AS float) * 9/5) + 32 AS ExternalTemp_F,
    TRY_CAST(EventProcessedUtcTime AS datetime) AS ProcessTimeUTC,
    TRY_CAST(EventEnqueuedUtcTime AS datetime) AS QueueTimeUTC,
    EventProcessedUtcTime AS  ProcessedTimeRaw,
    EventEnqueuedUtcTime AS EnqueuedTimeRaw,