susanli2016

fig = px.line(df_g7_melt, x="DATE", y="VALUE", color='COUNTRY', title='Real Residential Property Prices - All G7 Countries')
fig.update_layout(title=dict(x=0.5,y=0.95))
fig.show()

susanli2016

countries = {'US':'United States', 'GB':'United Kingdom', 'FR':'France', 'CA':'Canada', 'DE':'Germany', 'JP':'Japan', 'IT':'Italy'}

def check_country(x):
    for country in countries:
        if country.lower() in x.lower():
            return countries[country]
    return ''

df_g7_melt['COUNTRY'] = df_g7_melt['SERIES_ID'].map(lambda x: check_country(x))

susanli2016

g7_list = df_g7['series_id'].tolist()

start_date = '1970-01-01'
end_date = '2021-10-01'

df_g7 = get_fred_data(series_list = g7_list, 
                          start_date = start_date,
                          end_date = end_date)
                          
df_g7_melt = pd.melt(df_g7, id_vars = ['DATE'], value_vars = g7_list, var_name = 'SERIES_ID', value_name = 'VALUE')

susanli2016

df_g7 = df_all_series_ids.loc[(df_all_series_ids['region'] == 'Canada') | (df_all_series_ids['region'] == 'France') | (df_all_series_ids['region'] == 'Germany') | (df_all_series_ids['region'] == 'Italy') | (df_all_series_ids['region'] == 'Japan') | (df_all_series_ids['region'] == 'United Kingdom') | (df_all_series_ids['region'] == 'United States')]

df_g7

susanli2016

fred_api_key = 'your API key'

def get_fred_series_data(api_key, series):
    url = "https://api.stlouisfed.org/geofred/series/data?series_id={0}&api_key={1}&file_type=json".format(series, api_key)
    response = requests.request("GET", url)
    return response
def transform_series_response(response):
    latest_date = list(response.json()['meta']['data'].keys())[0]
    return pd.DataFrame(response.json()['meta']['data'][latest_date])
    

susanli2016

fig = px.line(df_canada, x="DATE", y='QCAR628BIS', title='Real Residential Property Prices for Canada')
fig.update_layout(title=dict(x=0.5,y=0.95))
fig.show();

susanli2016

import pandas_datareader as pdr
import pandas as pd
import requests
import plotly.express as px
from datetime import datetime

def get_fred_data(series_list, start_date, end_date):
    df = pdr.DataReader(series_list, 'fred', start_date, end_date)
    return df.reset_index()
    

susanli2016

grouped_df.loc[grouped_df['customer_id'] == 2].sort_values('Quantity', ascending=False)[['customer_id', 'Description', 'Quantity']].head(20)

susanli2016

def recommend(customer_id, sparse_customer_item, customer_vecs, item_vecs, num_items=10):
    
    customer_interactions = sparse_customer_item[customer_id,:].toarray()
    customer_interactions = customer_interactions.reshape(-1) + 1
    customer_interactions[customer_interactions > 1] = 0
    
    rec_vector = customer_vecs[customer_id,:].dot(item_vecs.T).toarray()
    
    min_max = MinMaxScaler()
    rec_vector_scaled = min_max.fit_transform(rec_vector.reshape(-1,1))[:,0]

susanli2016

item_id = 1319
n_similar = 10

item_vecs = model.item_factors
customer_vecs = model.user_factors

item_norms = np.sqrt((item_vecs * item_vecs).sum(axis=1))

scores = item_vecs.dot(item_vecs[item_id]) / item_norms
top_idx = np.argpartition(scores, -n_similar)[-n_similar:]