Valuations API Examples¶

Run this example in Colab¶

Setup¶

Install the Signal Ocean SDK:

pip install signal-ocean

Set your subscription key acquired here: https://apis.signalocean.com/profile

In [ ]:

pip install signal-ocean

pip install signal-ocean

In [1]:

signal_ocean_api_key = '' #replace with your subscription key

signal_ocean_api_key = '' #replace with your subscription key

Use Case 1: Average valuation per vessel class.¶

In this Use Case we will use Valuations API to get latest valuations for tanker vessels and visualize them by vessel class for vessel classes 84, 85, 86, 87.

Import helpful modules¶

In [2]:

from datetime import date, timedelta
from tqdm import tqdm

import requests
import json
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
from matplotlib.collections import PolyCollection
import numpy as np
sns.set_theme()
sns.set_style("whitegrid")

from datetime import date, timedelta from tqdm import tqdm import requests import json import pandas as pd import seaborn as sns import matplotlib.pyplot as plt from matplotlib.collections import PolyCollection import numpy as np sns.set_theme() sns.set_style("whitegrid")

Get imo numbers for the vessel classes of interest using Vessel API.¶

Initialize Vessels API¶

In [3]:

from signal_ocean import Connection
from signal_ocean.vessels import VesselsAPI

connection = Connection(signal_ocean_api_key)
vessels_api = VesselsAPI(connection)

from signal_ocean import Connection from signal_ocean.vessels import VesselsAPI connection = Connection(signal_ocean_api_key) vessels_api = VesselsAPI(connection)

Get list of imo numbers per vessel class¶

In [4]:

vessel_classes = [84, 85, 86, 87]

vessel_class_vessels = {
    vessel_class: [v.imo for v in vessels_api.get_vessels_by_vessel_class(vessel_class)] 
    for vessel_class in vessel_classes
}

vessel_classes = [84, 85, 86, 87] vessel_class_vessels = { vessel_class: [v.imo for v in vessels_api.get_vessels_by_vessel_class(vessel_class)] for vessel_class in vessel_classes }

Initialize Vessels API¶

In [5]:

from signal_ocean.vessel_valuations import VesselValuationsAPI

valuations_api = VesselValuationsAPI(connection)

from signal_ocean.vessel_valuations import VesselValuationsAPI valuations_api = VesselValuationsAPI(connection)

Create a list of Vessel Valuations calling Valuations API¶

In [6]:

vessel_valuations_list = []

for vc, imo_list in vessel_class_vessels.items():
    # Get list of valuation objects for these vessel class
    vessel_class_valuations = valuations_api.get_latest_valuations_for_list_of_vessels(imo_list)
    # Cast vessel valuations to dict objects and add the vessel class key
    vessel_class_valuations = [dict(valuation.to_dict(), **{'vesselClass':vc}) for valuation in vessel_class_valuations]
    # Update the list of all vessels
    vessel_valuations_list.extend(vessel_class_valuations)

valuations_df = pd.DataFrame.from_records(vessel_valuations_list)

vessel_valuations_list = [] for vc, imo_list in vessel_class_vessels.items(): # Get list of valuation objects for these vessel class vessel_class_valuations = valuations_api.get_latest_valuations_for_list_of_vessels(imo_list) # Cast vessel valuations to dict objects and add the vessel class key vessel_class_valuations = [dict(valuation.to_dict(), **{'vesselClass':vc}) for valuation in vessel_class_valuations] # Update the list of all vessels vessel_valuations_list.extend(vessel_class_valuations) valuations_df = pd.DataFrame.from_records(vessel_valuations_list)

Visuals¶

Create box plots for valuations per vessel class¶

In [7]:

def set_violins_transparency(violins, alpha):
  for violin in violins.get_children():
    if isinstance(violin, PolyCollection):
      violin.set_alpha(alpha)

def set_violins_transparency(violins, alpha): for violin in violins.get_children(): if isinstance(violin, PolyCollection): violin.set_alpha(alpha)

In [8]:

plt.figure(figsize=(16,8))
valuations_df.sort_values('vesselClass', inplace=True)

violins = sns.violinplot(x='vesselClass', y='valuationPrice', data=valuations_df, color=".0")

set_violins_transparency(violins, 0.3)

sns.boxplot(x='vesselClass', y='valuationPrice', data=valuations_df)
sns.stripplot(x='vesselClass', y='valuationPrice', data=valuations_df, color=".25")

plt.ylabel("Valuation in million $", fontsize=14)
plt.xlabel("Vessel Class", fontsize=14)
plt.title("Valuation per Vessel Class", fontsize=18)
plt.show()

plt.figure(figsize=(16,8)) valuations_df.sort_values('vesselClass', inplace=True) violins = sns.violinplot(x='vesselClass', y='valuationPrice', data=valuations_df, color=".0") set_violins_transparency(violins, 0.3) sns.boxplot(x='vesselClass', y='valuationPrice', data=valuations_df) sns.stripplot(x='vesselClass', y='valuationPrice', data=valuations_df, color=".25") plt.ylabel("Valuation in million $", fontsize=14) plt.xlabel("Vessel Class", fontsize=14) plt.title("Valuation per Vessel Class", fontsize=18) plt.show()

Use Case 2: Historical and Current Vessel Valuation.¶

Get historical valuations for 2 tanker vessels and store to Dataframes¶

In [9]:

vessel_1 = 9453315
valuation_history = valuations_api.get_all_historical_valuations_by_imo(vessel_1)
vessel_1_df = pd.DataFrame.from_records([valuation.to_dict() for valuation in valuation_history])

vessel_2 = 9318084
valuation_history = valuations_api.get_all_historical_valuations_by_imo(vessel_2)
vessel_2_df = pd.DataFrame.from_records([valuation.to_dict() for valuation in valuation_history])

vessel_1 = 9453315 valuation_history = valuations_api.get_all_historical_valuations_by_imo(vessel_1) vessel_1_df = pd.DataFrame.from_records([valuation.to_dict() for valuation in valuation_history]) vessel_2 = 9318084 valuation_history = valuations_api.get_all_historical_valuations_by_imo(vessel_2) vessel_2_df = pd.DataFrame.from_records([valuation.to_dict() for valuation in valuation_history])

Feature Engineering¶

Convert string date column to datetime and save oldest and latest date, valuation and scrap valuation for the 2 vessels

In [10]:

vessel_1_df['valueFrom'] = pd.to_datetime(vessel_1_df['valueFrom'])
vessel_1_oldest_date, vessel_1_latest_date = vessel_1_df['valueFrom'].min(), vessel_1_df['valueFrom'].max()

vessel_1_oldest_valuation = vessel_1_df[vessel_1_df['valueFrom']==vessel_1_oldest_date]['valuationPrice'].iloc[0]
vessel_1_latest_valuation = vessel_1_df[vessel_1_df['valueFrom']==vessel_1_latest_date]['valuationPrice'].iloc[0]

vessel_1_oldest_scrap = vessel_1_df[vessel_1_df['valueFrom']==vessel_1_oldest_date]['scrapPrice'].iloc[0]
vessel_1_latest_scrap = vessel_1_df[vessel_1_df['valueFrom']==vessel_1_latest_date]['scrapPrice'].iloc[0]


vessel_2_df['valueFrom'] = pd.to_datetime(vessel_2_df['valueFrom'])
vessel_2_oldest_date, vessel_2_latest_date = vessel_2_df['valueFrom'].min(), vessel_2_df['valueFrom'].max()

vessel_2_oldest_valuation = vessel_2_df[vessel_2_df['valueFrom']==vessel_2_oldest_date]['valuationPrice'].iloc[0]
vessel_2_latest_valuation = vessel_2_df[vessel_2_df['valueFrom']==vessel_2_latest_date]['valuationPrice'].iloc[0]

vessel_2_oldest_scrap = vessel_2_df[vessel_2_df['valueFrom']==vessel_2_oldest_date]['scrapPrice'].iloc[0]
vessel_2_latest_scrap = vessel_2_df[vessel_2_df['valueFrom']==vessel_2_latest_date]['scrapPrice'].iloc[0]

vessel_1_df['valueFrom'] = pd.to_datetime(vessel_1_df['valueFrom']) vessel_1_oldest_date, vessel_1_latest_date = vessel_1_df['valueFrom'].min(), vessel_1_df['valueFrom'].max() vessel_1_oldest_valuation = vessel_1_df[vessel_1_df['valueFrom']==vessel_1_oldest_date]['valuationPrice'].iloc[0] vessel_1_latest_valuation = vessel_1_df[vessel_1_df['valueFrom']==vessel_1_latest_date]['valuationPrice'].iloc[0] vessel_1_oldest_scrap = vessel_1_df[vessel_1_df['valueFrom']==vessel_1_oldest_date]['scrapPrice'].iloc[0] vessel_1_latest_scrap = vessel_1_df[vessel_1_df['valueFrom']==vessel_1_latest_date]['scrapPrice'].iloc[0] vessel_2_df['valueFrom'] = pd.to_datetime(vessel_2_df['valueFrom']) vessel_2_oldest_date, vessel_2_latest_date = vessel_2_df['valueFrom'].min(), vessel_2_df['valueFrom'].max() vessel_2_oldest_valuation = vessel_2_df[vessel_2_df['valueFrom']==vessel_2_oldest_date]['valuationPrice'].iloc[0] vessel_2_latest_valuation = vessel_2_df[vessel_2_df['valueFrom']==vessel_2_latest_date]['valuationPrice'].iloc[0] vessel_2_oldest_scrap = vessel_2_df[vessel_2_df['valueFrom']==vessel_2_oldest_date]['scrapPrice'].iloc[0] vessel_2_latest_scrap = vessel_2_df[vessel_2_df['valueFrom']==vessel_2_latest_date]['scrapPrice'].iloc[0]

Visuals¶

Plot valuations time series.¶

In [11]:

plt.figure(figsize=(16,8))

sns.lineplot(y='valuationPrice', x='valueFrom', data=vessel_1_df, label=f'Vessel Valuation: {vessel_1}', color='b')
plt.scatter(x=vessel_1_oldest_date, y=vessel_1_oldest_valuation, color='b')
plt.scatter(x=vessel_1_latest_date, y=vessel_1_latest_valuation, color='b')
plt.annotate(str(round(vessel_1_oldest_valuation, 3)), xy=(vessel_1_oldest_date, vessel_1_oldest_valuation+1),weight='bold', fontsize=10)
plt.annotate(str(round(vessel_1_latest_valuation, 3)), xy=(vessel_1_latest_date, vessel_1_latest_valuation+1),weight='bold', fontsize=10)

sns.lineplot(y='scrapPrice', x='valueFrom', data=vessel_1_df, label=f'Scrap Valuation: {vessel_1}', color='orange')
plt.scatter(x=vessel_1_oldest_date, y=vessel_1_oldest_scrap, color='orange')
plt.scatter(x=vessel_1_latest_date, y=vessel_1_latest_scrap, color='orange')
plt.annotate(str(round(vessel_1_oldest_scrap, 3)), xy=(vessel_1_oldest_date, vessel_1_oldest_scrap+1),weight='bold', fontsize=10)
plt.annotate(str(round(vessel_1_latest_scrap, 3)), xy=(vessel_1_latest_date, vessel_1_latest_scrap+1),weight='bold', fontsize=10)

sns.lineplot(y='valuationPrice', x='valueFrom', data=vessel_2_df, label=f'Vessel Valuation: {vessel_2}', color='g')
plt.scatter(x=vessel_2_oldest_date, y=vessel_2_oldest_valuation, color='g')
plt.scatter(x=vessel_2_latest_date, y=vessel_2_latest_valuation, color='g')
plt.annotate(str(round(vessel_2_oldest_valuation, 3)), xy=(vessel_2_oldest_date, vessel_2_oldest_valuation+1),weight='bold', fontsize=10)
plt.annotate(str(round(vessel_2_latest_valuation, 3)), xy=(vessel_2_latest_date, vessel_2_latest_valuation+1),weight='bold', fontsize=10)

sns.lineplot(y='scrapPrice', x='valueFrom', data=vessel_2_df, label=f'Scrap Valuation: {vessel_2}', color='r')
plt.scatter(x=vessel_1_oldest_date, y=vessel_2_oldest_scrap, color='r')
plt.scatter(x=vessel_1_latest_date, y=vessel_2_latest_scrap, color='r')
plt.annotate(str(round(vessel_2_oldest_scrap, 3)), xy=(vessel_2_oldest_date, vessel_2_oldest_scrap-2),weight='bold', fontsize=10)
plt.annotate(str(round(vessel_2_latest_scrap, 3)), xy=(vessel_2_latest_date, vessel_2_latest_scrap-2),weight='bold', fontsize=10)


plt.ylabel("Valuation in million $", fontsize=14)
plt.xlabel("Valuation Date", fontsize=14)
plt.show()

plt.figure(figsize=(16,8)) sns.lineplot(y='valuationPrice', x='valueFrom', data=vessel_1_df, label=f'Vessel Valuation: {vessel_1}', color='b') plt.scatter(x=vessel_1_oldest_date, y=vessel_1_oldest_valuation, color='b') plt.scatter(x=vessel_1_latest_date, y=vessel_1_latest_valuation, color='b') plt.annotate(str(round(vessel_1_oldest_valuation, 3)), xy=(vessel_1_oldest_date, vessel_1_oldest_valuation+1),weight='bold', fontsize=10) plt.annotate(str(round(vessel_1_latest_valuation, 3)), xy=(vessel_1_latest_date, vessel_1_latest_valuation+1),weight='bold', fontsize=10) sns.lineplot(y='scrapPrice', x='valueFrom', data=vessel_1_df, label=f'Scrap Valuation: {vessel_1}', color='orange') plt.scatter(x=vessel_1_oldest_date, y=vessel_1_oldest_scrap, color='orange') plt.scatter(x=vessel_1_latest_date, y=vessel_1_latest_scrap, color='orange') plt.annotate(str(round(vessel_1_oldest_scrap, 3)), xy=(vessel_1_oldest_date, vessel_1_oldest_scrap+1),weight='bold', fontsize=10) plt.annotate(str(round(vessel_1_latest_scrap, 3)), xy=(vessel_1_latest_date, vessel_1_latest_scrap+1),weight='bold', fontsize=10) sns.lineplot(y='valuationPrice', x='valueFrom', data=vessel_2_df, label=f'Vessel Valuation: {vessel_2}', color='g') plt.scatter(x=vessel_2_oldest_date, y=vessel_2_oldest_valuation, color='g') plt.scatter(x=vessel_2_latest_date, y=vessel_2_latest_valuation, color='g') plt.annotate(str(round(vessel_2_oldest_valuation, 3)), xy=(vessel_2_oldest_date, vessel_2_oldest_valuation+1),weight='bold', fontsize=10) plt.annotate(str(round(vessel_2_latest_valuation, 3)), xy=(vessel_2_latest_date, vessel_2_latest_valuation+1),weight='bold', fontsize=10) sns.lineplot(y='scrapPrice', x='valueFrom', data=vessel_2_df, label=f'Scrap Valuation: {vessel_2}', color='r') plt.scatter(x=vessel_1_oldest_date, y=vessel_2_oldest_scrap, color='r') plt.scatter(x=vessel_1_latest_date, y=vessel_2_latest_scrap, color='r') plt.annotate(str(round(vessel_2_oldest_scrap, 3)), xy=(vessel_2_oldest_date, vessel_2_oldest_scrap-2),weight='bold', fontsize=10) plt.annotate(str(round(vessel_2_latest_scrap, 3)), xy=(vessel_2_latest_date, vessel_2_latest_scrap-2),weight='bold', fontsize=10) plt.ylabel("Valuation in million $", fontsize=14) plt.xlabel("Valuation Date", fontsize=14) plt.show()