Skip to content
Snippets Groups Projects
Commit 26ff7105 authored by maxvaisto's avatar maxvaisto
Browse files

Merge branch 'dash_plots' of... 

Merge branch 'dash_plots' of https://version.helsinki.fi/maxvaist/steam-data-project into dash_plots
parents fd2fea26 b21ab399
No related branches found
No related tags found
No related merge requests found
Hide whitespace changes
Inline Side-by-side
dash_plot_generation/dash_plot_generator.py
+ 120
− 53
View file @ 26ff7105
import math
import os
import pandas
from dash import dash, dcc, html, ctx, callback
from dash import dash, dcc, html, Output, Input
from collections import Counter
from dash_plot_generation.utils import split_companies, extract_unique_companies, convert_owners_to_limits, \
get_owner_means
DEVELOPER_DROPDOWN = "developer_dropdown"
DEV_TOP_GENRES_LABEL = "dev_top_genres"
DEV_CCU_LABEL = "dev_ccu"
DEV_GAME_COUNT_LABEL = "dev_game_count"
DEV_REV_PER_GAME_LABEL = "dev_rev_per_game"
DEV_REVENUE_LABEL = "dev_revenue"
csv_path = os.path.normpath(os.getcwd() + os.sep + os.pardir + os.sep + "api_exploration")
split_csv_path = os.path.join(csv_path, "file_segments")
df = None
APP = None
DASH_ASSETS_PATH = "dash_assets"
APP = dash.Dash(
name=__name__,
assets_folder=DASH_ASSETS_PATH,
external_stylesheets=['https://codepen.io/chriddyp/pen/bWLwgP.css']
)
def initialize_data():
global df
files = os.listdir(split_csv_path)
......@@ -15,10 +40,31 @@ def initialize_data():
for file in os.listdir(split_csv_path):
file_path = os.path.join(split_csv_path, file)
dataframe = pandas.concat([dataframe, pandas.read_csv(file_path)]) if dataframe is not None \
else pandas.read_csv(file_path)
else pandas.read_csv(file_path)
df = dataframe
@APP.callback([Output(DEV_REVENUE_LABEL, "children"),
Output(DEV_TOP_GENRES_LABEL, "children"),
Output(DEV_CCU_LABEL, "children"),
Output(DEV_GAME_COUNT_LABEL, "children"),
Output(DEV_REV_PER_GAME_LABEL, "children")],
inputs=[Input(DEVELOPER_DROPDOWN, "value")])
def update_dev_info(dev_name):
global df
mask = df.developer.apply(lambda x: dev_name in x if isinstance(x,str) else False)
dev_data = df[mask]
ccu = sum(dev_data["ccu"])
dev_data["owner_means"] = dev_data["owners"].apply(lambda x: get_owner_means(convert_owners_to_limits(x)))
dev_data["game_prices"] = dev_data["price"]
game_revenue = [owner_count*game_price for (owner_count, game_price) in
zip(dev_data["owner_means"],dev_data["game_prices"]) if
not (pandas.isna(game_price) or pandas.isna(owner_count))]
genre_totals = [genre for genre_list in dev_data["genres"] if isinstance(genre_list,str) for genre in genre_list]
genre_counts = Counter(genre_totals).keys()
return str(sum(game_revenue)), "hello2", str(ccu), str(dev_data.shape[0]), str(round(sum(game_revenue)/len(game_revenue))
def initialize_dash(host: str = "0.0.0.0", **kwargs):
"""
Runs the Dash server.
......@@ -32,73 +78,94 @@ def initialize_dash(host: str = "0.0.0.0", **kwargs):
global APP, df
unique_developers = [dev for dev_list in df["developer"].fillna("Valve").unique() for dev in dev_list.split(",")]
unique_publishers = df["developer"].fillna("Valve").unique()
unique_publishers = extract_unique_companies(df["publisher"].apply(lambda x: split_companies(x)))
unique_developers = extract_unique_companies(df["developer"].apply(lambda x: split_companies(x)))
# unique_publishers = ["Valve"]
# unique_developers = ["Valve"]
# This is ok but the results are still unsatisfactory due to the game company names containing a comma
# unique_publishers = [publisher for publisher_list in df["developer"].fillna("Valve").unique() for publisher
# unique_developers = [dev for dev_list in df["developer"].fillna("Valve").unique() for dev in dev_list.split(",")]
APP = dash.Dash(
name=__name__,
assets_folder=DASH_ASSETS_PATH,
external_stylesheets=['https://codepen.io/chriddyp/pen/bWLwgP.css']
)
APP.layout = html.Div(children=[
html.Nav(className="nav nav-pills", children=[
html.H6("STEAM-SAVVY", style={"margin-left": "30px", "width": "20%", "display": "inline-block"}),
html.A('About', className="nav-item nav-link btn", href='/apps/App1',
style={"margin-left": "300px"}),
html.A('Process paper', className="nav-item nav-link active btn", href='/apps/App2',
html.A('Technical report', className="nav-item nav-link active btn", href='/apps/App2',
style={"margin-left": "150px"})
],
style={"background-color": "rgb(30,30,30)", "color": "rgb(255,255,255)"}),
html.H6("TEXT"),
style={"background-color": "rgb(30,30,30)", "color": "rgb(255,255,255)"}),
html.Div(className="row", children=[
html.Div(children = [
dcc.Tabs(id="tabs_main_plots1", value="tab1", children=[
dcc.Tab(label="Genre performance", value="tab1"),
dcc.Tab(label="Game popularity", value="tab2"),
dcc.Tab(label="Company revenues", value="tab3"),
dcc.Tab(label="Market performance", value="tab4"),
],
style={'width': '80%', 'display': 'inline-block', "outline": "2px dotted black"}),
], style={'width': '48%', 'display': 'inline-block'}),
html.Div(children=[
dcc.Tabs(id="tabs_main_plots2", value="tab3", children=[
dcc.Tab(label="Developer infromation", value="Valve", children=[
dcc.Dropdown(id="developer_dropdown",
options=[{"label": developer, "value": developer} for developer in unique_developers]
)
]),
dcc.Tab(label="Publisher information", value="tab4", children= [
dcc.Dropdown(id="publisher_dropdown",
options=[{"label": publisher, "value": publisher} for publisher in unique_publishers]
),
])
],
style={'width': '48%', 'display': 'inline-block'}),
dcc.Tabs(id="tabs_main_plots1", value="tab1", children=[
dcc.Tab(label="Genre performance", value="tab1"),
dcc.Tab(label="Game popularity", value="tab2"),
dcc.Tab(label="Company revenues", value="tab3"),
dcc.Tab(label="Market performance", value="tab4"),
],
style={'width': '80%', 'display': 'inline-block', "outline": "2px dot'ted black"}),
], style={'width': '60%', 'display': 'inline-block'}),
html.Div(children=[
dcc.Tabs(id="tabs_main_plots2", value="tab3", children=[
dcc.Tab(label="Developer infromation", value="Valve", children=[
html.Div(children=[
dcc.Dropdown(id=DEVELOPER_DROPDOWN, value="Valve",
options=[{"label": developer, "value": developer} for developer in
unique_developers]
),
html.Div(children=[
html.H6("General statistics"),
html.Div( # Revenue
children=[html.P("Revenue"),
html.Div(children=[
html.P("Total game sale revenue"),
html.Small(id=DEV_REVENUE_LABEL, children="524.245.000€"),
html.P("Total game sale revenue per game"),
html.Small(id=DEV_REV_PER_GAME_LABEL, children="92.625.000€"),
html.P("Highest game sale revenue games:"),
html.Small("Half life 2"),
html.Small("CS GO"),
html.Small("Portal 2")
])
],
style={'width': '48%', 'display': 'inline-block'}
),
html.Div(children=[
html.Div(children=[
html.P("Number of games"),
html.Small(id=DEV_GAME_COUNT_LABEL, children="5"),
html.P("Number of concurrent users"),
html.Small(id=DEV_CCU_LABEL, children="92.625.000€"),
html.P("Most common game genres"),
html.Small(id=DEV_TOP_GENRES_LABEL, children="FPS, Action, Puzzle"),
html.P("Average game rating"),
])
], style={'width': '48%', 'display': 'inline-block'}
)
])
],
)
]),
dcc.Tab(label="Publisher information", value="tab4", children=[
dcc.Dropdown(id="publisher_dropdown", value="Valve",
options=[{"label": publisher, "value": publisher} for publisher in
unique_publishers]
),
])
],
style={'width': '100%', 'display': 'inline-block'}),
],
style={'width': '48%', 'display': 'inline-block', "outline": "2px dotted black"})
style={'width': '30%', 'display': 'inline-block', "outline": "2px dotted black"})
],
style={'width': '100%'}),
])
style={'width': '100%', "padding-top": "15px"}),
])
APP.run(host=host, **kwargs)
print("The server has closed!")
if __name__ == "__main__":
initialize_data()
initialize_dash()
initialize_dash(debug=True)
print(csv_path)
dash_plot_generation/utils.py
+ 26
− 1
View file @ 26ff7105
from typing import Sequence, Optional
from typing import Sequence, Optional, Any
import pandas
DEFAULT_ILLEGAL_CONTINUATIONS = {"INC.", "LLC", "CO.", "LTD.", "S.R.O."}
def get_owner_means(owner_limits: Sequence[Any]):
if not isinstance(owner_limits, list):
return owner_limits
else:
return (owner_limits[0]+owner_limits[1])/2
def convert_owners_to_limits(owner_limit):
if not isinstance(owner_limit, str):
return owner_limit
owners_raw = [rev.replace(" ", "") for rev in owner_limit.split(" .. ")]
owners_clean = []
for owner_limit in owners_raw:
owner_limit = owner_limit.replace("M", "0"*6)
owner_limit = owner_limit.replace("k", "0"*3)
owners_clean.append(int(owner_limit))
return owners_clean
def split_companies(arr, illegal_continuations: Optional[Sequence[str]] = None):
"""
Splits the given string at comma sign as long as following the comma none of the illegal
......@@ -44,3 +59,13 @@ def split_companies(arr, illegal_continuations: Optional[Sequence[str]] = None):
results_list.append(arr[start_index:index + 1].strip())
return results_list
def extract_unique_companies(nested_companies):
full_company_list = [dev for company_list in nested_companies
if isinstance(company_list, list) for dev in company_list]
unique_companies = []
for company in full_company_list:
if company not in unique_companies:
unique_companies.append(company)
return unique_companies
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment