Yeah, sure. Here is the layout of the code:
UPLOAD_DIRECTORY = "Excel_Files"
if not os.path.exists(UPLOAD_DIRECTORY):
os.makedirs(UPLOAD_DIRECTORY)
@server.route("/download/<path:path>")
def download(path):
"""Serve a file from the upload directory."""
return send_from_directory(UPLOAD_DIRECTORY, path, as_attachment=True)
#Uploading file definitions
def save_file(name, content):
"""Decode and store a file uploaded with Plotly Dash."""
data = content.encode("utf8").split(b";base64,")[1]
with open(os.path.join(UPLOAD_DIRECTORY, name), "wb") as fp:
fp.write(base64.decodebytes(data))
def uploaded_files():
"""List the files in the upload directory."""
files = []
for filename in os.listdir(UPLOAD_DIRECTORY):
path = os.path.join(UPLOAD_DIRECTORY, filename)
if os.path.isfile(path):
files.append(filename)
return files
def file_download_link(filename):
"""Create a Plotly Dash 'A' element that downloads a file from the app."""
location = "/download/{}".format(urlquote(filename))
return html.A(filename, href=location)
@app.callback(
Output("file-list", "children"),
[Input("upload-data", "filename"), Input("upload-data", "contents")],)
def update_output(uploaded_filenames, uploaded_file_contents):
"""Save uploaded files and regenerate the file list."""
if uploaded_filenames is not None and uploaded_file_contents is not None:
for name, data in zip(uploaded_filenames, uploaded_file_contents):
save_file(name, data)
files = uploaded_files()
if len(files) == 0:
return [html.Li("No files yet!")]
else:
return [html.Li(file_download_link(filename)) for filename in files]
#Callback for dropdown menu
@app.callback(dash.dependencies.Output('dd-output-container', 'children'),[dash.dependencies.Input('Categories', 'value')])
def update_output_dropdown(value):
return 'You have selected "{}"'.format(value)
def calculation():
#Read the uploaded file and calculate the category from dropdown
for file in uploaded_files():
with open(file) as f:
dfs = pd.read_excel(file)
xl = pd.ExcelFile(dfs)
if xl.parse(sheet_name= 'Spikes'):
# Define if we analyzing Premium or Cut Spikes
# Define if the data is reliable
dfs.loc[(dfs['Fastener_Presence'] !='X') & (dfs['Concrete_or_Timber'] == 'Wooden'), 'Timber_Fastener_Type'] = 'Premium'
dfs.loc[(dfs['Fastener_Presence'] =='X') & (dfs['Concrete_or_Timber'] == 'Wooden'), 'Timber_Fastener_Type'] = 'Cut Spikes'
dfs.loc[(dfs['Spike_Presence'] =='X') & (dfs['Fastener_Presence'] == 'X'), 'Timber_Fastener_Type'] = 'Error'
# dfs.to_excel("Test_Type.xlsx")
# Cleaning the data
dfs = dfs[dfs.Concrete_or_Timber != 'Concrete']
dfs = dfs[dfs.Timber_Fastener_Type != 'Error']
dfs = dfs.reset_index(drop=True)
dfs['Spike_Presence'] = dfs.Spike_Presence.replace('X',0)
dfs['Spike_Presence'] = dfs.Spike_Presence.replace('x',0)
dfs['Fastener_Presence'] = dfs.Fastener_Presence.replace('X',0)
dfs['Fastener_Presence'] = dfs.Fastener_Presence.replace('x',0)
# dfs.to_excel("Test_Type.xlsx")
# Make sure we are working with floats
dfs['Spike_Presence'] = dfs['Spike_Presence'].astype(float)
dfs['Fastener_Presence'] = dfs['Fastener_Presence'].astype(float)
#Function to calculate how many fasteners in 5 ties window
def g(column,i):
a = np.sum(column[i-2:i+3].to_list())
return a
# Create new dataframe to receive all the calculations from function g
dfsnew = pd.DataFrame(columns=['Spike_Presence_5_Ties','Fastener_Presence_5_Ties'])
# Passing function in the dataframe
for j in range(0, len(dfs)):
Spike_Presence_5_Ties = g(dfs['Spike_Presence'], i=j)
Fastener_Presence_5_Ties = g(dfs['Fastener_Presence'], i=j)
dfsnew = dfsnew.append({'Spike_Presence_5_Ties':Spike_Presence_5_Ties,
'Fastener_Presence_5_Ties': Fastener_Presence_5_Ties
}, ignore_index=True, verify_integrity=False, sort=False)
# dfsnew.to_excel("Test.xlsx")
dfs1 = pd.concat([dfs, dfsnew], axis=1)
# Rename dataframe dfs1
dfspike = dfs1
# Define rules. Data x means percentage to consider a bad spot. Data y means the value from 0 to 1 for each one of the percentages
# Cut Spikes Rules
datax = [0,5,10,15,20,25,30,35,38,40]
datay = [1,1,1,1,0.8,0.6,0.4,0.2,0.1,0]
# Define rules. Data x means percentage to consider a bad spot. Data y means the value from 0 to 1 for each one of the percentages
# Premium Rules
datax1 = [0,2,4,6,8,10,12,14,16,18,20]
datay1 = [1,1,1,0.8,0.6,0.5,0.4,0.35,0.3,0,0]
x1 = np.array(datax1)
y1=np.array(datay1)
z1 = np.polyfit(x1,y1,2)
print ('{0}x^2 + {1}x + {2}'.format(*z1))
x = np.array(datax)
y=np.array(datay)
z = np.polyfit(x,y,2)
print ('{0}x^2 + {1}x + {2}'.format(*z))
def print_graph_cut_spikes(S):
# Plotting the degradation function - Cut Spikes
S = lambda x: z[0]*x**2 + z[1]*x + z[2]
xs = np.linspace(0,40,240)
ys = S(xs)
#fig = plt.figure(figsize=(20,12))
fig4 = px.line(x = xs, y = ys, title= 'Cut Spikes Degradation Function')
return fig4
def print_graph_premium_spikes(S1):
# Plotting the degradation function - Premium E-Clips
S1 = lambda x: z1[0]*x**2 + z1[1]*x + z1[2]
xs = np.linspace(0,20,240)
ys = S1(xs)
fig5 = px.line(x = xs, y = ys, title= 'Cut Spikes Degradation Function')
return fig5
def FSpikes(x):
res = z[0]*(x)**2 + z[1]*(x) + z[2]
if res<0:
res=0
else:
res = res
if res>1:
res=1
return res
def FPremium (s,c): # s = number of spikes, c = number of e-clips
res_s = z[0]*(s)**2 + z[1]*(s) + z[2]
res_c = z1[0]*(c)**2 + z1[1]*(c) + z1[2]
resf = 0.4*res_s + 0.6*res_c #Spikes represent 40% of the strenght and e-clips 60%
if resf<0:
resf=0
else:
resf = resf
if resf>1:
resf=1
return resf
dfspike['Health'] = dfspike.apply(lambda row : FSpikes(row['Spike_Presence_5_Ties']) if row['Timber_Fastener_Type'] == 'Cut Spikes' else FPremium(row['Spike_Presence_5_Ties'],row['Fastener_Presence_5_Ties'] ) , result_type='expand', axis=1)
dfspike = dfspike.reset_index(drop=True)
DF1 = dfspike.drop(['Spike_Presence','Fastener_Presence','Timber_Fastener_Type','Spike_Presence_5_Ties','Fastener_Presence_5_Ties'], axis=1)
df2 = pd.read_excel (r'TQI_Data.xlsx', sheet_name='Fasteners')
# Cleaning the data
df2 = df2[df2.Concrete_or_Timber != 'Wooden']
df2 = df2.reset_index(drop=True)
df2['Fastener_Presence_Correct'] = df2.Fastener_Presence.replace('X',4)
#Function to calculate how many fasteners in 5 ties window
def f(column,i):
a = np.sum(column[i-2:i+3].to_list())
return a
# Create new dataframe to receive results from funtion f
df2new = pd.DataFrame(columns=['Fastener_Presence_5_Ties'])
df2['Fastener_Presence_Correct'] = df2['Fastener_Presence_Correct'].astype(float)
# Passing function in the dataframe
for j in range(0, len(df2)):
Fastener_Presence_5_Ties = f(df2['Fastener_Presence_Correct'], i=j)
df2new = df2new.append({'Fastener_Presence_5_Ties':Fastener_Presence_5_Ties
}, ignore_index=True, verify_integrity=False, sort=False)
# Concatanate the two dataframes
df2 = pd.concat([df2, df2new], axis=1)
# Define rules. Data x means percentage to consider a bad spot. Data y means the value from 0 to 1 for each one of the percentages
# Concrete Ties
datax2 = [0,2,4,6,8,10,12,14,16,18,20]
datay2 = [1,1,1,0.6,0.4,0.35,0.3,0.25,0.1,0,0]
x2 = np.array(datax2)
y2=np.array(datay2)
z2 = np.polyfit(x2,y2,2)
print ('{0}x^2 + {1}x + {2}'.format(*z2))
def concrete_graph(H):
# Plotting the degradation function
H = lambda x: z2[0]*x**2 + z2[1]*x + z2[2]
xs = np.linspace(0,20,10)
ys = H(xs)
fig6 = px.line(x = xs, y = ys, title= 'Concrete Ties')
return fig6
#Defining the function that pass the measured value, divide it by the design height (resulting the porcentage) and return the track health based on the created rule
def FConcrete(x):
resconc = z2[0]*(x)**2 + z2[1]*(x) + z2[2]
if resconc<0:
resconc=0
if resconc>1:
resconc=1
else:
resconc = resconc
return resconc
# Calculate Health Index for Concrete
df2['Health'] = df2.apply(lambda row : FConcrete(row['Fastener_Presence_5_Ties']), result_type='expand', axis=1)
DF2 = df2.drop(['Fastener_Presence','Fastener_Presence_Correct','Fastener_Presence_5_Ties'], axis=1)
DF2
# Merge DF1 and DF2.
# This will result the totality of the loop Fasteners Health Index.
DF = pd.concat([DF1, DF2],axis=0)
DF = DF.sort_values(by=['Stationing'],axis=0)
DF = DF.reset_index(drop=True)
# Gambiarra do carai pra n corrigir entre stationings 3810.368 e 4525.509
df_gambiarra = pd.DataFrame({'Stationing':[4525],'Concrete_or_Timber':['Wooden'],'Health':[0]})
DF = pd.concat([DF, df_gambiarra],axis=0)
DF = DF.sort_values(by=['Stationing'],axis=0)
DF = DF.reset_index(drop=True)
print(DF)
#DF.to_excel("Test.xlsx")
def Health_Graph(DF):
ax_DF = px.line(DF, x = 'Stationing', y = 'Health', title= 'Fastener Health Index')
return ax_DF
Health_Graph(DF)
#Callback for submitting everything.
@app.callback(Output('target', 'children'), Input('submit', 'n_clicks'))
def callback(n_clicks):
if (n_clicks != 0):
#xl = pd.ExcelFile(dfs)
calculation()
#if xl.parse(sheet_name= 'Spikes'):
app.layout = html.Div(
children=[
html.H1(children='RailTec Dashboard'),
#Uploading the files to the board. We want to make sure that two files can be uploaded, read, and grab the same categories to the direct coorelated category.
#For choosing files between tabs
html.Div(
[
html.H2("Upload"),
dcc.Upload(
id="upload-data",
children=html.Div(
["Drag and drop or click to select a file to upload."]
),
style={
"width": "100%",
"height": "60px",
"lineHeight": "60px",
"borderWidth": "1px",
"borderStyle": "dashed",
"borderRadius": "5px",
"textAlign": "center",
"margin": "10px",
},
multiple=True,
),
html.H2("File List"),
html.Ul(id="file-list"),
],
style={"max-width": "500px"},
),
#Drop Down Menu for the categories.
dcc.Dropdown(
id = 'Categories',
options=[
{'label': 'Spikes', 'value': 'Spikes'},
{'label': 'Fasteners', 'value': 'Fasteners'},
{'label': 'Ballast Level', 'value': 'Ballast Level'},
{'label': 'Spikes Med', 'value': 'Spikes'},
{'label': 'Fasteners Med', 'value': 'Fasteners'},
{'label': 'Ballast Level Med', 'value': 'Ballast Level'},
{'label': 'Spikes Bad', 'value': 'Spikes'},
{'label': 'Fasteners Bad', 'value': 'Fasteners'},
{'label': 'Ballast Level Bad', 'value': 'Ballast Level'}
],
placeholder="Select a category",
value = "Spikes"
),
html.Div(id='dd-output-container', children = []),
html.Form([
#Enter button Event to trigger app callback for dropdown and upload data to submit what to calculate.
html.Div(id='target'),
html.Button('Submit',id='submit', n_clicks=0),
]),
dcc.Graph(
id='Concrete_Health_method',
figure= calculation(),
#calculator_1.graph_concrete(df2)
style ={'display': 'inline-block'})
])
