Skip to content
GitLab
Commit ec1ae277 authored by Saad Jbabdi's avatar Saad Jbabdi
Browse files

prettify report

parent 72a269dd
Hide whitespace changes
Inline Side-by-side
fsl_mrs/utils/plotting.py
View file @ ec1ae277
......@@ -453,8 +453,8 @@ def plotly_fit(mrs,res,ppmlim=None,proj='real',metabs = None,phs=(0,0)):
preds *= phaseTerm
resid *= phaseTerm
base = project(base,proj)
data = project(data,proj)
base = project(base,proj)
data = project(data,proj)
preds = project(preds,proj)
resid = project(resid,proj)
......@@ -479,15 +479,12 @@ def plotly_fit(mrs,res,ppmlim=None,proj='real',metabs = None,phs=(0,0)):
concstr = f'/{res.concScalings["internalRef"]}'
df[concstr] = np.round(res.getConc(scaling='internal'),decimals=2)
fig = ff.create_table(df, height_constant=50)
for i in range(len(fig.layout.annotations)):
fig.layout.annotations[i].font.size = 16
tab = create_table(df)
colors = dict(data='rgb(67,67,67)',
pred='rgb(253,59,59)',
base='rgb(170,170,170)',
base='rgb(0,150,242)',
resid='rgb(170,170,170)')
line_size = dict(data=1,
pred=2,
......@@ -497,45 +494,50 @@ def plotly_fit(mrs,res,ppmlim=None,proj='real',metabs = None,phs=(0,0)):
mode='lines',
name='data',
line=dict(color=colors['data'],width=line_size['data']),
xaxis='x2', yaxis='y2')
)
trace2 = go.Scatter(x=axis, y=preds,
mode='lines',
name='model',
line=dict(color=colors['pred'],width=line_size['pred']),
xaxis='x2', yaxis='y2')
)
trace3 = go.Scatter(x=axis, y=base,
mode='lines',
name='baseline',
line=dict(color=colors['base'],width=line_size['base']),
xaxis='x2', yaxis='y2')
)
trace4 = go.Scatter(x=axis, y=resid,
mode='lines',
name='residuals',
line=dict(color=colors['resid'],width=line_size['resid']),
xaxis='x2', yaxis='y2')
)
fig.add_traces([trace1,trace2,trace3,trace4])
fig['layout']['xaxis2'] = {}
fig['layout']['yaxis2'] = {}
fig = make_subplots(rows=1, cols=2,
column_widths=[0.4, 0.6],
horizontal_spacing=0.03,
specs=[[{'type':'table'},{'type':'scatter'}]])
fig.add_trace(tab,row=1,col=1)
fig.add_trace(trace1,row=1,col=2)
fig.add_trace(trace2,row=1,col=2)
fig.add_trace(trace3,row=1,col=2)
fig.add_trace(trace4,row=1,col=2)
fig.layout.xaxis.update({'domain': [0, .35]})
fig.layout.xaxis2.update({'domain': [0.4, 1.]})
fig.layout.xaxis2.update(title_text='Chemical shift (ppm)',
tick0=2, dtick=.5,
range=[ppmlim[1],ppmlim[0]])
# The graph's yaxis MUST BE anchored to the graph's xaxis
fig.layout.yaxis2.update({'anchor': 'x2'})
fig.layout.yaxis2.update(zeroline=True,
zerolinewidth=1,
zerolinecolor='Gray',
showgrid=False,showticklabels=False,
range=[ymin,ymax])
# Update the margins to add a title and see graph x-labels.
fig.layout.margin.update({'t':50, 'b':100})
fig.layout.update({'title': 'Fitting summary'})
fig.update_layout(template = 'plotly_white')
fig.update_xaxes({'domain': [0.4, 1.]},row=1,col=2)
fig.update_xaxes(title_text='Chemical shift (ppm)',
tick0=2, dtick=.5,
range=[ppmlim[1],ppmlim[0]])
fig.update_yaxes(zeroline=True,
zerolinewidth=1,
zerolinecolor='Gray',
showgrid=False,showticklabels=False,
range=[ymin,ymax])
fig.layout.update({'height':800})
return fig
......@@ -567,14 +569,15 @@ def plot_dist_approx(res,refname='Cr'):
showlegend=False,
font=dict(size=10),
title='Approximate marginal distributions (ref={})'.format(refname),
height=800,width=800)
height=700,width=700)
for i in fig['layout']['annotations']:
i['font'] = dict(size=10,color='#ff0000')
fig.update_layout(autosize=True)
return fig
def plot_mcmc_corr(res,corr=None):
def plot_corr(res,corr=None,title='Correlation'):
#Greys,YlGnBu,Greens,YlOrRd,Bluered,RdBu,Reds,Blues,
#Picnic,Rainbow,Portland,Jet,Hot,Blackbody,Earth,
......@@ -591,9 +594,9 @@ def plot_mcmc_corr(res,corr=None):
fig.update_layout(template = 'plotly_white',
font=dict(size=10),
title='MCMC correlation',
width = 700,
height = 700,
title=title,
width = 600,
height = 600,
yaxis = dict(
scaleanchor = "x",
scaleratio = 1,
......@@ -609,7 +612,7 @@ def plot_mcmc_corr(res,corr=None):
method="restyle"
),
dict(
args=[{"z":[corrabs],"colorscale":'Reds'}],
args=[{"z":[corrabs],"colorscale":'Picnic'}],
label="Abs",
method="restyle"
)
......@@ -622,7 +625,8 @@ def plot_mcmc_corr(res,corr=None):
yanchor="top"
),
])
fig.update_layout(autosize=True)
return fig
def plot_dist_mcmc(res,refname='Cr'):
......@@ -656,7 +660,8 @@ def plot_dist_mcmc(res,refname='Cr'):
for i in fig['layout']['annotations']:
i['font'] = dict(size=10,color='#ff0000')
fig.update_layout(autosize=True)
return fig
def plot_real_imag(mrs,res,ppmlim=(.2,4.2)):
......@@ -743,7 +748,7 @@ def plot_real_imag(mrs,res,ppmlim=(.2,4.2)):
# Update the margins to add a title and see graph x-labels.
# fig.layout.margin.update({'t':50, 'b':100})
fig.layout.update({'title': 'Fitting summary Real/Imag'})
#fig.layout.update({'title': 'Fitting summary Real/Imag'})
fig.update_layout(template = 'plotly_white')
# fig.layout.update({'height':800,'width':1000})
......@@ -818,7 +823,7 @@ def plot_indiv_stacked(mrs,res,ppmlim=(.2,4.2)):
# Update the margins to add a title and see graph x-labels.
# fig.layout.margin.update({'t':50, 'b':100})
fig.layout.update({'title': 'Individual Fitting summary'})
#fig.layout.update({'title': 'Individual Fitting summary'})
fig.update_layout(template = 'plotly_white')
# fig.layout.update({'height':800,'width':1000})
......@@ -874,70 +879,169 @@ def plot_indiv(mrs,res,ppmlim=(.2,4.2)):
showgrid=False,showticklabels=False)
return fig
# def plot_table_fitparams(res):
def plot_table_fitparams(res):
# Single parameter measures
header=["Static phase (deg)", "Linear phase (deg/ppm)"]
p0,p1 = res.getPhaseParams(phi0='degrees',phi1='deg_per_ppm')
values=[np.round(p0,decimals=5),
np.round(p1,decimals=5)]
table1 = dict(type='table',
header=dict(values=header,font=dict(size=10),align="center"),
cells=dict(values=values,align = "right"),
columnorder = [1,2,3],
columnwidth = [80,80,80],
)
# Fitting group Gamma/Eps
header2 = ['group','linewidth (Hz)','shift (ppm)','metab groups']
values2 = [[],[],[],[]]
def create_table(df):
"""
Generate plotly graphical Table from pandas dataframe
"""
n = df.shape[0]
colors = ['#F2F2F2','white']*n # alternating row colors
header = dict(values=['<b>'+x+'</b>' for x in list(df.columns)],
fill_color='#41476C',
align='left',
font={'color':'white'})
cells = dict(values=[df[x] for x in list(df.columns)],
fill_color=[colors*2],
align='left')
tab = go.Table(header=header,
cells=cells,visible=True)
return tab
def plot_table_lineshape_phase(res):
shift = res.getShiftParams(units='ppm')
lw = res.getLineShapeParams(units='Hz')[0] # Only take combined values
for g in range(res.g):
values2[0].append(g)
values2[1].append(np.round(lw[g],decimals=3))
values2[2].append(np.round(shift[g],decimals=5))
lw = res.getLineShapeParams(units='Hz')[0] # Only take combined values
# Get the lineshape params
header = ['Metab group','linewidth (Hz)','shift (ppm)']
values = [[],[],[]]
for g in range(res.g):
values[1].append(np.round(lw[g],decimals=3))
values[2].append(np.round(shift[g],decimals=5))
metabs = []
for i,m in enumerate(res.original_metabs):
if res.metab_groups[i] == g:
metabs.append(m)
values2[3].append(metabs)
table2 = go.Table(header=dict(values=header2,font=dict(size=10),align="center"),
cells=dict(values=values2,align = "right"),
columnorder = [1,2,3,4],
columnwidth = [12,12,12,64]
)
fig = make_subplots(rows=1,cols=2,
specs= [[{"type": "table"},{"type": "table"}]])
fig.add_trace(table1,row=1, col=1)
fig.add_trace(table2,row=1, col=2)
values[0].append(', '.join(metabs))
# Fill dataframe
df1 = pd.DataFrame()
for h,v in zip(header,values):
df1[h] = v
# create table
tab1 = create_table(df1)
# Get phase params
p0,p1 = res.getPhaseParams(phi0='degrees',phi1='deg_per_ppm')
# Fill dataframe
df = pd.DataFrame()
df['Static phase (deg)'] = [np.round(p0,decimals=5)]
df['Linear phase (deg/ppm)'] = [np.round(p1,decimals=5)]
# create table
tab2 = create_table(df)
fig = make_subplots(rows=1, cols=2,
column_widths=[3/5, 2/5],
horizontal_spacing=0.03,
specs=[[{'type':'table'},{'type':'table'}]],
subplot_titles=['Lineshape params','Phase params'])
fig.add_trace(tab1,row=1,col=1)
fig.add_trace(tab2,row=1,col=2)
fig.update_layout(template = 'plotly_white')
fig.layout.update({'height':300})
fig.update_layout(autosize=True,margin=dict(l=0,r=0,t=0,b=0))
return fig
def plot_table_lineshape(res):
"""
Creates a table summarising fitted lineshape parameters for each metab group
"""
shift = res.getShiftParams(units='ppm')
lw = res.getLineShapeParams(units='Hz')[0] # Only take combined values
df = pd.DataFrame()
header = ['Metab group','linewidth (Hz)','shift (ppm)']
values = [[],[],[]]
for g in range(res.g):
values[1].append(np.round(lw[g],decimals=3))
values[2].append(np.round(shift[g],decimals=5))
metabs = []
for i,m in enumerate(res.original_metabs):
if res.metab_groups[i] == g:
metabs.append(m)
values[0].append(', '.join(metabs))
for h,v in zip(header,values):
df[h] = v
tab = create_table(df)
fig = go.Figure(data=[tab])
fig.update_layout(autosize=True,margin=dict(l=0,r=0,t=0,b=0))
return fig
# fig = ff.create_table(df, height_constant=50)
# for i in range(len(fig.layout.annotations)):
# fig.layout.annotations[i].font.size = 16
# fig.layout.update(title={'text': 'Lineshape parameters'},font={'size':20})
# fig.update_layout(template = 'plotly_white')
# return fig
def plot_table_phase(res):
"""
Creates a table summarising the fitted phase parameters
"""
p0,p1 = res.getPhaseParams(phi0='degrees',phi1='deg_per_ppm')
df = pd.DataFrame()
df['Static phase (deg)'] = [np.round(p0,decimals=5)]
df['Linear phase (deg/ppm)'] = [np.round(p1,decimals=5)]
tab = create_table(df)
fig = go.Figure(data=[tab])
fig.update_layout(autosize=True,margin=dict(l=0,r=0,t=0,b=0))
return fig
# fig = ff.create_table(df, height_constant=50)
# for i in range(len(fig.layout.annotations)):
# fig.layout.annotations[i].font.size = 16
# fig.layout.update(title={'text': 'Phase parameters'},font={'size':20})
# fig.update_layout(template = 'plotly_white')
# fig.update_layout()
#return fig
def plot_table_qc(res):
# Peak by peak snr and fwhm
snr,fwhm = res.getQCParams()
df = pd.DataFrame([snr.to_numpy(),fwhm.to_numpy()],columns=res.original_metabs).T
df.columns = ['SNR','FWHM (Hz)']
df = pd.DataFrame()
df['Metab'] = res.original_metabs
df['SNR'] = snr.to_numpy()
df['FWHM (Hz)'] = fwhm.to_numpy()
df['SNR'] = df['SNR'].map('{:.1f}'.format)
df['FWHM (Hz)'] = df['FWHM (Hz)'].map('{:.1f}'.format)
fig = go.Figure(data=[go.Table(
header=dict(values=["Metabolites"]+list(df.columns),
align='left'),
cells=dict(values=[df.index,df.SNR, df['FWHM (Hz)']],
align='left',
format=('','0.1f','0.1f')))
])
fig.update_layout(template = 'plotly_white')
# fig.layout.update({'width':1000,'height':200})
df['FWHM (Hz)'] = df['FWHM (Hz)'].map('{:.1f}'.format)
tab = create_table(df)
fig = go.Figure(data=[tab])
fig.update_layout(autosize=True,margin=dict(l=0,r=0,t=0,b=0))
return fig
#fig = ff.create_table(df, height_constant=30)
#for i in range(len(fig.layout.annotations)):
# fig.layout.annotations[i].font.size = 16
#fig.layout.update(title={'text': 'QC parameters'},font={'size':20})
#fig.update_layout(template = 'plotly_white')
#return fig
# ----------- Imaging
......
fsl_mrs/utils/report.py
View file @ ec1ae277
......@@ -99,41 +99,40 @@ import nibabel as nib
def create_plotly_div(mrs,res):
divs=[]
divs={}
def to_div(fig):
return plotly.offline.plot(fig, output_type='div',include_plotlyjs='cdn')
return plotly.offline.plot(fig,
output_type='div',
include_plotlyjs='cdn')
# Summary plot
fig = plotting.plotly_fit(mrs,res)
divs.append(to_div(fig))
divs['summary'] = to_div(fig)
fig = plotting.plot_table_fitparams(res)
divs.append(to_div(fig))
# tables
# nuisance
fig = plotting.plot_table_lineshape_phase(res)
divs['table-lineshape-phase'] = to_div(fig)
# qc
fig = plotting.plot_table_qc(res)
divs.append(to_div(fig))
# Approximate (Laplace) marginals
fig = plotting.plot_dist_approx(res,refname=res.concScalings['internalRef'])
div2 = plotly.offline.plot(fig, output_type='div',include_plotlyjs='cdn')
divs.append(div2)
divs['table-qc'] = to_div(fig)
# MCMC results (if available)
if res.method == 'MH':
fig = plotting.plot_mcmc_corr(res)
div3 = plotly.offline.plot(fig, output_type='div',include_plotlyjs='cdn')
# divs.append(div3)
fig = plotting.plot_dist_mcmc(res,refname=res.concScalings['internalRef'])
div4 = plotly.offline.plot(fig, output_type='div',include_plotlyjs='cdn')
divs.append([div3,div4])
fig1 = plotting.plot_corr(res,title='MCMC Correlations')
fig2 = plotting.plot_dist_mcmc(res,refname=res.concScalings['internalRef'])
else:
fig1 = plotting.plot_corr(res,corr=res.corr,title='Laplace approx Correlations')
fig2 = plotting.plot_dist_approx(res,refname=res.concScalings['internalRef'])
divs['corr'] = to_div(fig1)
divs['posteriors'] = to_div(fig2)
fig = plotting.plot_real_imag(mrs,res,ppmlim=(.2,4.2))
div = plotly.offline.plot(fig, output_type='div',include_plotlyjs='cdn')
divs.append(div)
divs['real-imag'] = to_div(fig) #plotly.offline.plot(fig, output_type='div',include_plotlyjs='cdn')
fig = plotting.plot_indiv_stacked(mrs,res,ppmlim=res.ppmlim)
div = plotly.offline.plot(fig, output_type='div',include_plotlyjs='cdn')
divs.append(div)
divs['indiv'] = to_div(fig) #plotly.offline.plot(fig, output_type='div',include_plotlyjs='cdn')
return divs
......@@ -180,13 +179,13 @@ def create_vox_plot(t1file,voxfile,outdir):
def create_report(mrs,res,filename,fidfile,basisfile,h2ofile,date):
divlist= create_plotly_div(mrs,res)
divs= create_plotly_div(mrs,res)
#if t1 is not None:
# create_vox_plot(t1file=t1,outdir=outdir)
template = """<!DOCTYPE html>
template = f"""<!DOCTYPE html>
<html>
<head>
<style>
......@@ -197,56 +196,98 @@ def create_report(mrs,res,filename,fidfile,basisfile,h2ofile,date):
white-space: pre;
margin: 1em 0;
}}
.container{{
display: flex;
}}
.fixed{{
width: 200px;
}}
.flex-item{{
flex-grow: 1;
}}
</style>
<script src="https://cdn.plot.ly/plotly-latest.min.js"></script>
</head>
<body>
<h1>FSL MRS Report</h1>
<div class="header">
<center><h1>FSL MRS Report</h1></center>
<hr>
<pre>
Date : {}
FID : {}
Basis : {}
H2O : {}
Date : {date}
FID : {fidfile}
Basis : {basisfile}
H2O : {h2ofile}
</pre>
</div>
<hr>
<center>
<a href="#summary">Summary</a> -
<a href="#nuisance">Nuisance</a> -
<a href="#qc">QC</a> -
<a href="#uncertainty">Uncertainty</a> -
<a href="#realimag">Real/Imag</a> -
<a href="#metabs">Metabs</a>
</center>
<hr>
""".format(date,fidfile,basisfile,h2ofile)
"""
# Summary section
section=f"""
<h1><a name="summary">Summary</a></h1>
<div id=fit>{divs['summary']}</div>
<hr>
"""
template+=section
# Tables section
section=f"""
<h1><a name="nuisance">Nuisance parameters</a></h1>
<div style="width:70%">{divs['table-lineshape-phase']}</div>
<hr>
<h1><a name="qc">QC parameters</a></h1>
<div style="width:70%">{divs['table-qc']}</div>
<hr>
"""
template+=section
# Dist section
section=f"""
<h1><a name="uncertainty">Uncertainties</a></h1>
<table width=100%>
<tr>
<th style="vertical-align:top">{divs['corr']}</th>
<th style="vertical-align:top">{divs['posteriors']}</th>
</tr>
</table>
<hr>
"""
template+=section
# Real/imag section
section=f"""
<h1><a name="realimag">Fitting summary (real/imag)</a></h1>
{divs['real-imag']}
<hr>
"""
template+=section
# Indiv spectra section
section=f"""
<h1><a name="metabs">Individual metabolite spectra</a></h1>
{divs['indiv']}
<hr>
"""
template+=section
# add voxplot?
#if t1 is not None:
# create_vox_plot(t1file,voxfile,outdir)
# voxplothtml='<p><img src="voxplot.png">voxplot</img></p>'
# template+=voxplothtml
for i,div in enumerate(divlist):
if isinstance(div,list):
section=f"""
<div id='divPlotly{i}' class='container'>
<div class='flex-item'>
{div[0]}
</div>
<div class='flex-item'>
{div[1]}
</div>
</div>
"""
template+=section
else:
section="""
<div id='divPlotly{}'>
{}
</div>
"""
template+=section.format(i,div)
# End of report
template+="""
</body>
</html>
......
Supports Markdown