Related
I want to reshape the array G in every iteration, and the shape for new dimensions are coming from vector (say) tensorSize. But the MATLAB reshape command does not accept the below given method,
tensorSize = [5,6,7,9,3,4];
r=[1,5,25,225,45,9,1];
G1(1) = {randn(1,tensorSize(1),r(2))};
G1(2) = {randn(r(2),tensorSize(2),r(3))};
G1(3) = {randn(r(3),tensorSize(3),r(4))};
G1(4) = {randn(r(4),tensorSize(4),r(5))};
G1(5) = {randn(r(5),tensorSize(5),r(6))};
G1(6) = {randn(r(6),tensorSize(6),1)};
for j = 1:length(tensorSize)-1
if j == 1
G = G1(j);
end
G = reshape(G,[],r(j+1));
H = reshape(G1(j+1),r(j+1),[]);
G = G*H;
G = reshape(G,tensorSize(1:j+1),[]);
end
I have also tried to use other alternatives like:
str2num(regexprep(num2str(tensorSize(1:j+1),),'\s+',','))
str2num(strjoin(cellstr(tensorSize(1:j+1)),','))
but they create a string and when converted to num, they are not comma separated. So the reshape option does not accept it.
Is there any work around?
Thanks to #beaker in the comment section below, for proposing this solution!
tensorSize = [5,6,7,9,3,4];
r=[1,5,25,225,45,9,1];
G1(1) = {randn(1,tensorSize(1),r(2))};
G1(2) = {randn(r(2),tensorSize(2),r(3))};
G1(3) = {randn(r(3),tensorSize(3),r(4))};
G1(4) = {randn(r(4),tensorSize(4),r(5))};
G1(5) = {randn(r(5),tensorSize(5),r(6))};
G1(6) = {randn(r(6),tensorSize(6),1)};
tensorSizeCell = {zeros(1,length(tensorSize))};
for i = 1:length(tensorSize)
tensorSizeCell(i) = {tensorSize(i)};
end
for j = 1:length(tensorSize)-1
if j == 1
G = cell2mat(G1(j));
end
G = reshape(G,[],r(j+1));
H = reshape(cell2mat(G1(j+1)),r(j+1),[]);
G = G*H;
G = reshape(G,tensorSizeCell{1:j+1},[]);
end
I have a function that gives me a value in the form of an array as an output when run but I need the output as an integer percentage for a results piece
def pred_datsci(file_path):
prev_precompute = learn.precompute
learn.precompute = False
try:
trn_tfms, val_tfms = tfms_from_model(arch,sz)
test_img = open_image(file_path)
im = val_tfms(test_img)
pred = learn.predict_array(im[None])
class_index = (np.exp(pred))
class_index1 = np.argmax(np.exp(pred))
print(class_index*100)
return data.classes[class_index1]
finally:
learn.precompute = prev_precompute
This is what the output looks like:
pred_datsci(f"data/dogscats1/valid/dogs/12501.jpg")
My question is how do I get these two values to display as :
Cat % = 15.81724%
Dog % = 84.18274%
You can use zip function as:
for z in zip(['Cat', 'Dog'], [15.3, 84.6]):
print('%s %% = %s%%'%(z[0], z[1]))
I get "IndexError: list is out of range" when I input this code. Also, the retmax is set at 614 because that's the total number of results when I make the request. Is there a way to make the retmode equal to the number of results using a variable that changes depending on the search results?
#!/usr/bin/env python
from Bio import Entrez
Entrez.email = "something#gmail.com"
handle1 = Entrez.esearch(db = "nucleotide", term = "dengue full genome", retmax = 614)
record = Entrez.read(handle1)
IdNums = [int(i) for i in record['IdList']]
while i >= 0 and i <= len(IdNums):
handle2 = Entrez.esearch(db = "nucleotide", id = IdNums[i], type = "gb", retmode = "text")
record = Entrez.read(handle2)
print(record)
i += 1
Rather than using a while loop, you can use a for loop...
from Bio import Entrez
Entrez.email = 'youremailaddress'
handle1 = Entrez.esearch(db = 'nucleotide', term = 'dengue full genome', retmax = 614)
record = Entrez.read(handle1)
IdNums = [int(i) for i in record['IdList']]
for i in IdNums:
print(i)
handle2 = Entrez.esearch(db = 'nucleotide', term = 'dengue full genome', id = i, rettype = 'gb', retmode = 'text')
record = Entrez.read(handle2)
print(record)
I ran it on my computer and it seems to work. The for loop solved the out of bounds, and adding the term to handle2 solved the calling error.
This question already has an answer here:
MATLAB : What is the mistake in my Ramachandran plot?
(1 answer)
Closed 8 years ago.
I am trying matlab to plot ramachandran plot, without using built in command. I have succeeded too. Now I wanted to spot the GLYCINEs alone in the scatter array. Any ideas how to do this? (link to 1UBQ.pdb file : http://www.rcsb.org/pdb/download/downloadFile.do?fileFormat=pdb&compression=NO&structureId=1UBQ)
% Program to plot Ramanchandran plot of Ubiquitin
close all; clear ; clc; % close all figure windows, clear variables, clear screen
pdb1 ='/home/devanandt/Documents/VMD/1UBQ.pdb';
p=pdbread(pdb1); % read pdb file corresponding to ubiquitin protein
atom={p.Model.Atom.AtomName};
n_i=find(strcmp(atom,'N')); % Find indices of atoms
ca_i=find(strcmp(atom,'CA'));
c_i=find(strcmp(atom,'C'));
X = [p.Model.Atom.X];
Y = [p.Model.Atom.Y];
Z = [p.Model.Atom.Z];
X_n = X(n_i(2:end)); % X Y Z coordinates of atoms
Y_n = Y(n_i(2:end));
Z_n = Z(n_i(2:end));
X_ca = X(ca_i(2:end));
Y_ca = Y(ca_i(2:end));
Z_ca = Z(ca_i(2:end));
X_c = X(c_i(2:end));
Y_c = Y(c_i(2:end));
Z_c = Z(c_i(2:end));
X_c_ = X(c_i(1:end-1)); % the n-1 th C (C of cabonyl)
Y_c_ = Y(c_i(1:end-1));
Z_c_ = Z(c_i(1:end-1));
V_c_ = [X_c_' Y_c_' Z_c_'];
V_n = [X_n' Y_n' Z_n'];
V_ca = [X_ca' Y_ca' Z_ca'];
V_c = [X_c' Y_c' Z_c'];
V_ab = V_n - V_c_;
V_bc = V_ca - V_n;
V_cd = V_c - V_ca;
phi=0;
for k=1:numel(X_c)
n1=cross(V_ab(k,:),V_bc(k,:))/norm(cross(V_ab(k,:),V_bc(k,:)));
n2=cross(V_bc(k,:),V_cd(k,:))/norm(cross(V_bc(k,:),V_cd(k,:)));
x=dot(n1,n2);
m1=cross(n1,(V_bc(k,:)/norm(V_bc(k,:))));
y=dot(m1,n2);
phi=cat(2,phi,-atan2d(y,x));
end
phi=phi(1,2:end);
X_n_ = X(n_i(2:end)); % (n+1) nitrogens
Y_n_ = Y(n_i(2:end));
Z_n_ = Z(n_i(2:end));
X_ca = X(ca_i(1:end-1));
Y_ca = Y(ca_i(1:end-1));
Z_ca = Z(ca_i(1:end-1));
X_n = X(n_i(1:end-1));
Y_n = Y(n_i(1:end-1));
Z_n = Z(n_i(1:end-1));
X_c = X(c_i(1:end-1));
Y_c = Y(c_i(1:end-1));
Z_c = Z(c_i(1:end-1));
V_n_ = [X_n_' Y_n_' Z_n_'];
V_n = [X_n' Y_n' Z_n'];
V_ca = [X_ca' Y_ca' Z_ca'];
V_c = [X_c' Y_c' Z_c'];
V_ab = V_ca - V_n;
V_bc = V_c - V_ca;
V_cd = V_n_ - V_c;
psi=0;
for k=1:numel(X_c)
n1=cross(V_ab(k,:),V_bc(k,:))/norm(cross(V_ab(k,:),V_bc(k,:)));
n2=cross(V_bc(k,:),V_cd(k,:))/norm(cross(V_bc(k,:),V_cd(k,:)));
x=dot(n1,n2);
m1=cross(n1,(V_bc(k,:)/norm(V_bc(k,:))));
y=dot(m1,n2);
psi=cat(2,psi,-atan2d(y,x));
end
psi=psi(1,2:end);
scatter(phi,psi)
box on
axis([-180 180 -180 180])
title('Ramachandran Plot for Ubiquitn Protein','FontSize',16)
xlabel('\Phi^o','FontSize',20)
ylabel('\Psi^o','FontSize',20)
grid
The output is :
EDIT : Is my plot correct? Biopython: How to avoid particular amino acid sequences from a protein so as to plot Ramachandran plot? has an answer which has slightly different plot.
The modified code is as below :
% Program to plot Ramanchandran plot of Ubiquitin with no glycines
close all; clear ; clc; % close all figure windows, clear variables, clear screen
pdb1 ='/home/devanandt/Documents/VMD/1UBQ.pdb';
p=pdbread(pdb1); % read pdb file corresponding to ubiquitin protein
atom={p.Model.Atom.AtomName};
n_i=find(strcmp(atom,'N')); % Find indices of atoms
ca_i=find(strcmp(atom,'CA'));
c_i=find(strcmp(atom,'C'));
X = [p.Model.Atom.X];
Y = [p.Model.Atom.Y];
Z = [p.Model.Atom.Z];
X_n = X(n_i(2:end)); % X Y Z coordinates of atoms
Y_n = Y(n_i(2:end));
Z_n = Z(n_i(2:end));
X_ca = X(ca_i(2:end));
Y_ca = Y(ca_i(2:end));
Z_ca = Z(ca_i(2:end));
X_c = X(c_i(2:end));
Y_c = Y(c_i(2:end));
Z_c = Z(c_i(2:end));
X_c_ = X(c_i(1:end-1)); % the n-1 th C (C of cabonyl)
Y_c_ = Y(c_i(1:end-1));
Z_c_ = Z(c_i(1:end-1));
V_c_ = [X_c_' Y_c_' Z_c_'];
V_n = [X_n' Y_n' Z_n'];
V_ca = [X_ca' Y_ca' Z_ca'];
V_c = [X_c' Y_c' Z_c'];
V_ab = V_n - V_c_;
V_bc = V_ca - V_n;
V_cd = V_c - V_ca;
phi=0;
for k=1:numel(X_c)
n1=cross(V_ab(k,:),V_bc(k,:))/norm(cross(V_ab(k,:),V_bc(k,:)));
n2=cross(V_bc(k,:),V_cd(k,:))/norm(cross(V_bc(k,:),V_cd(k,:)));
x=dot(n1,n2);
m1=cross(n1,(V_bc(k,:)/norm(V_bc(k,:))));
y=dot(m1,n2);
phi=cat(2,phi,-atan2d(y,x));
end
phi=phi(1,2:end);
X_n_ = X(n_i(2:end)); % (n+1) nitrogens
Y_n_ = Y(n_i(2:end));
Z_n_ = Z(n_i(2:end));
X_ca = X(ca_i(1:end-1));
Y_ca = Y(ca_i(1:end-1));
Z_ca = Z(ca_i(1:end-1));
X_n = X(n_i(1:end-1));
Y_n = Y(n_i(1:end-1));
Z_n = Z(n_i(1:end-1));
X_c = X(c_i(1:end-1));
Y_c = Y(c_i(1:end-1));
Z_c = Z(c_i(1:end-1));
V_n_ = [X_n_' Y_n_' Z_n_'];
V_n = [X_n' Y_n' Z_n'];
V_ca = [X_ca' Y_ca' Z_ca'];
V_c = [X_c' Y_c' Z_c'];
V_ab = V_ca - V_n;
V_bc = V_c - V_ca;
V_cd = V_n_ - V_c;
psi=0;
for k=1:numel(X_c)
n1=cross(V_ab(k,:),V_bc(k,:))/norm(cross(V_ab(k,:),V_bc(k,:)));
n2=cross(V_bc(k,:),V_cd(k,:))/norm(cross(V_bc(k,:),V_cd(k,:)));
x=dot(n1,n2);
m1=cross(n1,(V_bc(k,:)/norm(V_bc(k,:))));
y=dot(m1,n2);
psi=cat(2,psi,-atan2d(y,x));
end
psi=psi(1,2:end);
res=strsplit(p.Sequence.ResidueNames,' ');
angle =[phi;psi];
angle(:,find(strcmp(res,'GLY'))-1)=[];
scatter(angle(1,:),angle(2,:))
box on
axis([-180 180 -180 180])
title('Ramachandran Plot for Ubiquitn Protein','FontSize',16)
xlabel('\Phi^o','FontSize',20)
ylabel('\Psi^o','FontSize',20)
grid
which gives output (with no GLY) as below :
I would change this code block to use logical indexing
res=strsplit(p.Sequence.ResidueNames,' ');
angle =[phi;psi];
angle(:,find(strcmp(res,'GLY'))-1)=[];
Instead:
residues = strsplit(p.Sequency.ResidueNames,' ');
glycine = ismember(residues,'GLY');
angle = [phi;psi];
angleNoGLY= angle(:,~glycine);
Doing it this way, if you wanted to highlight glycine (or any other residue) you can easily call it out:
angleGLY = angle(:,glycine);
plot(angleNoGLY(1,:),angleNoGLY(2,:),'ob')
line(angleGLY(1,:),angleGLY(2,:),'Marker','o','Color','r','LineStyle','none')
Below is a section for my code.
The user inputs a value for Energy, if the value is on the array ARRDAT(1,x), then I want it to assign the related variables ARRDAT(2,x), ARRDAT(3,x), and ARRDAT(4,x).
If the data is not on the array, I will interpolate (note: only values between two of my given points will be inputted).
I am able to compile it with a warning
warning 179 - Comparning floating point quantities for equality may give misleading results
and as soon as I input any value, whether it's in the table or not I get
error 11, array subscript(s) out of bounds
The error occurs on the line
IF (ARRDAT(1,x) == Energy) THEN
I haven't been able to fix it.
IMPLICIT NONE
REAL :: Dose, Energy, Source, FeAtt, PbAtt, Resp, Distance, Mu, Thickness
INTEGER :: x, y, upp, low
CHARACTER(LEN=4) :: Material
REAL, PARAMETER :: PI = 3.1415926
real, dimension(1:4,1:25) :: ARRDAT
!Energy MeV
ARRDAT(1,1) = 0.01
ARRDAT(1,2) = 0.015
ARRDAT(1,3) = 0.02
ARRDAT(1,4) = 0.03
ARRDAT(1,5) = 0.04
ARRDAT(1,6) = 0.05
ARRDAT(1,7) = 0.06
ARRDAT(1,8) = 0.08
ARRDAT(1,9) = 0.10
ARRDAT(1,10) = 0.15
ARRDAT(1,11) = 0.20
ARRDAT(1,12) = 0.30
ARRDAT(1,13) = 0.40
ARRDAT(1,14) = 0.50
ARRDAT(1,15) = 0.60
ARRDAT(1,16) = 0.80
ARRDAT(1,17) = 1.0
ARRDAT(1,18) = 1.5
ARRDAT(1,19) = 2.0
ARRDAT(1,20) = 3.0
ARRDAT(1,21) = 4.0
ARRDAT(1,22) = 5.0
ARRDAT(1,23) = 6.0
ARRDAT(1,24) = 8.0
ARRDAT(1,25) = 10.0
!Response Function Sv*cm2
ARRDAT(2,1) = 0.062
ARRDAT(2,2) = 0.157
ARRDAT(2,3) = 0.238
ARRDAT(2,4) = 0.329
ARRDAT(2,5) = 0.365
ARRDAT(2,6) = 0.384
ARRDAT(2,7) = 0.400
ARRDAT(2,8) = 0.451
ARRDAT(2,9) = 0.533
ARRDAT(2,10) = 0.777
ARRDAT(2,11) = 1.03
ARRDAT(2,12) = 1.56
ARRDAT(2,13) = 2.06
ARRDAT(2,14) = 2.54
ARRDAT(2,15) = 2.99
ARRDAT(2,16) = 3.83
ARRDAT(2,17) = 4.60
ARRDAT(2,18) = 6.24
ARRDAT(2,19) = 7.66
ARRDAT(2,20) = 10.2
ARRDAT(2,21) = 12.5
ARRDAT(2,22) = 14.7
ARRDAT(2,23) = 16.7
ARRDAT(2,24) = 20.8
ARRDAT(2,25) = 24.7
!mu/rho for iron cm2/g
ARRDAT(3,1) = 169.4
ARRDAT(3,2) = 56.33
ARRDAT(3,3) = 25.16
ARRDAT(3,4) = 7.891
ARRDAT(3,5) = 3.450
ARRDAT(3,6) = 1.833
ARRDAT(3,7) = 1.113
ARRDAT(3,8) = 0.5391
ARRDAT(3,9) = 0.3340
ARRDAT(3,10) = 0.1786
ARRDAT(3,11) = 0.1357
ARRDAT(3,12) = 0.1051
ARRDAT(3,13) = 0.09131
ARRDAT(3,14) = 0.08241
ARRDAT(3,15) = 0.07583
ARRDAT(3,16) = 0.06631
ARRDAT(3,17) = 0.05951
ARRDAT(3,18) = 0.04863
ARRDAT(3,19) = 0.04254
ARRDAT(3,20) = 0.03616
ARRDAT(3,21) = 0.03309
ARRDAT(3,22) = 0.03144
ARRDAT(3,23) = 0.03056
ARRDAT(3,24) = 0.02991
ARRDAT(3,25) = 0.02994
!mu/rho for lead cm2/g
ARRDAT(4,1) = 125.7
ARRDAT(4,2) = 108.3
ARRDAT(4,3) = 84.02
ARRDAT(4,4) = 28.94
ARRDAT(4,5) = 13.44
ARRDAT(4,6) = 7.386
ARRDAT(4,7) = 4.531
ARRDAT(4,8) = 2.112
ARRDAT(4,9) = 5.337
ARRDAT(4,10) = 1.910
ARRDAT(4,11) = 0.9359
ARRDAT(4,12) = 0.3732
ARRDAT(4,13) = 0.2148
ARRDAT(4,14) = 0.1499
ARRDAT(4,15) = 0.1167
ARRDAT(4,16) = 0.08408
ARRDAT(4,17) = 0.06803
ARRDAT(4,18) = 0.05087
ARRDAT(4,19) = 0.04530
ARRDAT(4,20) = 0.04200
ARRDAT(4,21) = 0.04178
ARRDAT(4,22) = 0.04260
ARRDAT(4,23) = 0.04382
ARRDAT(4,24) = 0.04670
ARRDAT(4,25) = 0.04969
WRITE(*,*) 'Please input the particle energy in MeV'
READ(*,*) Energy
x=0
DO x = 0, 25, 1
IF (ARRDAT(1,x) == Energy) THEN !if the data is already on our table
ARRDAT(2,x) = Resp
ARRDAT(3,x) = FeAtt
ARRDAT(4,x) = PbAtt
WRITE(*,*) 'CHECK 1'
ELSE IF (ARRDAT(1,x) < Energy) THEN !if the data is between two points
upp = x
low = x - 1
Energy = ((ARRDAT(1,upp) + ARRDAT(1,low))/2)
WRITE(*,*) Energy
ELSE
WRITE(*,*) 'Invalid Entry'
END IF
END DO
You declare your array to be 1-indexed with real, dimension(1:4,1:25) :: ARRDAT, but your DO loop begins with x = 0, so the first iteration of the loop looks for ARRDAT(1,0) == Energy, which is out of bounds. Start your DO loop with x = 1 and you should be fine.
Side note: Fortran automatically indexes arrays beginning with 1, so you can declare your array as real, dimension(4,25) :: ARRDAT, and it will have the same effect.
EDIT: As mentioned by #user5713492 in the comments, this code will still crash if the input Energy is greater than ARRDAT(1,1). You will either have to fix the lower bound of the linear interpolation or skip the interpolation entirely in the first iteration of the loop. For example:
DO x = 1, 25
IF (ARRDAT(1,x) == Energy) THEN
! ...
ELSE IF (ARRDAT(1,x) < Energy) THEN
upp = x
low = x - 1
IF (low < 1) low = 1 ! <-- make sure we don't go out of bounds on the array
! ...
END IF
END DO
... or ...
DO x = 1, 25
IF (ARRDAT(1,x) == Energy) THEN
! ...
ELSE IF (x > 1 .AND. ARRDAT(1,x) < Energy) THEN ! <-- don't interpolate for the first iteration
! ...
END IF
END DO
Now, about that warning concerning comparison of floating point numbers for equality: please read this and try to avoid ==, >=, and <= operators when comparing reals.