I want to ask a question, I have trained a model with two gpus and stored this model with
model.module.state_dict(), now I want to load this model in one gpu, can I directly load this trained model with model.state_dict()?
Thanks in advance!
You can refer to this question.
You can either adding an nn.DataParallel for loading purpose. Or change the key naming like
# original saved file with DataParallel
state_dict = torch.load('myfile.pth.tar')
# create new OrderedDict that does not contain `module.`
from collections import OrderedDict
new_state_dict = OrderedDict()
for k, v in state_dict.items():
name = k[7:] # remove `module.`
new_state_dict[name] = v
# load params
model.load_state_dict(new_state_dict)
But as you save the model with model.module.state_dict() instead of model.state_dict() it possible that the names may differed. If the two methods above don't work try print the saved dict and model to see what you need to change. Like
state_dict = torch.load('myfile.pth.tar')
print(state_dict)
print(model)
Related
I have large data sets which i want to work with in matlab.
I have a struct called Trail containing serveral structures called trail1, trail2 ...
which then contain several matrices. I now want to add another point to for instance trail1
I can do that with Trail.trail1.a2rotated(i,:) = rotpoint'; the problem is that i have to do it in a loop where the trail number as well as the a2rotated changes to e.g. a3rot...
I tired to do it like that
name ="trail"+num2str(z)+".a2rotated"+"("+i+",:)";
name = convertStringsToChars(name);
Trail.(name) = rotpoint'
But that gives me the error: Invalid field name: 'trail1.a2rotated(1,:)'.
Does someone have a solution?
The name in between brackets after the dot must be the name of a field of the struct. Other indexing operations must be done separately:
Trail.("trail"+z).a2rotated(i,:)
But you might be better off making trail(z) an array instead of separate fields with a number in the name.
I have a list of .txt datafiles to import. Suppose they are called like that
file100data.txt file101data.txt ... file109data.txt I want to import them all using readtable.
I tried using the for to specify a vector a = [0:9] through which matlab could loop the readtable command but I cannot make it work.
for a = [0:9]
T_a_ = readtable('file10_a_data.txt')
end
I know I cannot just put _a_ where I want the vector to loop through, so my question is how can I actually do it?
Thank you in advance!
Here is a solution that should work even if you have missing files in your folder (e.g. you have file100data.txt to file107data.txt, but you are missing file file108data.txt and file109data.txt):
files=dir('file10*data.txt'); %list all data files in your folder
nof=size(files,1); %number of files
for i=1:nof %loop over the number of files
table_index=files(i).name(7) %recover table index from data filename
eval(sprintf('T%s = readtable(files(i).name)', table_index)); %read table
end
Now, please note that is it generally regarded as poor practice to dynamically name variables in Matlab (see this post for example). You may want to resort to structures or cells to store your data.
You need to convert the value of a into a string and combine strings together, like this:
Tables = struct();
for a = 0:9
% note: using dynamic structure field names to store the imported tables
fname = ['file10_' num2str(a) '_data'];
Tables.(fname) = readtable([fname '.txt']);
end
I do not manage to create an Orange table with StringVariables.
The following code:
d = Orange.data.Domain([Orange.data.StringVariable("s")])
makes this error:
TypeError: variables must be primitive
It seems that StringVariable is for metadata only. So I'm worried about this because my data has a lot of strings that it would be crazy to put in a discrete structure (each string value is different).
Is there a solution for putting strings in a table ?
Thanks in advance for the answers,
Best,
mike
This question might be old but I found it via Google and wanted to provide a simple example of how to use meta "columns".
You need to specify the meta variables the same way you specify the "normal" variables just do it inside the metas parameter inside the Domain constructor.
from Orange.data import *
taskid = StringVariable(name="taskid")
logdata = StringVariable(name="logdata")
domain = Domain([] , metas=[taskid, logdata])
data = Table(domain, [
["uuid1","some more stuff"],
["uuid2","some more stuff"]
]);
out_data = data;
I need to know how to save just one branch of a structure in MATLAB.
The structure contains more levels with more fields per level. For example:
data.level1.level21
data.level1.level22
I want now to save the branches data.level1.level21 and data.level1.level21 individually. I have tried the following but it doesn't work:
firstLevelName = fieldnames(data);
secondLevelNames = fieldnames(data.(firstLevelName{1}));
for pL = 1:length(secondLevelNames)
save([filename '.mat'], '-struct', 'data', firstLevelName{1}, secondLevelNames{pL});
end
The structure-saving method that you're trying to use doesn't work quite the way you are expecting. All the arguments after your struct variable name are fields of that struct to save.
The way MATLAB is interpreting your code is that you're trying to save the level1, and level21 fields of data which obviously doesn't work since level21 is a subfield of level1 not data.
To save the nested fields, the easiest thing is probably to create a new variable pointing to the structure data.level and then call save on that and specify the specific fields to save.
level1 = data.level1;
for pL = 1:numel(secondLevelNames)
save(filename, '-struct', 'level1', secondLevelNames{pL});
end
If you actually want the double nesting in the saved data, you would need to create a new structure containing only the data that you wanted and then save that.
for pL = 1:numel(secondLevelNames)
newstruct = struct(firstLevelName{1}, struct());
newstruct.(secondLevelNames{pL}) = data.(firstLevelName{1}).(secondLevelNames{pL});
save(filename, '-struct', 'newstruct')
end
I want to create a structure with a variable name in a matlab script. The idea is to extract a part of an input string filled by the user and to create a structure with this name. For example:
CompleteCaseName = input('s');
USER WRITES '2013-06-12_test001_blabla';
CompleteCaseName = '2013-06-12_test001_blabla'
casename(12:18) = struct('x','y','z');
In this example, casename(12:18) gives me the result test001.
I would like to do this to allow me to compare easily two cases by importing the results of each case successively. So I could write, for instance :
plot(test001.x,test001.y,test002.x,test002.y);
The problem is that the line casename(12:18) = struct('x','y','z'); is invalid for Matlab because it makes me change a string to a struct. All the examples I find with struct are based on a definition like
S = struct('x','y','z');
And I can't find a way to make a dynamical name for S based on a string.
I hope someone understood what I write :) I checked on the FAQ and with Google but I wasn't able to find the same problem.
Use a structure with a dynamic field name.
For example,
mydata.(casename(12:18)) = struct;
will give you a struct mydata with a field test001.
You can then later add your x, y, z fields to this.
You can use the fields later either by mydata.test001.x, or by mydata.(casename(12:18)).x.
If at all possible, try to stay away from using eval, as another answer suggests. It makes things very difficult to debug, and the example given there, which directly evals user input:
eval('%s = struct(''x'',''y'',''z'');',casename(12:18));
is even a security risk - what happens if the user types in a string where the selected characters are system(''rm -r /''); a? Something bad, that's what.
As I already commented, the best case scenario is when all your x and y vectors have same length. In this case you can store all data from the different files into 2 matrices and call plot(x,y) to plot each column as a series.
Alternatively, you can use a cell array such that:
c = cell(2,nufiles);
for ii = 1:numfiles
c{1,ii} = import x data from file ii
c{2,ii} = import y data from file ii
end
plot(c{:})
A structure, on the other hand
s.('test001').x = ...
s.('test001').y = ...
Use eval:
eval(sprintf('%s = struct(''x'',''y'',''z'');',casename(12:18)));
Edit: apologies, forgot the sprintf.