Too many input arguments.

 clc;
 close all;
 clear all;
%load train dataset
 train=xlsread('trainchanged.csv','A2:AG1000');
 class_label=train(:,33);
 
%exclude eventid,weight from init_table
 train_sample=train(:,2:31); 
 
%appending class label
  final_train=[train_sample class_label];
  
  x=final_train(:,1:30);
  xx=array2table(x);
  yy=categorical(final_train(:,end));
  
  
 %lstm definition
  layers = [ ...
    sequenceInputLayer(1)
    bilstmLayer(100,'OutputMode','last')
    fullyConnectedLayer(2)
    softmaxLayer
    classificationLayer
    ];
 options = trainingOptions('adam', ...
    'MaxEpochs',10, ...
    'MiniBatchSize', 150, ...
    'InitialLearnRate', 0.01, ...
    'SequenceLength', 1000, ...
    'GradientThreshold', 1, ...
    'ExecutionEnvironment',"auto",...
    'plots','training-progress', ...
    'Verbose',false);
%train the network
 net = trainNetwork(xx,yy,layers,options);

how to resolve the error: Too many input arguments.

[Adam Danz]

How about you make it easy for us to help you by telling us what line is producing the code. Sharing the entire copy-pasted error message would give us a lot of info, too.

BTW, you know that we cannot run that code, right?

[Chandini Govind]

well here is my whole error

Error in study8 (line 38)
 net = trainNetwork(xx,yy,layers,options);

Caused by:
    Error using trainNetwork>iParseInputArguments (line 274)
    Too many input arguments.

i know we can't run the code. i want know if it is due to some logical mistake.

[Walter Roberson]

Which MATLAB version are you using?

[Chandini Govind]

2018b

[Adam Danz]

"i know we can't run the code. i want know if it is due to some logical mistake."

@ Chandini Govind , I was able to identify the error because I happen to have r2018b on my laptop. Note that r2019b, the most recent release, does not have the same code for this function and we could not have identified the problem using the current release.

If you would have provided us with inputs, almost any experienced volunteer here could have instantly seen that your first input is a table while you're using numeric syntax. This is why it's almost always helpful to provide inputs that allow us to reproduce the problem.

Answer 1

[Adam Danz]

Understand the input options

In your version of Matlab (r2018b) as well as in the current release (r2019b), there are multiple syntax options for this function that allow for various input types:

• image datastore inputs net = trainNetwork(imds,layers,options) 3 inputs
• datastore objects net = trainNetwork(ds,layers,options) 3 inputs
• numeric arrays net = trainNetwork(X,Y,layers,options) 4 inputs
• and tables net = trainNetwork(tbl,layers,options) 3 inputs

Your code uses the numeric input syntax

net = trainNetwork(xx,yy,layers,options);

where xx and yy are expected to be numeric arrays.

What does your error tell us?

Your error in r2018b is thrown at line 274 in trainNetwork().

Error using trainNetwork>iParseInputArguments (line 274)

That line is within a conditional statement that indicates that your first input xx is actually a table. As you can see in the image below. If you look at the "table" syntax above, you'll see that it only has 3 inputs.

Solution

Use the right syntax for tables.

[Chandini Govind]

Error in study8 (line 39)
 net = trainNetwork(xx,layers,options);

Caused by:

  Error using cellfun
    Input #2 expected to be a cell array, was double instead.

i'm getting this error(snapshot from workspace).is there any example for handling neural network with table?

[Adam Danz]

".is there any example for handling neural network with table?"

Unfortunately and ironically, the table input syntax is the only syntax in the documentation that does not have an example :D

I'd be happy to look into this further if you attach a mat file containing the inputs to trainNetwork.

save('trainNetworkInputs.mat','xx','layers','options') 
net = trainNetwork(xx,layers,options);

[Walter Roberson]

For a table, the first variable must consist either of 3D numeric arrays (which would require cell array to store in tables), or else of path names to image files, or else of path names to .mat file. Instead your first column is numeric (scalar for each row).

[Chandini Govind]

" I'd be happy to look into this further if you attach a mat file containing the inputs to trainNetwork."

Tnk u.i have attached what u requested.

[Adam Danz]

The reason for your error is clearly described in the documentation. It just requires a little digging to understand the problem. I'll point you in the right direction so you can become more familiar with the documentation.

For table inputs there are actually two syntaxes. In my answer I only provided the first one below but that link would have shown you the second one, too.

net = trainNetwork(tbl,layers,options)

net = trainNetwork(tbl,responseName,layers,options)

In both syntaxes, it is expected that the predictor variables are in the first column of the table (see Walter's comment above). The second syntax specifies responses as the second input. If the first syntax is used instead, responses are expected to be in the second column of the table. This is all explained in this description of the table input (the link is to a different section of the same page as the previous links).

The syntax you're using fits neither description. Here are the first few rows of your table.

head(xx)

  8×30 table
      x1        x2        x3        x4       x5        x6       x7       x8        x9       x10       x11      x12       x13      x14       x15       x16       x17       x18       x19       x20       x21       x22      x23     x24       x25       x26       x27       x28      x29       x30  
    ______    ______    ______    ______    _____    ______    _____    _____    ______    ______    _____    ______    _____    ______    ______    ______    ______    ______    ______    ______    ______    ______    ___    ______    ______    ______    ______    _____    ______    ______
    138.47    51.655    97.827     27.98     0.91    124.71    2.666    3.064    41.928    197.76    1.582     1.396      0.2    32.638     1.017     0.381    51.626     2.273    -2.414    16.824    -0.277    258.73     2     67.435      2.15     0.444    46.062     1.24    -2.475     113.5
    160.94    68.768    103.23    48.146     -999      -999     -999    3.473     2.078    125.16    0.879     1.414     -999    42.014     2.039    -3.011    36.918     0.501     0.103    44.704    -1.916    164.55     1     46.226     0.725     1.158      -999     -999      -999    46.226
      -999    162.17    125.95    35.635     -999      -999     -999    3.148     9.336    197.81    3.776     1.414     -999    32.154    -0.705    -2.093    121.41    -0.953     1.052    54.283    -2.186    260.41     1     44.251     2.053    -2.028      -999     -999      -999    44.251
    143.91    81.417    80.943     0.414     -999      -999     -999     3.31     0.414    75.968    2.354    -1.285     -999    22.647    -1.655      0.01    53.321    -0.522      -3.1    31.082      0.06    86.062     0       -999      -999      -999      -999     -999      -999         0
    175.86    16.915    134.81    16.405     -999      -999     -999    3.891    16.405    57.983    1.056    -1.385     -999    28.209    -2.197    -2.231    29.774     0.798     1.569     2.723    -0.871    53.131     0       -999      -999      -999      -999     -999      -999         0
    89.744     13.55    59.149    116.34    2.636    284.58    -0.54    1.362    61.619    278.88    0.588     0.479    0.975    53.651     0.371     1.329    31.565    -0.884     1.857    40.735     2.237    282.85     3     90.547    -2.412    -0.653    56.165    0.224     3.106    193.66
    148.75    28.862    107.78    106.13    0.733    158.36    0.113    2.941     2.545    305.97    3.371     1.393    0.791     28.85     1.113     2.409     97.24     0.675    -0.966    38.421    -1.443    294.07     2     123.01     0.864      1.45    56.867    0.131    -2.767    179.88
    154.92    10.418    94.714    29.169     -999      -999     -999    2.897     1.526    138.18    0.365    -1.305     -999      78.8     0.654     1.547     28.74     0.506    -1.347    22.275    -1.761     187.3     1     30.638    -0.715    -1.724      -999     -999      -999    30.638

[Chandini Govind]

" it is expected that the predictor variables are in the first column of the table ."

does this means for 30 features i have to extract them and train along with reponse?.doesn't it look tedious?

well after converting to cell array,i done this

 for i=1:30
     xx=table(ft(:,i),yy);
     net = trainNetwork(xx,layers,options);
 end

i get this error

Error using trainNetwork (line 150)
Invalid training data. For table input, sequences must be in the first column as file paths.

Error in study8 (line 38)
     net = trainNetwork(xx,layers,options); 

[Adam Danz]

Since you're having trouble with the table input syntax and since there are no examples for that syntax, why don't you try using the numeric array input, if possible?

net = trainNetwork(X,Y,layers,options)

[Chandini Govind]

because i'm getting the following error and couldn't resolve it.

Error using trainNetwork (line 150)
Invalid training data. Predictors must be a N-by-1 cell array of sequences, where N is the number of sequences. All
sequences must have the same feature dimension and at least one time step.

Error in study8 (line 37)
     net = trainNetwork(ft,yy,layers,options);

[Kenan]

For information there is an example for this method :

net = trainNetwork(tbl,responseName,layers,options)