I'm fairly new to machine learning, so I don't know the correct terminology, but I converted two categorical columns into numbers the following way. These columns are part of my features inputs, akin to the sex column in the titanic database.
(They are not the target data y which I have already created)
changed p_changed
Date
2010-02-17 0.477182 0 0
2010-02-18 0.395813 0 0
2010-02-19 0.252179 1 1
2010-02-22 0.401321 0 1
2010-02-23 0.519375 1 1
Now the rest of my data Xlooks something like this
Open High Low Close Volume Adj Close log_return \
Date
2010-02-17 2.07 2.07 1.99 2.03 219700.0 2.03 -0.019513
2010-02-18 2.03 2.03 1.99 2.03 181700.0 2.03 0.000000
2010-02-19 2.03 2.03 2.00 2.02 116400.0 2.02 -0.004938
2010-02-22 2.05 2.05 2.02 2.04 188300.0 2.04 0.009852
2010-02-23 2.05 2.07 2.01 2.05 255400.0 2.05 0.004890
close_open Daily_Change 30_Avg_Vol 20_Avg_Vol 15_Avg_Vol \
Date
2010-02-17 0.00 -0.04 0.909517 0.779299 0.668242
2010-02-18 0.00 0.00 0.747470 0.635404 0.543015
2010-02-19 0.00 -0.01 0.508860 0.417706 0.348761
2010-02-22 0.03 -0.01 0.817274 0.666903 0.562414
2010-02-23 0.01 0.00 1.078411 0.879007 0.742730
As you can see the rest of my data is continuous (containing many variables) as opposed to the two categorical columns which only have two values (0 and 1).
I was planning to preprocess all this data in one shot via this simple preprocess method
X_scaled = preprocessing.scale(X)
I was wondering if this is mistake? Is there something else I need to do to the categorical values before using this simple preprocessing?
EDIT: I tried two ways; First I tried scaling the full data, including the categorical data converted to 1's and 0's.
Full_X = OPK_df.iloc[:-5, 0:-5]
Full_X_scaled = preprocessing.scale( Full_X) # First way, which scales everything in one shot.
Then I tried dropping the last two columns, scaling, then adding the dropped columns via this code.
X =OPK_df.iloc[:-5, 0:-7] # Here I'm dropping both -7 while originally the offset was only till -5, which means two extra columns were dropped.
I created another dataframe which has those two columns I dropped
x2 =OPK_df.iloc[:-5, -7:-5]
x2 = np.array(x2) # convert it to an array
# preprocessing the data without last two columns
from sklearn import preprocessing
X_scaled = preprocessing.scale(X)
# Then concact the X_scaled with x2(originally dropped columns)
X =np.concatenate((X_scaled, x2), axis =1)
#Creating a classifier
from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier(n_neighbors=5)
knn2 = KNeighborsClassifier(n_neighbors=5)
knn.fit(X_scaled, y)
knn2.fit(X,y)
knn.score(Full_X_scaled, y)
0.71396522714526078
knn2.score(X, y)
0.71789119461581608
So there is a higher score when I do indeed drop the two columns during standarization.
You're doing pretty well so far. Do not scale your classification data. Since those appear to be binary classifications, think of this as "Yes" and "No". What does it mean to scale these?
Even worse, consider that you might have classifications such as flower types: you've coded Zinnia=0, Rose=1, Orchid=2, etc. What does it meant to scale those? It doesn't make any sense to re-code these as Zinnia=-0.257, Rose=+0.448, etc.
Scaling your input data is the necessary part: it keeps the values within comparable ranges (mathematical influence), allowing you to readily use a single treatment for your loss function. Otherwise, the feature with the largest spread of values would have the greatest influence on training, until your model's weights learned how to properly discount the large values.
For your beginning explorations, don't do any other preprocessing: just scale the input data and start your fitting exercises.
Related
Summary & Questions
I'm using liblinear 2.30 - I noticed a similar issue in prod, so I tried to isolate it through a simple reduced training with 2 classes, 1 train doc per class, 5 features with same weight in my vocabulary and 1 simple test doc containing only one feature which is present only in class 2.
a) what's the feature value being used for?
b) I wanted to understand why this test document containing a single feature which is only present in one class is not being strongly predicted into that class?
c) I'm not expecting to have different values per features. Is there any other implications by increasing each feature value from 1 to something-else? How can I determine that number?
d) Could my changes affect other more complex trainings in a bad way?
What I tried
Below you will find data related to a simple training (please focus on feature 5):
> cat train.txt
1 1:1 2:1 3:1
2 2:1 4:1 5:1
> train -s 0 -c 1 -p 0.1 -e 0.01 -B 0 train.txt model.bin
iter 1 act 3.353e-01 pre 3.333e-01 delta 6.715e-01 f 1.386e+00 |g| 1.000e+00 CG 1
iter 2 act 4.825e-05 pre 4.824e-05 delta 6.715e-01 f 1.051e+00 |g| 1.182e-02 CG 1
> cat model.bin
solver_type L2R_LR
nr_class 2
label 1 2
nr_feature 5
bias 0
w
0.3374141436539016
0
0.3374141436539016
-0.3374141436539016
-0.3374141436539016
0
And this is the output of the model:
solver_type L2R_LR
nr_class 2
label 1 2
nr_feature 5
bias 0
w
0.3374141436539016
0
0.3374141436539016
-0.3374141436539016
-0.3374141436539016
0
1 5:10
Below you will find my model's prediction:
> cat test.txt
1 5:1
> predict -b 1 test.txt model.bin test.out
Accuracy = 0% (0/1)
> cat test.out
labels 1 2
2 0.416438 0.583562
And here is where I'm a bit surprised because of the predictions being just [0.42, 0.58] as the feature 5 is only present in class 2. Why?
So I just tried with increasing the feature value for the test doc from 1 to 10:
> cat newtest.txt
1 5:10
> predict -b 1 newtest.txt model.bin newtest.out
Accuracy = 0% (0/1)
> cat newtest.out
labels 1 2
2 0.0331135 0.966887
And now I get a better prediction [0.03, 0.97]. Thus, I tried re-compiling my training again with all features set to 10:
> cat newtrain.txt
1 1:10 2:10 3:10
2 2:10 4:10 5:10
> train -s 0 -c 1 -p 0.1 -e 0.01 -B 0 newtrain.txt newmodel.bin
iter 1 act 1.104e+00 pre 9.804e-01 delta 2.508e-01 f 1.386e+00 |g| 1.000e+01 CG 1
iter 2 act 1.381e-01 pre 1.140e-01 delta 2.508e-01 f 2.826e-01 |g| 2.272e+00 CG 1
iter 3 act 2.627e-02 pre 2.269e-02 delta 2.508e-01 f 1.445e-01 |g| 6.847e-01 CG 1
iter 4 act 2.121e-03 pre 1.994e-03 delta 2.508e-01 f 1.183e-01 |g| 1.553e-01 CG 1
> cat newmodel.bin
solver_type L2R_LR
nr_class 2
label 1 2
nr_feature 5
bias 0
w
0.19420510395364846
0
0.19420510395364846
-0.19420510395364846
-0.19420510395364846
0
> predict -b 1 newtest.txt newmodel.bin newtest.out
Accuracy = 0% (0/1)
> cat newtest.out
labels 1 2
2 0.125423 0.874577
And again predictions were still ok for class 2: 0.87
a) what's the feature value being used for?
Each instance of n features is considered as a point in an n-dimensional space, attached with a given label, say +1 or -1 (in your case 1 or 2). A linear SVM tries to find the best hyperplane to separate those instance into two sets, say SetA and SetB. A hyperplane is considered better than other roughly when SetA contains more instances labeled with +1 and SetB contains more those with -1. i.e., more accurate. The best hyperplane is saved as the model. In your case, the hyperplane has formulation:
f(x)=w^T x
where w is the model, e.g (0.33741,0,0.33741,-0.33741,-0.33741) in your first case.
Probability (for LR) formulation:
prob(x)=1/(1+exp(-y*f(x))
where y=+1 or -1. See Appendix L of LIBLINEAR paper.
b) I wanted to understand why this test document containing a single feature which is only present in one class is not being strongly predicted into that class?
Not only 1 5:1 gives weak probability such as [0.42,0.58], if you predict 2 2:1 4:1 5:1 you will get [0.337417,0.662583] which seems that the solver is also not very confident about the result, even the input is exactly the same as the training data set.
The fundamental reason is the value of f(x), or can be simply seen as the distance between x and the hyperplane. It can be 100% confident x belongs to a certain class only if the distance is infinite large (see prob(x)).
c) I'm not expecting to have different values per features. Is there any other implications by increasing each feature value from 1 to something-else? How can I determine that number?
TL;DR
Enlarging both training and test set is like having a larger penalty parameter C (the -c option). Because larger C means a more strict penalty on error, intuitively speaking, the solver has more confidence with the prediction.
Enlarging every feature of the training set is just like having a smaller C.
Specifically, logistic regression solves the following equation for w.
min 0.5 w^T w + C ∑i log(1+exp(−yi w^T xi))
(eq(3) of LIBLINEAR paper)
For most instance, yi w^T xi is positive and larger xi implies smaller ∑i log(1+exp(−yi w^T xi)).
So the effect is somewhat similar to having a smaller C, and a smaller C implies smaller |w|.
On the other hand, enlarging the test set is the same as having a large |w|. Therefore, the effect of enlarging both training and test set is basically
(1). Having smaller |w| when training
(2). Then, having larger |w| when testing
Because the effect is more dramatic in (2) than (1), overall, enlarging both training and test set is like having a larger |w|, or, having a larger C.
We can run on the data set and multiply every features by 10^12. With C=1, we have the model and probability
> cat model.bin.m1e12.c1
solver_type L2R_LR
nr_class 2
label 1 2
nr_feature 5
bias 0
w
3.0998430106024949e-12
0
3.0998430106024949e-12
-3.0998430106024949e-12
-3.0998430106024949e-12
0
> cat test.out.m1e12.c1
labels 1 2
2 0.0431137 0.956886
Next we run on the original data set. With C=10^12, we have the probability
> cat model.bin.m1.c1e12
solver_type L2R_LR
nr_class 2
label 1 2
nr_feature 5
bias 0
w
3.0998430101989314
0
3.0998430101989314
-3.0998430101989314
-3.0998430101989314
0
> cat test.out.m1.c1e12
labels 1 2
2 0.0431137 0.956886
Therefore, because larger C means more strict penalty on error, so intuitively the solver has more confident with prediction.
d) Could my changes affect other more complex trainings in a bad way?
From (c) we know your changes is like having a larger C, and that will result in a better training accuracy. But it almost can be sure that the model is over fitting the training set when C goes too large. As a result, the model cannot endure the noise in training set and will perform badly in test accuracy.
As for finding a good C, a popular way is by cross validation (-v option).
Finally,
it may be off-topic but you may want to see how to pre-process the text data. It is common (e.g., suggested by the author of liblinear here) to instance-wise normalize the data.
For document classification, our experience indicates that if you normalize each document to unit length, then not only the training time is shorter, but also the performance is better.
I am working on a dataset which has a feature that has multiple categories for a single example.
The feature looks like this:-
Feature
0 [Category1, Category2, Category2, Category4, Category5]
1 [Category11, Category20, Category133]
2 [Category2, Category9]
3 [Category1000, Category1200, Category2000]
4 [Category12]
The problem is similar to the this question posted:- Encode categorical features with multiple categories per example - sklearn
Now, I want to vectorize this feature. One solution is to use MultiLabelBinarizer as suggested in the answer of the above similar question. But, there are around 2000 categories, which results into a sparse and very high dimentional encoded data.
Is there any other encoding that can be used? Or any possible solution for this problem. Thanks.
Given an incredibly sparse array one could use a dimensionality reduction technique such as PCA (Principal component analysis) to reduce the feature space to the top k features that best describe the variance.
Assuming the MultiLabelBinarizered 2000 features = X
from sklearn.decomposition import PCA
k = 5
model = PCA(n_components = k, random_state = 666)
model.fit(X)
Components = model.predict(X)
And then you can use the top K components as a smaller dimensional feature space that can explain a large portion of the variance for the original feature space.
If you want to understand how well the new smaller feature space describes the variance you could use the following command
model.explained_variance_
In many cases when I encountered the problem of too many features being generated from a column with many categories, I opted for binary encoding and it worked out fine most of the times and hence is worth a shot for you perhaps.
Imagine you have 9 features, and you mark them from 1 to 9 and now binary encode them, you will get:
cat 1 - 0 0 0 1
cat 2 - 0 0 1 0
cat 3 - 0 0 1 1
cat 4 - 0 1 0 0
cat 5 - 0 1 0 1
cat 6 - 0 1 1 0
cat 7 - 0 1 1 1
cat 8 - 1 0 0 0
cat 9 - 1 0 0 1
This is the basic intuition behind Binary Encoder.
PS: Given that 2 power 11 is 2048 and you may have 2000 categories or so, you can reduce your categories to 11 feature columns instead of many (for example, 1999 in the case of one-hot)!
I also encountered these same problems but I solved using Countvectorizer from sklearn.feature_extraction.text just by giving binary=True, i.e CounterVectorizer(binary=True)
for my thesis I have to calculate the number of workers at risk of substitution by machines. I have calculated the probability of substitution (X) and the number of employee at risk (Y) for each occupation category. I have a dataset like this:
X Y
1 0.1300 0
2 0.1000 0
3 0.0841 1513
4 0.0221 287
5 0.1175 3641
....
700 0.9875 4000
I tried to plot a histogram with this command:
hist(dataset1$X,dataset1$Y,xlim=c(0,1),ylim=c(0,30000),breaks=100,main="Distribution",xlab="Probability",ylab="Number of employee")
But I get this error:
In if (freq) x$counts else x$density
length > 1 and only the first element will be used
Can someone tell me what is the problem and write me the right command?
Thank you!
It is worth pointing out that the message displayed is a Warning message, and should not prevent the results being plotted. However, it does indicate there are some issues with the data.
Without the full dataset, it is not 100% obvious what may be the problem. I believe it is caused by the data not being in the correct format, with two potential issues. Firstly, some values have a value of 0, and these won't be plotted on the histogram. Secondly, the observations appear to be inconsistently spaced.
Histograms are best built from one of two datasets:
A dataframe which has been aggregated grouped into consistently sized bins.
A list of values X which in the data
I prefer the second technique. As originally shown here The expandRows() function in the package splitstackshape can be used to repeat the number of rows in the dataframe by the number of observations:
set.seed(123)
dataset1 <- data.frame(X = runif(900, 0, 1), Y = runif(900, 0, 1000))
library(splitstackshape)
dataset2 <- expandRows(dataset1, "Y")
hist(dataset2$X, xlim=c(0,1))
dataset1$bins <- cut(dataset1$X, breaks = seq(0,1,0.01), labels = FALSE)
I have found that during training my model vw shows very big (much more than my features count ) feature number count in it's log.
I have tried to reproduce it using some small example:
simple.test:
-1 | 1 2 3
1 | 3 4 5
then "vw simple.test" command says that it have used 8 features. +one feature is constant but what are the other ? And in my real exmaple difference between my features and features used in wv is abot x10 more.
....
Num weight bits = 18
learning rate = 0.5
initial_t = 0
power_t = 0.5
using no cache
Reading datafile = t
num sources = 1
average since example example current current current
loss last counter weight label predict features
finished run
number of examples = 2
weighted example sum = 2
weighted label sum = 3
average loss = 1.9179
best constant = 1.5
total feature number = 8 !!!!
total feature number displays a sum of feature counts from all observed examples. So it's 2*(3+1 constant)=8 in your case. The number of features in current example is displayed in current features column. Note that only 2^Nth example is printed on screen by default. In general observations can have unequal number of features.
I want to find the standard deviation:
Minimum = 5
Mean = 24
Maximum = 84
Overall score = 90
I just want to find out my grade by using the standard deviation
Thanks,
A standard deviation cannot in general be computed from just the min, max, and mean. This can be demonstrated with two sets of scores that have the same min, and max, and mean but different standard deviations:
1 2 4 5 : min=1 max=5 mean=3 stdev≈1.5811
1 3 3 5 : min=1 max=5 mean=3 stdev≈0.7071
Also, what does an 'overall score' of 90 mean if the maximum is 84?
I actually did a quick-and-dirty calculation of the type M Rad mentions. It involves assuming that the distribution is Gaussian or "normal." This does not apply to your situation but might help others asking the same question. (You can tell your distribution is not normal because the distance from mean to max and mean to min is not close). Even if it were normal, you would need something you don't mention: the number of samples (number of tests taken in your case).
Those readers who DO have a normal population can use the table below to give a rough estimate by dividing the difference of your measured minimum and your calculated mean by the expected value for your sample size. On average, it will be off by the given number of standard deviations. (I have no idea whether it is biased - change the code below and calculate the error without the abs to get a guess.)
Num Samples Expected distance Expected error
10 1.55 0.25
20 1.88 0.20
30 2.05 0.18
40 2.16 0.17
50 2.26 0.15
60 2.33 0.15
70 2.38 0.14
80 2.43 0.14
90 2.47 0.13
100 2.52 0.13
This experiment shows that the "rule of thumb" of dividing the range by 4 to get the standard deviation is in general incorrect -- even for normal populations. In my experiment it only holds for sample sizes between 20 and 40 (and then loosely). This rule may have been what the OP was thinking about.
You can modify the following python code to generate the table for different values (change max_sample_size) or more accuracy (change num_simulations) or get rid of the limitation to multiples of 10 (change the parameters to xrange in the for loop for idx)
#!/usr/bin/python
import random
# Return the distance of the minimum of samples from its mean
#
# Samples must have at least one entry
def min_dist_from_estd_mean(samples):
total = 0
sample_min = samples[0]
for sample in samples:
total += sample
sample_min = min(sample, sample_min)
estd_mean = total / len(samples)
return estd_mean - sample_min # Pos bec min cannot be greater than mean
num_simulations = 4095
max_sample_size = 100
# Calculate expected distances
sum_of_dists=[0]*(max_sample_size+1) # +1 so can index by sample size
for iternum in xrange(num_simulations):
samples=[random.normalvariate(0,1)]
while len(samples) <= max_sample_size:
sum_of_dists[len(samples)] += min_dist_from_estd_mean(samples)
samples.append(random.normalvariate(0,1))
expected_dist = [total/num_simulations for total in sum_of_dists]
# Calculate average error using that distance
sum_of_errors=[0]*len(sum_of_dists)
for iternum in xrange(num_simulations):
samples=[random.normalvariate(0,1)]
while len(samples) <= max_sample_size:
ave_dist = expected_dist[len(samples)]
if ave_dist > 0:
sum_of_errors[len(samples)] += \
abs(1 - (min_dist_from_estd_mean(samples)/ave_dist))
samples.append(random.normalvariate(0,1))
expected_error = [total/num_simulations for total in sum_of_errors]
cols=" {0:>15}{1:>20}{2:>20}"
print(cols.format("Num Samples","Expected distance","Expected error"))
cols=" {0:>15}{1:>20.2f}{2:>20.2f}"
for idx in xrange(10,len(expected_dist),10):
print(cols.format(idx, expected_dist[idx], expected_error[idx]))
Yo can obtain an estimate of the geometric mean, sometimes called the geometric mean of the extremes or GME, using the Min and the Max by calculating the GME= $\sqrt{ Min*Max }$. The SD can be then calculated using your arithmetic mean (AM) and the GME as:
SD= $$\frac{AM}{GME} * \sqrt{(AM)^2-(GME)^2 }$$
This approach works well for log-normal distributions or as long as the GME, GM or Median is smaller than the AM.
In principle you can make an estimate of standard deviation from the mean/min/max and the number of elements in the sample. The min and max of a sample are, if you assume normality, random variables whose statistics follow from mean/stddev/number of samples. So given the latter, one can compute (after slogging through the math or running a bunch of monte carlo scripts) a confidence interval for the former (like it is 80% probable that the stddev is between 20 and 40 or something like that).
That said, it probably isn't worth doing except in extreme situations.