No user data is collected.
Right click on pal6 -> properties -> language -> simplified Chinese Then its done! Its that easy! Dont download the installers just DIY it as those usually contains malware.
Make sure you have pandas and seaborn installed
fig, (ax0,ax1) = plt.subplots(1, 2, sharex=True, sharey=True)
cbar_ax = fig.add_axes([.91,.3,.03,.4])
sns.heatmap(pd1.corr(),ax=ax0,cbar=True,vmin=-1,vmax=1,cbar_ax = cbar_ax)
sns.heatmap(pd2.corr(),ax=ax1,cbar=True,vmin=-1,vmax=1,cbar_ax = cbar_ax)
#saving figure for publication if needed
thats it! I guess the most important thing here is
cbar_ax = fig.add_axes([.91,.3,.03,.4]) and make sure you have a fixed vmin and vmax.
Normally for us, signing PDFs should be easy. But thats why you come here.
The trick for the article is to help you get an unsignable PDF signed electronically and relatively unpainful. Follow the steps below you will get your pdf signed, and sent back. Assuming you have ghostscript, gimp installed, and assuming you are using mac. (both are free btw)
if you dont know how to install ghostscript and gimp, refer to my other post
install ghostscript and gimp
- You should do as much possible to fill in the information, up till you cannot fill any more. Assuming you have ghostscript, gimp installed. (both are free btw)
At this point, save your pdf and then use the command (assuming your pdf file is called convert.pdf)
gs -q -dSafer -dBatch -dNOPAUSE -sDEVICE=tiff32nc -r300 -sOutputFile=convert.tif convert.pdf -c quit
Your converted picture file will be called convert.tif and it is in 300 dpi.
- Now open the convert.tif in gimp, sign using the draw tool. When you are finished, go to file->export and export as result.pdf (type it is easy).
- Now that you finished, close all windows, send the file to the email address requesting the signed pdf. Done. You have saved a tree branch for the earth. LOL
I made a combined weak classifier model, needed a custom estimator and custom scorer. I went through a few stack overflow articles however none actually targeted specifically for cross validation in sklearn.
Then I figured I would try to implement baseestimator class, and make my own scorer. It WORKED. :>
Therefore, I am posting instructions here on how to use it, hopefully its gonna be useful to you.
- Write your own estimator class, just make sure to implement base estimator (or extend I am not sure how this works in python but its similar. base estimator is like an interface or abstract class provides basic functionalities for estimator)
- Write your loss function or gain function, and then make your own scorer
- Use the sklearn api to do cross validation. Using whatever you have created in 1 and 2.
Code: Please read comments. Important.
#create a custom estimator class
#Keep in mind. This is just a simplified version. You can treat it as any other class, just make sure the signitures should stay same, or you should add default value to other parameters
from sklearn.base import BaseEstimator
from sklearn import tree
from sklearn.cluster import KMeans
import numpy as np
from sklearn.cluster import KMeans
#Kmeans clustering model
__clusters = None
#decision tree model
__tree = None
__X = None
__y = None
#columns selected from pandas dataframe.
__columns = None
def fit(self, X, y, **kwargs):
result_kmeans = self.__clusters.predict(X)
result_tree = self.__tree.predict(X)
result = result_tree
clusters = KMeans(n_clusters=4, random_state=0).fit(X)
#the error center should have the lowest number of labels.(implementation not shown here)
self.__clusters = clusters
temp_tree = tree.DecisionTreeClassifier(criterion='entropy',max_depth=3)
self.__tree = temp_tree
Now we have our class. We need to build hit/loss function:
#again, feel free to change any thing in the hit function. As long as the function signature remain the same.
def seg_tree_hit_func(ground_truth, predictions):
total_hit = 0
total_number = 0
for i in xrange(len(predictions)):
total_hit += (1-abs(ground_truth[i]-predictions[i]))
print 'skipped: ',len(predictions)- total_number,'/',len(predictions),'instances'
return total_hit/total_number if total_number!=0 else 0
Now we still need to build scorer.
from sklearn.metrics.scorer import make_scorer
#make our own scorer
score = make_scorer(seg_tree_hit_func, greater_is_better=True)
We have our scorer, our estimator, and so we can start doing cross-validation task:
#change the 7 to whatever fold validation you are running.
scores = cross_val_score(custom_classifier(), X, Y, cv=7, scoring=score)
There it is! You have your own scorer and estimator, and you can use sklearn api to plug it in anything from sklearn easily.
Hope this helps.
I got really frustrated when trying to convert my charts and excel worksheets into publishable content for my paper. Went through several stack overflow questions and used ghostscript to make it possible.
All softwares used in this post are free — $0. It is ideal for Ph.D students or home use for publishing.
- Install ghostscript, using homebrew just run brew install GhostScript and you are good.
- Install Gimp. From homebrew all you have to do is: brew install Caskroom/cask/gimp
- Open your excel worksheet. Set the page setup to be vertical or horizontal, select your area as print area, and then save as pdf. Make sure you save the workbook as pdf. E.g. now I have saved the workbook to be figure1.pdf
- In your command line, type
gs -q -dSafer -dBatch -dNOPAUSE -sDEVICE=tiff32nc -r300 -sOutputFile=figure1.tif figure1.pdf -c quit
- Open the figure1.tif in Gimp, select the area you want, and then choose image -> fit canvas to selection
- In file->export as simple export the file as figure1.tiff you are all set. If you look at the properties it has 300dpi.
You can easily use bash to do batch processing for pdf files, but for more accurate image selection you still need gimp.
Hope this will help other people!
docker rm `docker ps --no-trunc -aq`
Mac .bashrc is not working. This happens when I get my new mac because it does not source the virtualenvwrapper I installed.
.bash_profile is the default bashrc on mac os.
I grew tired of using a shared serve resource where everyone is submitting jobs. I decided to build my own home server.
Used ubuntu 14.04 using vmware 300G disk space, 4 cpu cores, 8G memory with all virtual technologies on.
1. Threading package actually slows down your system because of GIL
2. Multithreading packages make use of processes and can have the expected speed gain
If you want to pass global dictionary, use Manager.dict()