Visios| Hey! What Are You Looking At?

Team Updates

## NASA Space Apps Challenge Project##
# #
# Dataset loading
library(ranger)
library(caret)
library(data.table)
library(CRAN)
library(Rtsne)
url<-"https://heasarc.gsfc.nasa.gov/FTP/nicer/data/obs/2018_01/*/auxil/ni*.att.gz"
download.file(url, destfile="heasarcfiles.gz", mod="wb")
spacedata<-read.gz("heasarcfiles.gz", sheetIndex=1)
head(spacedata)
# Data exploration / cleaning #
dim(space_data)
head(space_data, 130)
tail(space_data, 130)
table(space_data$Class)
summary(space_data$Amount)
names(space_data)
var(space_data$Amount)
sd(space_data$Amount)
space_data %>%
mutate(id=n(nrows)),
mutate(Class=as.Integer)
# Data wrangling #
head(space_data)
space_data$Amount=scale(space_data$Amount)
NewData=space_data[,-c(1)]
head(NewData)
tsne_out<- Rtsne(as.matrix(select(space_data)),
pca=FALSE,
verbose=TRUE,
theta=0.35,
max_iter=2500,
Y_init=NULL,)
# Data modeling #
library(caTools)
set.seed(123)
data_sample= sample.split(NewData$Class,SplitRatio=0.80)
train_data= subset(NewData,data_sample==TRUE)
test_data= subset(NewData,data_sample==FALSE)
dim(train_data)
dim(test_data)
# Logistic regression model #
Logistic_Model= glm(Class~.,test_data,family=binomial())
summary(Logistic_Model)
plot(Logistic_Model)
library(pROC)
lr.predict<- predict(Logistic_Model,train_data, probability=TRUE)
auc.gbm= roc(test_data$Class, lr.predict, plot=TRUE, col="blue")
# Decision Tree model #
library(rpart)
library(rpart.plot)
decisionTree_model<- rpart(Class~. , space_data, method='class')
predicted_val<- predict(decisionTree_model, space_data, type='class')
probability<- predict(decisionTree_model, space_data, type='prob')
rpart.plot(decisionTree_model)
# Artificial Neural Network #
library(neuralnet)
ABE_model=neuralnet (Class~.,train_data,linear.output=FALSE)
plot(ABE_model)
predABE=compute(ABE_model,test_data)
resultABE=predABE$net.result
resultABE=ifelse(resultABE>0.5,1,0)
# Gradient boosting #
library(gbm, quietly=TRUE)
# Get the time to train the GBM model
system.time(
model_gbm<- gbm(Class~.
, distribution="bernoulli"
, data= rbind(train_data, test_data)
, n.trees=500
, interaction.depth=3
, n.minobsinnode=100
, shrinkage=0.01
, bag.fraction=0.5
, train.fraction= nrow(train_data) / (nrow(train_data) + nrow(test_data))
)
)
# Determine best iteration based on test data #
gbm.iter= gbm.perf(model_gbm, method="test")
model.influence= relative.influence(model_gbm, n.trees=gbm.iter, sort.=TRUE)
#Plot the gbm model #
plot(model_gbm)
# Plot and calculate AUC on test data
gbm_test= predict(model_gbm, newdata=test_data, n.trees=gbm.iter)
gbm_auc= roc(test_data$Class, gbm_test, plot=TRUE, col="red")
print(gbm_auc)
view raw NASASpaceAppsML2.R hosted with ❤ by GitHub
A
Abraham Mateos
.aladin-container{position:relative;border:1px solid #ddd;height:80%;}
.aladin-imageCanvas{position:absolute;z-index:1;left:0;top:0}
.aladin-catalogCanvas{position:absolute;z-index:2;left:0;top:0}
.aladin-reticleCanvas{position:absolute;z-index:3;left:0;top:0}
.aladin-logo-container{position:absolute;bottom:2px;right:5px;z-index:20;min-width:30px;max-width:80px}
.aladin-logo-small{padding:0;background:url(data:image/gif;base64,R0lGODlhIAAgAJEAAJIsLdEwJAdMmP///yH5BAkAAAMALAAAAAAgACAAAAjMAAcIHEiwoMGDCBMqXMiwocOHECMaFCCxYkKKAAoK2MiRo0UBAEKKFOkxYUaCIEMSHBlyo0OQCke6HHDyJEWBKgcG2MlzoEyFMAXyHNqTZsubNFGeHLDT4FCcLREGZUqwaFGRUk82FfqUaQCoSH0OCLqVqlCuX42u9Kl1a1qzXnGGVaozLdG6cpMWxOrVblm4AOYOTNn2L1efYZdu5Eu0cV6cE0fW7QqV4WK+CAMLPnhZMtvAEDmy/CkWMtCOHVFaXC2VtevXsGPLZhgQADs=);background-size:100%;background-repeat:no-repeat;padding-top:100%;}
.aladin-logo-large{padding:0;background:url('data:image/png;base64,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');background-size:100%;background-repeat:no-repeat;padding-top:58.45%;}
.aladin-location{z-index:20;position:absolute;top:0;padding:2px4px2px4px;background-color:rgba(250,250,250,0.5);font-size:11px}
.aladin-location-text{font-size:13px;margin-left:4px}
.aladin-measurement-div{z-index:147;position:absolute;bottom:0;background-color:rgba(250,250,250,0.8);font-family:monospace;font-size:12px;display:none;width:100%;overflow:auto}
.aladin-measurement-divtable{padding:2px4px2px4px;table-layout:fixed;white-space:nowrap}
.aladin-measurement-divthead{background-color:#e0e0e0;color:#000}
.aladin-measurement-divtd,.aladin-measurement-divth{padding:.3em.3em;white-space:nowrap;overflow:hidden;text-overflow:ellipsis;text-align:left}
.aladin-marker-measurement{max-height:230px;overflow-y:auto;overflow-x:hidden;font-family:monospace;font-size:11px;color:#000}
.aladin-marker-measurementtable{table-layout:fixed;width:100%}
.aladin-marker-measurementtabletrtd{word-wrap:break-word}
.aladin-marker-measurementtr:nth-child(even){background-color:#ddd}
.aladin-marker-measurementtd:first-child{font-weight:bold}
.aladin-fov{z-index:20;position:absolute;padding:0;font-size:12px;font-weight:bold;bottom:0;padding:2px;color:#321bdf;background-color:rgba(255,255,255,0.5)}
.aladin-maximize{position:absolute;top:6px;right:3px;z-index:20;width:30px;height:30px;background-image:url('data:image/gif;base64,R0lGODlhFgAWAOMJAAAAAAEBAQICAgMDAwUFBAUFBQcHBwgICAsLCv///////////////////////////yH5BAEKAA8ALAAAAAAWABYAAARm8MlJabr41i0T+CCQcJz3CYMwklsSEAUhspVnHEYw0x1w558VzYQTBXm24sgjJCUQykmT9dzxWlNq9vrwILYtqe/wRc6SBqszOE6DLZ/AT00FyKNcD4wQeLdQAiB+cCFHVxkZXA8RADs=');background-repeat:no-repeat;background-position:center center}
.aladin-layersControl-container{position:absolute;top:30px;left:4px;cursor:pointer;z-index:20;background:rgba(250,250,250,0.8);border-radius:4px}
.aladin-layersControl-container:hover{background:rgba(210,210,210,0.8)}
.aladin-layersControl{width:32px;height:34px;background-image:url('data:image/gif;base64,R0lGODlhGQAcAMIAAAAAADQ0NKahocvFxf///wAAAAAAAAAAACH5BAEKAAcALAAAAAAZABwAAANneLoH/hCwyaJ1dDrCuydY1gBfyYUaaZqosq0r+sKxNNP1pe98Hy2OgXBILLZGxWRSBlA6iZjgczrwWa9WIEDA7Xq/R8d3PGaSz97oFs0WYN9wiDZAr9vvYcB9v2fy/3ZqgIN0cYZYCQA7');background-repeat:no-repeat;background-position:center center}
.aladin-layerBox{top:50px;padding:4px4px10px10px}
.aladin-box{display:none;z-index:30;position:absolute;background:#eee;left:4px;border-radius:4px;font-size:14px;font-family:Verdana,Lucida,Arial;line-height:1.3;color:#222}
.aladin-gotoControl-container{position:absolute;top:68px;left:4px;cursor:pointer;z-index:20;background:rgba(245,245,245,0.8);border-radius:4px}
.aladin-gotoControl-container:hover{background:rgba(210,210,210,0.8)}
.aladin-gotoControl{width:32px;height:34px;background-image:url('data:image/png;base64,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');background-repeat:no-repeat;background-position:center center}
.aladin-gotoBox{top:48px;padding:4px}
.aladin-target-form{padding:5px}
.aladin-shareControl-container{position:absolute;top:106px;left:4px;cursor:pointer;z-index:20;background:rgba(250,250,250,0.8);border-radius:4px}
.aladin-shareControl-container:hover{background:rgba(210,210,210,0.8)}
.aladin-shareControl{width:42px;height:44px;background-image:url('data:image/png;base64,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');background-repeat:no-repeat;background-position:center center}
.aladin-shareBox{top:106px;padding:4px}
.aladin-target-forminput{width:140px;margin-left:5px}
.aladin-cb-listul{margin:0;padding-left:4px;list-style:none}
.aladin-cb-listlabel{display:inline}
.aladin-cb-listinput[type="checkbox"]{margin:3px}
.aladin-closeBtn{float:right;margin-top:002px0;cursor:pointer;color:#605f61;border:1px solid #aeaeae;border-radius:3px;background:#fff;font-size:15px;font-weight:bold;display:inline-block;line-height:0;padding:8px2px}
.aladin-closeBtn:hover{color:#201c22}
.aladin-label{font-weight:bold;padding-bottom:4px}
.aladin-box-separator{height:0;border-top:1px solid #bbb;margin:5px05px-4px}
.aladin-blank-separator{height:10px}
.aladin-restore{position:absolute;top:6px;right:3px;z-index:20;width:30px;height:30px;background-image:url('data:image/gif;base64,R0lGODlhFgAWAOMJAAAAAAEBAQICAgMDAwUFBAUFBQcHBwgICAsLCv///////////////////////////yH5BAEKAA8ALAAAAAAWABYAAARk8MlJ60zJapsAyhsFPp1xfKHUZeVhAOPWwYD5xjIABDacXrseLtQhFAiBIVEwEOh8P9VzquRgrhgrNhudTaHdJ1NQ1SQCRgL41zIE1sSa6w0361y0eqXtW7EReCNlIgh6XYMbEQA7');background-repeat:no-repeat;background-position:center center}
.aladin-fullscreen{position:fixed !important;z-index:9999;top:0;left:0;height:100%!important;width:100%!important;border:0!important;max-width:none !important;background:#fff;}
.aladin-zoomControl{z-index:20;position:absolute;top:50%;height:48px;right:8px;padding:0;margin:-24px000;font-weight:bold;font-size:18px;font-family:Arial,Arial,Arial}
.aladin-zoomControla{width:20px;height:20px;line-height:18px;display:block;background-color:rgba(250,250,250,0.8);margin:1px;text-align:center;border-radius:4px;border:1px solid #aaa;text-decoration:none;color:#222}
.aladin-zoomControla:hover{background-color:rgba(210,210,210,0.8)}
.aladin-surveySelection{width:100px}
.aladin-cmSelection{width:60px;margin-right:10px}
.aladin-layerIcon{width:4px;height:12px;background:#f00;display:inline-block}
.aladin-btn{display:inline-block;padding:6px8px;margin-bottom:0;font-size:12px;font-weight:normal;text-align:center;white-space:nowrap;vertical-align:middle;cursor:pointer;border:1px solid transparent;border-radius:3px;color:#fff;background-color:#428bca;border-color:#357ebd}
.aladin-btn-small{font-size:10px}
.aladin-button:hover{color:#fff;background-color:#3276b1;border-color:#285e8e}
.aladin-unknownObject{border:3px solid #f00}
.aladin-popup-container{z-index:25;position:absolute;width:150px;display:none;line-height:1.3}
.aladin-popup{font-family:Verdana,Lucida,Arial;font-size:13px;background:#fff;border:1px solid #bbb;border-radius:4px;padding:4px;top:80px;left:110px}
.aladin-popup-arrow{display:block;border-color:#fff transparent transparent;border-style:solid;border-width:12px;width:0;height:0;margin-top:-1px;margin-left:auto;margin-right:auto;}
.aladin-popupTitle{font-weight:bold}
.aladin-layer-label{padding:04px04px;color:#ddd;border-bottom-left-radius:8px;border-top-left-radius:8px;border-bottom-right-radius:8px;border-top-right-radius:8px;cursor:pointer}
view raw WebsiteStructure.css hosted with ❤ by GitHub
A
Abraham Mateos
defcompute_probabilities(X, betha, image_selection):
"""
Computes, for each datapoint X[i], the probability that X[i] is labeled as j
for j = 0, 1, ..., k-1
Args:
X = (n, d) NumPy array (n datapoints within the datasets from Agencies each with d features)
betha = (k, d) NumPy array, where row j represents the parameters of our model for label j
image_selection = the temperature parameter of softmax function (scalar)
Returns:
H = (k, n) NumPy array, where each entry H[j][i] is the probability that X[i] (image) is labeled as j
and included inside the list of results after the user's choice
"""
itemp=1/image_selection
selection_channel==2*X^i
selection_channel=selection_channel%>%sum(itemp+betha)
dot_products=itemp*betha.dot(X.T)
max_of_columns=dot_products.max(axis=0)
shifted_dot_products=dot_products-max_of_columns
exponentiated=np.exp(shifted_dot_products)
col_sums=exponentiated.sum(axis=0)
returnexponentiated/col_sums
view raw MLCodeUserMachineSelection.py hosted with ❤ by GitHub
A
Abraham Mateos