machine-learning


What is happen when the number of file smaller than batch in HDF5


I have a hdf5 layer that read the information from the list.txt as
layer {
name: "data"
type: "HDF5Data"
top: "data"
top: "label"
include {
phase: TEST
}
hdf5_data_param {
source: "./list.txt"
batch_size: 4
shuffle: true
}
}
where list.txt contains two path files
/home/user/file1.h5
/home/user/file2.h5
while the batch size is 4. What is happen with above code? Can the data choose 4 files to feed the network?
You have two hdf5 files, but each file may contain more than a single training example. Thus, effectively, you may have far more than batch_size: 4 examples.
Caffe does not really cares about the actual number of training examples: when it finishes to process all the examples (aka "epoch") it simply starts over reading the samples again. Caffe cycles through all the samples until number of training/testing iteration is reached.

Related Links

How to handle conditional features with SVM?
Automatic text / HTML annotation / highlighting
SMOTE oversampling and cross-validation
How to encode dependency path as a feature for classification?
Can Machine Learning be used for Natural Language Understanding
How to visualize telecom churn analysis using hostogram or scatterplot in ipython
weka multiple vectors for same instance
Why is linear regression called 'linear'?
scikitlearn - how to model a single features composed of multiple independant values
Evaluating the value of the multipliers of L1 and L2 type
Inference in structured prediction
What is Sequence to Sequence Learning with Neural Networks
Using Dense Training Data for Prediction on Sparse Testing Data for SVD yields poor performance
Difference between random forest and random tree algorithm
When doing cross validation, what changes if you ensure that the class distribution in the training and test set is equal to the whole set?
How to emphasize the importance of a title in an article in text mining?

Categories

HOME
centos
inno-setup
sitecore8
magento-1.9
ignite
owasp
set
entities
cakephp-3.0
attributes
conditional
mips32
asp.net-web-api2
xhtml
amazon-cloudsearch
ccache
batch-normalization
emu8086
sdl
uidocumentinteraction
clang-format
react-boilerplate
gulp-sass
google-nativeclient
revit
webcam
xml-configuration
size-classes
postgresql-8.3
pdb-files
aspose-cells
dos
card-flip
lift
fluent-nhibernate
instrumentation
smo
javassist
oltp
istanbul
code-signing-entitlements
express-stormpath
reportviewer2008
nested-function
jodconverter
virtual-directory
nokia
arithmetic-expressions
rouge
oql
wns
struts2-jquery-grid
r1soft
scrapyd
cosmos
stunnel
gridsplitter
ti-nspire
euclidean-distance
acl2
cpanm
.a
ironscheme
uploadifive
class-relationship
typescript1.4
nxt-python
sharepoint-clientobject
out
kiwi
n-tier
apparmor
chaplinjs
microsoft-commerce-server
stubs
problem-steps-recorder
websitespark
ereg
facebook-events
mysql-5.0
ejbql
memory-mapped-files
osx-gatekeeper
snk
ipod-nano
microsoft.build

Resources

Encrypt Message



code
soft
python
ios
c
html
jquery
cloud
mobile