Fixed render issues
Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
parent
7a4c932dcd
commit
11ed4e2b93
|
@ -206,7 +206,7 @@ fact I do want to reserve free time.
|
||||||
\textbf{Due date:} 20th March
|
\textbf{Due date:} 20th March
|
||||||
|
|
||||||
\begin{description}
|
\begin{description}
|
||||||
\item[Download SceneNet RGB-D to cvpc\{7,8\} computer]
|
\item[Download SceneNet RGB-D to cvpc\{7,8\} computer] \hfill \\
|
||||||
Requires external resource.
|
Requires external resource.
|
||||||
\end{description}
|
\end{description}
|
||||||
|
|
||||||
|
@ -215,14 +215,14 @@ fact I do want to reserve free time.
|
||||||
\textbf{Due date:} 5th April
|
\textbf{Due date:} 5th April
|
||||||
|
|
||||||
\begin{description}
|
\begin{description}
|
||||||
\item[Download pre-trained weights of SSD for MS COCO]
|
\item[Download pre-trained weights of SSD for MS COCO] \hfill \\
|
||||||
This is trivial. Takes not more than two hours.
|
This is trivial. Takes not more than two hours.
|
||||||
\item[Modify SSD Keras implementation to work inside masterthesis package]
|
\item[Modify SSD Keras implementation to work inside masterthesis package] \hfill \\
|
||||||
Should be possible to achieve within one day.
|
Should be possible to achieve within one day.
|
||||||
\item[Implement integration of SSD into masterthesis package]
|
\item[Implement integration of SSD into masterthesis package] \hfill \\
|
||||||
Implementing the glue code between the git submodule and
|
Implementing the glue code between the git submodule and
|
||||||
my own code. Should be doable within one day.
|
my own code. Should be doable within one day.
|
||||||
\item[Group SceneNet RGB-D classes to MS COCO classes]
|
\item[Group SceneNet RGB-D classes to MS COCO classes] \hfill \\
|
||||||
SceneNet contains more classes than COCO. Miller et al have
|
SceneNet contains more classes than COCO. Miller et al have
|
||||||
grouped, for example, various chair classes in SceneNet into
|
grouped, for example, various chair classes in SceneNet into
|
||||||
one chair class of COCO. This grouping involves researching
|
one chair class of COCO. This grouping involves researching
|
||||||
|
@ -231,15 +231,15 @@ fact I do want to reserve free time.
|
||||||
|
|
||||||
All in all this could take up a full day and perhaps slip
|
All in all this could take up a full day and perhaps slip
|
||||||
into a second one.
|
into a second one.
|
||||||
\item[Implement variant of SSD with dropout layers (Bayesian SSD)]
|
\item[Implement variant of SSD with dropout layers (Bayesian SSD)] \hfill \\
|
||||||
This is a rather trivial task as it only involves adding two
|
This is a rather trivial task as it only involves adding two
|
||||||
Keras dropout layers into SSD. Can be done in one hour.
|
Keras dropout layers into SSD. Can be done in one hour.
|
||||||
\item[Fine-tune vanilla SSD on SceneNet RGB-D]
|
\item[Fine-tune vanilla SSD on SceneNet RGB-D] \hfill \\
|
||||||
Requires external resource and length of required training is
|
Requires external resource and length of required training is
|
||||||
unknown. Due to two unknown factors (availability of resource,
|
unknown. Due to two unknown factors (availability of resource,
|
||||||
and length of training) this task can be considered a project
|
and length of training) this task can be considered a project
|
||||||
risk.
|
risk.
|
||||||
\item[Fine-tune Bayesian SSD on SceneNet RGB-D]
|
\item[Fine-tune Bayesian SSD on SceneNet RGB-D] \hfill \\
|
||||||
Similar remarks like the previous task.
|
Similar remarks like the previous task.
|
||||||
\end{description}
|
\end{description}
|
||||||
|
|
||||||
|
@ -259,30 +259,29 @@ so that the training time is used as efficiently as possible.
|
||||||
\textbf{Due date:} 12th April
|
\textbf{Due date:} 12th April
|
||||||
|
|
||||||
\begin{description}
|
\begin{description}
|
||||||
\item[Adapt GPND implementation for SceneNet RGB-D using MS COCO classes]
|
\item[Adapt GPND implementation for SceneNet RGB-D using COCO classes] \hfill \\
|
||||||
Requires research to figure out the exact architecture needed
|
Requires research to figure out the exact architecture needed
|
||||||
for a different data set. The code is not well
|
for a different data set. The code is not well
|
||||||
documented and some logical variables like image size are
|
documented and some logical variables like image size are
|
||||||
sometimes hard-coded, which makes this adaption difficult
|
sometimes hard-coded, which makes this adaption difficult
|
||||||
and error-prone.
|
and error-prone.
|
||||||
|
|
||||||
Furthermore, some trial-and-error regarding training successes
|
Furthermore, some trial-and-error regarding training successes
|
||||||
is likely needed, which makes this task a project risk.
|
is likely needed, which makes this task a project risk.
|
||||||
If the needed architecture was known the time to implement
|
If the needed architecture was known the time to implement
|
||||||
it would be at most one day. The uncertainty therefore
|
it would be at most one day. The uncertainty therefore
|
||||||
lies with the research part.
|
lies with the research part.
|
||||||
\item[Implement novelty score calculation for GPND]
|
\item[Implement novelty score calculation for GPND] \hfill \\
|
||||||
There is an implementation for this in the original
|
There is an implementation for this in the original
|
||||||
author's implementation. It would have to be ported
|
author's implementation. It would have to be ported
|
||||||
to Tensorflow and integrated into the package structure.
|
to Tensorflow and integrated into the package structure.
|
||||||
Takes likely one day or two.
|
Takes likely one day or two.
|
||||||
\item[Apply insights of GAN stability to GPND implementation]
|
\item[Apply insights of GAN stability to GPND implementation] \hfill \\
|
||||||
The insights from the GAN stability\footnote{\url{https://avg.is.tuebingen.mpg.de/publications/meschedericml2018}} research should be applied to
|
The insights from the GAN stability\footnote{\url{https://avg.is.tuebingen.mpg.de/publications/meschedericml2018}} research should be applied to
|
||||||
my GPND implementation. Requires research what, if any,
|
my GPND implementation. Requires research what, if any,
|
||||||
insights can be used for this thesis. The research is
|
insights can be used for this thesis. The research is
|
||||||
doable within one day and the application of it
|
doable within one day and the application of it
|
||||||
within another.
|
within another.
|
||||||
\item[Train GPND on SceneNet RGB-D]
|
\item[Train GPND on SceneNet RGB-D] \hfill \\
|
||||||
Requires external resource.
|
Requires external resource.
|
||||||
In contrast to the SSD network, there are no pre-trained
|
In contrast to the SSD network, there are no pre-trained
|
||||||
weights available for the GPND. Therefore it has to be
|
weights available for the GPND. Therefore it has to be
|
||||||
|
@ -304,32 +303,32 @@ the aggressive date I will work towards.
|
||||||
\textbf{Due date:} 10th May
|
\textbf{Due date:} 10th May
|
||||||
|
|
||||||
\begin{description}
|
\begin{description}
|
||||||
\item[Implement evaluation pipeline for vanilla SSD]
|
\item[Implement evaluation pipeline for vanilla SSD] \hfill \\
|
||||||
Involves the implementation of the evaluation steps
|
Involves the implementation of the evaluation steps
|
||||||
according to the chosen metrics. Takes likely two days.
|
according to the chosen metrics. Takes likely two days.
|
||||||
\item[Implement evaluation pipeline for Bayesian SSD]
|
\item[Implement evaluation pipeline for Bayesian SSD] \hfill \\
|
||||||
Involves the implementation of the evaluation steps
|
Involves the implementation of the evaluation steps
|
||||||
for the Bayesian variant. As more is has to be done,
|
for the Bayesian variant. As more is has to be done,
|
||||||
it will likely take three days.
|
it will likely take three days.
|
||||||
\item[Implement evaluation pipeline for vanilla SSD with GPND for novelty score]
|
\item[Implement evaluation pipeline for SSD with GPND for novelty score] \hfill \\
|
||||||
The implementation of this pipeline will take probably two
|
Implementation of the evaluation steps for my approach.
|
||||||
days.
|
It will probably take two days.
|
||||||
\item[Run vanilla SSD on test data]
|
\item[Run vanilla SSD on test data] \hfill \\
|
||||||
The trained network is run on the test data and the results
|
The trained network is run on the test data and the results
|
||||||
are stored. Requires external resource but should be
|
are stored. Requires external resource but should be
|
||||||
far quicker than the training and will probably be done
|
far quicker than the training and will probably be done
|
||||||
in two days at most.
|
in two days at most.
|
||||||
\item[Run Bayesian SSD on test data]
|
\item[Run Bayesian SSD on test data] \hfill \\
|
||||||
Similar remarks to previous task.
|
Similar remarks to previous task.
|
||||||
\item[Run vanilla SSD detections through GPND]
|
\item[Run vanilla SSD detections through GPND] \hfill \\
|
||||||
For my approach the SSD detections need to be run through
|
For my approach the SSD detections need to be run through
|
||||||
the GPND to have all the relevant data for evaluation.
|
the GPND to have all the relevant data for evaluation.
|
||||||
Requires external resource. Will take likely two days.
|
Requires external resource. Will take likely two days.
|
||||||
\item[Calculate evaluation metrics for vanilla SSD]
|
\item[Calculate evaluation metrics for vanilla SSD] \hfill \\
|
||||||
Takes one day.
|
Takes one day.
|
||||||
\item[Calculate evaluation metrics for Bayesian SSD]
|
\item[Calculate evaluation metrics for Bayesian SSD] \hfill \\
|
||||||
Takes one day.
|
Takes one day.
|
||||||
\item[Calculate evaluation metrics for vanilla SSD with GPND]
|
\item[Calculate evaluation metrics for vanilla SSD with GPND] \hfill \\
|
||||||
Takes one day
|
Takes one day
|
||||||
\end{description}
|
\end{description}
|
||||||
|
|
||||||
|
@ -341,19 +340,18 @@ happen on the CPU as all the data is already there by then.
|
||||||
|
|
||||||
\subsection*{Visualizations created}
|
\subsection*{Visualizations created}
|
||||||
|
|
||||||
\textbf{Due date:} 31st May
|
\textbf{Due date:} 31st May\\
|
||||||
|
|
||||||
I won't be able to work on the thesis between May 13th and May 26th
|
I won't be able to work on the thesis between May 13th and May 26th
|
||||||
due to the election campaign. I am involved into the campaign already
|
due to the election campaign. I am involved into the campaign already
|
||||||
as of this writing but I hope that up until May 10th both thesis
|
as of this writing but I hope that up until May 10th both thesis
|
||||||
and campaign can somewhat co-exist.
|
and campaign can somewhat co-exist.
|
||||||
|
|
||||||
The visualizations should be creatable within one week from May 27th
|
The visualizations should be creatable within one week from May 27th
|
||||||
to May 31st.
|
to May 31st.
|
||||||
|
|
||||||
\subsection*{Stretch goals}
|
\subsection*{Stretch goals}
|
||||||
|
|
||||||
\textbf{Due date:} 27th June
|
\textbf{Due date:} 27th June\\
|
||||||
|
|
||||||
As I mentioned earlier, there are no specific tasks for the
|
As I mentioned earlier, there are no specific tasks for the
|
||||||
stretch goals. If the critical path is finished by the end
|
stretch goals. If the critical path is finished by the end
|
||||||
|
@ -364,24 +362,24 @@ writing period.
|
||||||
|
|
||||||
\subsection*{Thesis writing}
|
\subsection*{Thesis writing}
|
||||||
|
|
||||||
\textbf{Due date:} 30th August
|
\textbf{Due date:} 30th August\\
|
||||||
|
|
||||||
A first complete draft of the thesis should be finished
|
A first complete draft of the thesis should be finished
|
||||||
at the latest by August 16th. The following week I am not
|
at the latest by August 16th. The following week I am not
|
||||||
able to work on the thesis but it can be used for feedback.
|
able to work on the thesis but it can be used for feedback.
|
||||||
|
|
||||||
The last week of August should allow for polishing of the
|
The last week of August should allow for polishing of the
|
||||||
thesis with a submission-ready candidate by August 30th.
|
thesis with a submission-ready candidate by August 30th.
|
||||||
|
|
||||||
\subsection*{Finishing touches}
|
\subsection*{Finishing touches}
|
||||||
|
|
||||||
\textbf{Due date:} 13th September
|
\begin{description}
|
||||||
|
\item[Due date:] 13th September
|
||||||
|
\end{description}
|
||||||
|
|
||||||
The submission requires three printed copies of the thesis,
|
The submission requires three printed copies of the thesis,
|
||||||
together with any digital components on a CD glued to the back
|
together with any digital components on a CD glued to the back
|
||||||
of the thesis. A non-editable CD ensures that the code submitted
|
of the thesis. A non-editable CD ensures that the code submitted
|
||||||
cannot be modified and will be exactly as submitted when reviewed.
|
cannot be modified and will be exactly as submitted when reviewed.
|
||||||
|
|
||||||
I will use these two weeks to print the copies and to make
|
I will use these two weeks to print the copies and to make
|
||||||
last publication steps for the code like improving the code
|
last publication steps for the code like improving the code
|
||||||
documentation and adding usage examples.
|
documentation and adding usage examples.
|
||||||
|
@ -401,3 +399,19 @@ The workload for the election campaign, in which I have an
|
||||||
organizational responsibility in addition to being a candidate
|
organizational responsibility in addition to being a candidate
|
||||||
myself, could come into conflict with the progress of the
|
myself, could come into conflict with the progress of the
|
||||||
thesis.
|
thesis.
|
||||||
|
|
||||||
|
Availability of the external resource can hinder the progress
|
||||||
|
and delay steps of the thesis. In such a case dependent
|
||||||
|
tasks cannot commence until the earlier task has been finished,
|
||||||
|
resulting in an overall delay of the thesis.
|
||||||
|
|
||||||
|
To deal with these risks, I have planned for one whole month
|
||||||
|
of buffer time that can account for many delays. Furthermore,
|
||||||
|
the writing time is intentionally that long as it is difficult
|
||||||
|
to predict how inspired I will be. I know from my bachelor
|
||||||
|
thesis that on some days it can be many pages you write and
|
||||||
|
on others you might barely make one page progress.
|
||||||
|
|
||||||
|
I would argue that the thesis success is largely dependent on
|
||||||
|
the first part of the work as it can make or break it. Once
|
||||||
|
the technical part is done, the way forward should be downhill.
|
||||||
|
|
Loading…
Reference in New Issue