Skip to content

Latest commit

 

History

History
549 lines (307 loc) · 13.5 KB

timelog.md

File metadata and controls

549 lines (307 loc) · 13.5 KB

Timelog

  • Deep neural networks for classification of multispectral images
  • Niklas Lindorfer
  • 2265846l
  • Professor Roderick Murray-Smith

Guidance

  • This file contains the time log for your project. It will be submitted along with your final dissertation.
  • YOU MUST KEEP THIS UP TO DATE AND UNDER VERSION CONTROL.
  • This timelog should be filled out honestly, regularly (daily) and accurately. It is for your benefit.
  • Follow the structure provided, grouping time by weeks. Quantise time to the half hour.

Week 1

25 Sept 2019

  • 0.5 hour Set up project repository on GitHub
  • 1.5 hours Downloaded LaTeX template and set up CD script on GitHub
  • 1.0 hour Set up timelog, plan, milestones, etc.

26 Sept 2019

  • 3.0 hours Reading and research (100 page ML book)

Week 2

01 Oct 2019

  • 0.5 hours Updated project plan
  • 1.0 hour Reading

04 Oct 2019

  • 0.5 hours 1st supervisor meeting

Week 3

08 Oct 2019

  • 3.5 hours Reading
  • 0.5 hours 2nd supervisor meeting

09 Oct 2019

  • 1.0 hour Reading
  • 1.0 hour Keras tests

10 Oct 2019

  • 2.0 hours Keras tests (image colorization)

12 Oct 2019

  • 2.5 hours Reading

15 Oct 2019

  • 1.0 hour Reading

16 Oct 2019

  • 1.5 hours Keras tests (image colorization)

Week 4

24 Oct 2019

  • 2 hours Reading

25 Oct 2019

  • 0.5 hours 3rd supervisor meeting

26 Oct 2019

  • 1.0 hour Requirement analysis

Week 5

29 Oct 2019

  • 4.0 hours Reproduced results of a similar previous project

30 Oct 2019

  • 3.5 hours Investigation into object detection strategies

31 Oct 2019

  • 0.5 hours 4th supervisor meeting

02 Nov 2019

  • 3.0 hours Exploration of YOLO object detection model

Week 6

05 Nov 2019

  • 2.0 hours Prepared extraction of thermal and RGB from FLIR sensor export
  • 2.0 hours Initialized mobile app and set up repository

06 Nov 2019

  • 1.0 hours Got FLIR support to activate my developer account
  • 2.0 hours Further work on YOLO model test

07 Nov 2019

  • 5.0 hours Familiarization with Flir One SDK

08 Nov 2019

  • 0.5 hours 5th supervisor meeting
  • 3.0 hours Work on YOLO activation function

Week 7

12 Nov 2019

  • 3.0 hours Classification task with Pytorch
  • 3.0 hours Started investigating FLIR One calibration

13 Nov 2019

  • 1.0 hour Investigaged FLIR One calibration
  • 1.0 hour Dissertation download link

14 Nov 2019

  • 3.0 hours Investigated FLIR One calibration

15 Nov 2019

  • 0.5 hours 6th supervisor meeting
  • 2.0 hours Worked on Android app

Week 8

20 Nov 2019

  • 3.0 hours Worked on Android app

21 Nov 2019

  • 2.5 hours Detection research
  • 1.5 hour Investigated FLIR One calibration depending on rotation
  • 1.5 hours Hands classification

22 Nov 2019

  • 0.5 hours 7th supervisor meeting
  • 4.0 hours Worked on Android app

24 Nov 2019

  • 6.0 hours NN in tensorflow

Week 9

29 November 2019

  • 0.5 hours 8th supervisor meeting

Week 10

Week 11

13 December 2019

  • 6.0 hours Refactored and improved Android app. Tensorflow integration should be possible now.

14 December 2019

  • 1.5 hours Took test dataset and created model to test on mobile
  • 7.5 hours Started adding Tensorflow Lite to mobile app

15 December 2019

  • 5.5 hours Worked on Tensorflow Lite integration of mobile app- fixed bugs and refactored image pre-processing

16 December 2019

  • 4.5 hours Mobile app can now use test model to classify people

17 December 2019

  • 2.5 hours Some minor tweaking to app

18 December 2019

  • 4.0 hours Android app now supports affine transformation of input images
  • 1.0 hour Test model now assumes previous affine transformation

Week 12

06 Jan 2020

  • 4.5 hours Worked on YOLO TF model

07 Jan 2020

  • 4.0 hours Worked on YOLO TF model

08 Jan 2020

  • 5.0 hours Worked on YOLO TF model- model architecture is complete, many bugs remain

09 Jan 2020

  • 4.0 hours Worked on YOLO TF model (data loading)

10 Jan 2020

  • 5.0 hours Debugged YOLO TF model

11 Jan 2020

  • 3.0 hours Debugged YOLO TF model

Week 13

13 Jan 2020

  • 4.5 hours Debugged YOLO TF model

14 Jan 2020

  • 1.5 hours Debugged YOLO TF model

15 Jan 2020

  • 2.5 hours Debugged YOLO TF model

17 Jan 2020

  • 3.5 hours Debugged YOLO TF model; model is working on MNIST dataset

Week 14

23 Jan 2020

  • 2.0 hours Tried applying YOLO TF model to KAIST dataset (unsatisfactory results)

24 Jan 2020

  • 4.0 hours Tried applying YOLO TF model to KAIST dataset.Work will have to be discontinued due to extended problems (unsatisfactory results, difficulty of debugging, lack of resources to train the model for a sufficient number of epochs, difficulty of deploying the model to TF Lite)

27 Jan 2020

  • 2.0 hours Worked on implementing the NTG (normalised total gradient) metric to quantify image alignment/registration

Week 15

28 Jan 2020

  • 3.0 hours Attempted other metrics for image alignment. Neither NTG nor any other metrics attempted yield good results.
  • 1.0 hour Started writing dissertation chapter about image alignment

31 Jan 2020

  • 0.5 hours 9th supervisor meeting

01 Feb 2020

  • 1.0 hours Added evidence and conclusion to dissertation cahpter about image alignment

02 Feb 2020

  • 6.0 hours Created autoencoder to predict the FIR images of humans from the grayscale version of the RGB images. The model is working very well with a dataset consisting of 2 people.

03 Feb 2020

  • 4.0 hours Tried improving autoencoder by adding more convolutional layers. This was not successful.
  • 1.5 hours Worked on visualization of autoencoder architecture

Week 16

04 Feb 2020

  • 5.0 hours Implemented a texture-based image registration metric

05 Feb 2020

  • 3.0 hours Further investigation of texture-based image registration
  • 1.5 hours Minor changes to autoencoder

10 Feb 2020

  • 3.0 hours Dissertation section on texture-based registration
  • 2.0 hours Dissertation section on spectral imaging

Week 17

11 Feb 2020

  • 5.0 hours Added more background sections to dissertation

12 Feb 2020

  • 3.0 hours Captured animal dataset with 12 classes and over 700 samples.
  • 3.0 hours Pre-processed dataset (downsampling and label generation)

14 Feb 2020

  • 0.5 hours 10th supervisor meeting
  • 2.0 hours Started implementing a classifier for dataset

15 Feb 2020

  • 1.0 hour Implemented residual block in Keras

16 Feb 2020

  • 1.0 hour Captured images of geese and cats to be added to dataset

Week 18

18 Feb 2020

  • 1.5 hours Worked on dissertation chapter about data augmentation
  • 4.0 hours Implemented affine data augmentation for dataset

19 Feb 2020

  • 2.0 hours Fixed issues with augmentation and regenerated dataset
  • 4.0 hours Trained autoencoder on animals dataset

20 Feb 2020

  • 1.0 hour Created U-Net based on autoencoder and trained on subset of animals dataset
  • 1.5 hours Evaluated autoencoder and U-Net on animal images from the internet

21 Feb 2020

  • 0.5 hours 11th supervisor meeting
  • 2.0 hours Fixed class balancing in data augmentation pipeline
  • 1.0 hour Re-generated augmented dataset

22 Feb 2020

  • 2.0 hours Successfully ran simple NN on augmented dataset. It appears as if visible light only performs better than VIS+LWIR. Further investigation is needed.
  • 3.0 hour Started analysis section in dissertation

23 Feb 2020

  • 1.5 hours Dissertation: mobile application analysis

24 Feb 2020

  • 2.0 hours Implemented a variation of AlexNet
  • 3.0 hours Familiarised myself with running jobs on the GPU cluster. Managed to run a simple classification job.

Week 19

25 Feb 2020

  • 2.5 hours Set up custom Docker image for GPU cluster
  • 3.0 hours Captured second animals dataset. This is to be able to perform a representative train/test split.
  • 2.0 hours Added and refactored various pre-processing scripts from test notebooks. These will make it easier to run various tasks on the cluster.
  • 1.0 hour Dissertation: dataset

26 Feb 2020

  • 2.0 hours Dissertation additions
  • 2.0 hours Performed various tests on combined datasets. Models appear to struggle with chicken and alpacas.
  • 1.0 hours Various improvements to pre-processing scripts

27 Feb 2020

  • 1.0 hours Experiments with subsets of dataset (e.g. chickena and alpacas only).
  • 1.0 hours Evaluated different data augmentation techniques
  • 2.0 hours Dissertation additions

28 Feb 2020

  • 0.5 hours 12th supervisor meeting

29 Feb 2020

  • 3.0 hours Captured third animals dataset, as classification accuracy was still too low with the existing data.

02 Mar 2020

  • 2.0 hours Created train/validation split from all 3 raw datasets. Started testing classification on Colab.

03 Mar 2020

  • 1.5 hours Refactored code for models.
  • 1.5 hour Created a grid-search script for running 5 different configurations of models.

Week 20

04 Mar 2020

  • 1.0 hours Added custom implementation of a residual network to models file.
  • 3.0 hours Ran grid search for AlexNet and ResNet. ResNet seems to perform better overall. RGB-only is better than LWIR-only, but combined models are better than both.

06 Mar 2020

  • 0.5 hours 13th supervisor meeting

07 Mar 2020

  • 3.5 hours Dissertation additions
  • 2.0 hours Added pre-made ResNet with pre-trained weights to grid search
  • 0.5 hours Started grid-search on off-the-shelf Resnet

09 Mar 2020

  • 1.5 hours Miscellaneous additions to dissertation
  • 2.0 hours Retrieved off-the-shelf dataset from FLIR One and wrote scripts to be able to load it.
  • 3.0 hours Trained ResNet50 on FLIR One dataset. Prepared transfer-training the model on the custom dataset.

10 Mar 2020

  • 0.5 hours Added custom CNN to grid search
  • 1.5 hours Added script for transfer learning from saved weights
  • 2.0 hours Determined 95 reference points from the train dataset. Determined a new transformation matrix for registration based on these points.
  • 1.5 hours Added registration results and evalution to dissertation

Week 21

11 Mar 2020

  • 2.0 hours Prepared first incomplete draft of dissertation
  • 0.5 hours Updated docker image to include easydict dependency
  • 2.0 hours Added transfer learning to grid search

12 Mar 2020

  • 0.5 hours Minor restructuring of directory structure

13 Mar 2020

  • 0.5 hours 14th supervisor meeting
  • 2.0 hours Added data augmentation to FLIR One dataset loader. Re-trained model on FLIR One dataset.
  • 2.0 hours Various fixes

14 Mar 2020

  • 5.0 hours Dissertation additions (classifiers and autoencoder)

15 Mar 2020

  • 4.0 hours Prepared various visualisations for the dissertation

16 Mar 2020

  • 2.0 hours Successfully deployed RGB-only ResNet to mobile device. It seems to work (tested on pictures). Deployment of other modes should be trivial, although evaluation will be more difficult.
  • 2.5 hours Various dissertation additions

17 Mar 2020

  • 1.5 hours Plotted loss and accuracy of grid search results. All metrics are very unstable and hardly interpretable.
  • 3.5 hours Experimented with different batch sizes and learning rates to fix the problem mentioned above. Reducing learning rates and increasing batch sizes seems to help a lot.

Week 22

18 Mar 2020

  • 2.0 hours Dissertation section on overfitting and regularisation
  • 2.0 hours Batch size evaluation (including dissertation)

19 Mar 2020

  • 4.0 hours Dissertation: Mobile app eval and train-test-split eval

20 Mar 2020

  • 0.5 hours 15th supervisor meeting

21 Mar 2020

  • 6.0 hours Dissertation: Literature review

22 Mar 2020

  • 3.0 hours Ran final grid search on IDA cluster

23 Mar 2020

  • 5.0 hours Dissertation: Modality analysis

24 Mar 2020

  • 4.5 hours Dissertation: Started evaluation section

Week 23

25 Mar 2020

  • 4.0 hours t-SNE and dissertation section

27 Mar 2020

  • 0.5 hours 16th supervisor meeting

28 Mar 2020

  • 3.0 hours Grad-CAM and dissertation section
  • 2.5 hours K-fold cross-validation

29 Mar 2020

  • 6.0 hours Dissertation: Various additions

Week 24

30 Mar 2020

  • 1.0 hours Fixed affine transformations (higher res to minimise interpolation errors)
  • 3.0 hours Dissertation: Autoencoder evaluation

31 Mar 2020

  • 5.5 hours Dissertation: Abstract, introduction, various additions

01 Apr 2020

  • 4.5 hours Dissertation: Various additions

02 Apr 2020

  • 6.0 hours Dissertation: Various additions

03 Apr 2020

  • 0.5 hours 17th supervisor meeting
  • 6.0 hours Dissertation review and quality improvements

04 Apr 2020

  • 5.5 hours Dissertation review and quality improvements
  • 1.5 hours Started presentation

05 Apr 2020

  • 4.0 hours Dissertation review and quality improvements
  • 3.0 hours Additions to presentation
  • 1.0 hour Tidied up repo

Week 25

06 Apr 2020

  • 3.0 hours Recorded presentation