summaryrefslogtreecommitdiff
path: root/examples/py-machine-learning
Commit message (Collapse)AuthorAge
* Improved Keras Autoencoder hyper parameter.Toni Uhlig2024-08-16
| | | | Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Fixed more SonarCloud complaints.Toni Uhlig2023-11-07
| | | | Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Fixed some SonarCloud complaints.Toni Uhlig2023-11-06
| | | | Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* sklearn-random-forest.py: Pretty print false positive/negative.Toni Uhlig2023-09-10
| | | | | | | * added max tree depth command line argument * print a note if loading an existing model while using --sklearn-* command line options Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* keras-autoencoder.py: TensorBoard, SGD optimizer, KLDivergence loss ↵Toni Uhlig2023-08-23
| | | | | | function, EarlyStopping Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* keras-autoencoder.py: Improved ModelToni Uhlig2023-08-20
| | | | | | | * added initial learning rate for Adam * plot some metrics using pyplot Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* keras-autoencoder.py: fixed invalid preprocessing of received base64 packet dataToni Uhlig2023-08-15
| | | | | | | * split logic into seperate jobs; nDPIsrvd and Keras * nDPIsrvd: break event processing and re-run `epoll_wait()` after client disconnected Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Keras AE supports loading/saving models.Toni Uhlig2023-07-22
| | | | | | * added training/batch size as cmdargs Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Added Keras based Autoencode (Work-in-Progress!)Toni Uhlig2023-07-16
| | | | | | * minor fixes Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* bump libnDPI to 8ea0eaa0d0c4a3be05f67ef7fa1d22c2579cf7d1Toni Uhlig2023-06-12
| | | | | | | | | | * added build fix for Gitlab CI * added friendly C11 check * set required libnDPI versionto 4.7 (ArchLinux ndpi-git sets version to 4.7, which is not released yet) * reduced sklearn-random-forest memory consumption by adjusting min. sample leaf Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Improved:Toni Uhlig2023-02-27
| | | | | | | | * Gitlab-CI: build nDPId executable from CLI * C-Simple: log affected JSON line on READ/PARSE error * Sklearn: quality of life changes Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* py-machine-learning: Print CSV line numbers for invalid lines (SKLearn ↵Toni Uhlig2023-01-30
| | | | | | | | Random Forest Classificator). * c-analysed: fix wrong length check Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* py-machine-learning / sklearn-random-forest: Quality Of Life improvmentsToni Uhlig2022-12-22
| | | | | | * fixed libnDPI submodule build on some platforms Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* py-machine-learning: load and save trained modelsToni Uhlig2022-10-15
| | | | | | * added link to a pre-trained model Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Removed example py-ja3-checker.Toni Uhlig2022-10-13
| | | | | | | * renamed sklearn-ml.py to sklearn-random-forest.py (there is more to come!) * force all protocol classes to lower case Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Improved py-machine-learning example.Toni Uhlig2022-10-11
| | | | | | | | * colorize/prettify output * added sklearn controls/tuning options * disable IAT/Packet-Length features as default Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Improved py-machine-learning example.Toni Uhlig2022-10-10
| | | | Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
* Improved py-machine-learning example.Toni Uhlig2022-10-09
| | | | | | | | | * c-analysed: fixed quoting bug * nDPId: fixed invalid iat storing/serialisation * nDPId: free data analysis after event was sent Signed-off-by: Toni Uhlig <matzeton@googlemail.com> Signed-off-by: lns <matzeton@googlemail.com>
* Improved flown analyse event:Toni Uhlig2022-09-22
* store packet directions * merged direction based IATs * merged direction based PKTLENs Signed-off-by: Toni Uhlig <matzeton@googlemail.com>
Powered by cgit, Themed by Ted Yin.