beat issueshttps://gitlab.idiap.ch/groups/beat/-/issues2015-11-20T11:27:21Zhttps://gitlab.idiap.ch/beat/beat.web/-/issues/235Cannot jump to algorithm/toolchain page from a (completed/failed) experiment ...2015-11-20T11:27:21ZLaurent EL SHAFEYCannot jump to algorithm/toolchain page from a (completed/failed) experiment pageWhen an experiment is being configured, it is possible to visualize the algorithm code and/or the toolchain.
In contrast, once an experiment has completed (or failed), the number of links on the webpage is very limited.
Besides, the ...When an experiment is being configured, it is possible to visualize the algorithm code and/or the toolchain.
In contrast, once an experiment has completed (or failed), the number of links on the webpage is very limited.
Besides, the algorithm name may only be partially visible by looking at the toolchain graph. It would be better to have entry points to both the algorithm and the toolchain pages, similarly to what we have when we set up an experiment.
Final Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/234Feature request2015-11-20T11:27:43ZSébastien MARCELFeature requestJust my 2 cent.
Will it be possible to give a user the possibility to report an issue on gitlab directly from popup window with the error message by clicking on a button in that windows ?Just my 2 cent.
Will it be possible to give a user the possibility to report an issue on gitlab directly from popup window with the error message by clicking on a button in that windows ?Final Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/230E-mail notification upon sharing2015-11-20T11:28:35ZAndré AnjosE-mail notification upon sharingIt would be a good feature to allow users to globally enable notification when something is shared with them.
As @philip.abbet pointed out, it would be good if e-mails could be summarized in case a lot has been just shared with a pers...It would be a good feature to allow users to globally enable notification when something is shared with them.
As @philip.abbet pointed out, it would be good if e-mails could be summarized in case a lot has been just shared with a person (e.g., when sharing an experiment, this normally includes algorithms, dataformats, the toolchain, libraries, etc). At this point, it would be best to receive a single e-mail listing all objects from the last sharing actions.
We should also think about the situation where various experiments are shared at once and avoid the e-mail spam once more on this case.
This feature could also have a hook to our to-be-implemented "activity" tab.Final Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/227E-mail notification when experiment done2015-11-05T08:39:39ZAndré AnjosE-mail notification when experiment doneIt would be an useful feature a check box on the experiment submission page allowing the user to be notified when that experiment is done. We already have the user e-mail in hand.It would be an useful feature a check box on the experiment submission page allowing the user to be notified when that experiment is done. We already have the user e-mail in hand.Final Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/221Few algorithms do directly use data_index attributes2015-11-10T11:58:55ZLaurent EL SHAFEYFew algorithms do directly use data_index attributesFew algorithms on the platform do directly use data_index{,_end} attributes of the inputs.
The I/O uses the SynchronizationListener class to ensure that outputs are always synchronized with the main input channel.
Therefore, the use of...Few algorithms on the platform do directly use data_index{,_end} attributes of the inputs.
The I/O uses the SynchronizationListener class to ensure that outputs are always synchronized with the main input channel.
Therefore, the use of these attributes in the algorithm side should not be neither required nor allowed.
See the following issue for more details: https://gitlab.idiap.ch/biometric/beat.examples/issues/11
Final Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/216Locked attestations with other user's components could be improved2015-11-20T11:32:41ZAndré AnjosLocked attestations with other user's components could be improvedWhen the user goes to unlock an attestation by clicking on the respective button at the attestation page, in the case none of the components belong to that user, the display could be improved (see image).
![image](https://gitlab.idiap...When the user goes to unlock an attestation by clicking on the respective button at the attestation page, in the case none of the components belong to that user, the display could be improved (see image).
![image](https://gitlab.idiap.ch/biometric/beat.web/uploads/2475fa071eb26ef4b83ea600d311924b/image.png)
From the other perspective, unlocking an attestation with components which belong to the user shows components which are already public (see the second image).
![image](https://gitlab.idiap.ch/biometric/beat.web/uploads/255ad59e963468252dbab1679f87f072/image.png)
I wonder if we should not have a more consistent display through these use cases. A suggestion:
1. Always display all components involved or leave the option to display them by following a link, for example (if space is an issue on this pop-up)
2. If a component is public already, indicate so
3. If an algorithm is public and/or open-source, indicate soFinal Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/215Possibility to delete locked attestations2015-11-20T11:31:10ZAndré AnjosPossibility to delete locked attestationsIt would be nice if the user could delete locked (still not public) attestations. IMO, there is no reason why a "private" attestation could not be deleted by the user requesting it.
Unlocked attestations should not be deletable as the...It would be nice if the user could delete locked (still not public) attestations. IMO, there is no reason why a "private" attestation could not be deleted by the user requesting it.
Unlocked attestations should not be deletable as they are already.Final Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/186Reports: Export to CSV, LaTeX, PNG2015-10-26T08:34:28ZPhilip ABBETReports: Export to CSV, LaTeX, PNGDepends on #185
In a Locked Report, the table generated should be exportable in LaTeX or CSV mode, so that a user can adapt it to its own publication.
At minimum: CSV + PNG
Depends on #185
In a Locked Report, the table generated should be exportable in LaTeX or CSV mode, so that a user can adapt it to its own publication.
At minimum: CSV + PNG
Final DocumentationFlavio TARSETTIFlavio TARSETTIhttps://gitlab.idiap.ch/beat/beat.web/-/issues/184Reports: Report page editor2015-10-02T15:47:34ZPhilip ABBETReports: Report page editorDepends on #183
The Report page should allow the user to display features of the
experiment. Preferably, it should be made flexible enough so that new
conditions can be added in a relatively easy way (i.e., w/o necessarily
requiring a ...Depends on #183
The Report page should allow the user to display features of the
experiment. Preferably, it should be made flexible enough so that new
conditions can be added in a relatively easy way (i.e., w/o necessarily
requiring a modification of large parts of the system). At first, apart
from what is already displayable at Search results, these are the other
conditions we'd thought are necessary. We've organized it in "Columns" of
interest, which can be mixed and matched upon user configuration.
- [x] Column "block computing time": Displays the computing time of a given block. Requires all experiments have a block with a given name
- [x] Column "total experiment time": Displays the total computing time for an experiment
- [x] Column "parameter value": Displays a parameter value in an algorithm for all experiments. Requires all experiments have an algorithm with a given value. If the algorithm has two values, displays a comma-separated list of values for each of those parameters.
- [x] Column "results": Displays any valid analyzer result.
- [x] Column "user label": Allows the user to define a label of its own for a given experiment.Final DocumentationFlavio TARSETTIFlavio TARSETTIhttps://gitlab.idiap.ch/beat/beat.web/-/issues/180Search: More flexible filters specification2015-11-20T11:33:46ZPhilip ABBETSearch: More flexible filters specificationAt the Search page, it would be good to be able to add logical-OR conditionals. Grouping of ANDs and ORs would be a plus.
**The report feature could be seen has a workaround for this feature!**At the Search page, it would be good to be able to add logical-OR conditionals. Grouping of ANDs and ORs would be a plus.
**The report feature could be seen has a workaround for this feature!**Final DocumentationPhilip ABBETPhilip ABBEThttps://gitlab.idiap.ch/beat/beat.web/-/issues/82Harmonisation of statistics and processing times sent by the scheduler2015-11-20T11:35:02ZAndré AnjosHarmonisation of statistics and processing times sent by the schedulerWe must harmonize the displaying of computing power informed by the scheduler and calculated by the worker in some way. Currently, we find some inconsistencies between the values displayed on the user page (experiments tab) and the "sequ...We must harmonize the displaying of computing power informed by the scheduler and calculated by the worker in some way. Currently, we find some inconsistencies between the values displayed on the user page (experiments tab) and the "sequential processing" times displayed on the experiment page.Final Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/78Locked attestations do not properly expire2015-11-20T11:36:56ZAndré AnjosLocked attestations do not properly expireThe `creation_date` is correctly set for all locked attestations. The `publication_date` is set for all public attestations. The only thing we need to implement to get this going is:
- [x] A warning on the locked attestation page that t...The `creation_date` is correctly set for all locked attestations. The `publication_date` is set for all public attestations. The only thing we need to implement to get this going is:
- [x] A warning on the locked attestation page that tells the user about the remaining unlock time (which is a constant on the `settings.py` file for the project.
- [x] A cronjob that runs a (very simple) script that, for all attestations w/o publication_date, compares the current date to the creation date and deletes that attestation (just the attestation, not the experiment) if the time difference pass a threshold.Final Documentationhttps://gitlab.idiap.ch/beat/beat.web/-/issues/42Multi-resolution support2015-11-20T11:36:14ZAndré AnjosMulti-resolution supportSome of our pages at the website are no well-adapated to smaller displays. When running representations on low-resolution projectors or using tablets and phones, it would be nice that the data presented is, at least, not overlapping.
...Some of our pages at the website are no well-adapated to smaller displays. When running representations on low-resolution projectors or using tablets and phones, it would be nice that the data presented is, at least, not overlapping.
A specialised page for small devices would be a plus.Final DocumentationPhilip ABBETPhilip ABBEThttps://gitlab.idiap.ch/beat/beat.web/-/issues/39Algorithm to reset the toolchain graph display2015-12-15T11:33:23ZAndré AnjosAlgorithm to reset the toolchain graph displayAs of today, we have toolchains that were created by hand. In that case, the toolchain editor cannot display the toolchain properly. There will be cases in the future as well, in which the user would like to reorganise a toolchain to its...As of today, we have toolchains that were created by hand. In that case, the toolchain editor cannot display the toolchain properly. There will be cases in the future as well, in which the user would like to reorganise a toolchain to its best possible representation. For those purposes, we need to implement an algorithm for finding the best graph given the toolchain and some constraints.
Possibilities:
1. A simple algorithm which would be better than stacking all blocks to the left as it is today:
* Put all databases, spaced to the left, one after the other
* All analyzers, spaced to the right
* All blocks equally randomly around the space left
2. Use a more complex graph-theory algorithm. Searching for "topological graph sorting" in Python I did find some resources:
* http://code.activestate.com/recipes/578272-topological-sort/, for acyclic graphs.
* A package in Python that can do it: https://pypi.python.org/pypi/toposort/1.1
* Sorting cyclic graphs is apparently an NP-hard optimisation problem, so we should not try to solve it.Final DocumentationPhilip ABBETPhilip ABBEThttps://gitlab.idiap.ch/beat/beat.web/-/issues/9Result display and search: missing features2015-11-10T12:02:44ZAndré AnjosResult display and search: missing featuresThe result search and display has been recently revamped.
Here is a list of missing features yet to be implemented:
- [X] Display only default (`+`) results
- [ ] The parameters, names of algorithms, versions, should be display-a...The result search and display has been recently revamped.
Here is a list of missing features yet to be implemented:
- [X] Display only default (`+`) results
- [ ] The parameters, names of algorithms, versions, should be display-able at the search results.
- [X] Filter out columns from the results on user request
- [ ] Order columns from the results on user request
- [x] Save searches (implement user micro-site tab)
- [x] Sharing of saved search results
- [ ] Modify front page to display public searches instead of public toolchains, which are less useful.
- [ ] It should be possible to export figures and tables in a search result for scientific publishing. Plotting should also be improved. It is far from scientific standards of today. E.g., it would be nice to get it exported into PDF, with nice fonts and configurable colors. A compromise would be to export the plotting data and let the user handle it the way they want.
- [ ] It should be possible to annex complementary search results when showing previous results. This gives users an overview if a certain result is still representative of the state-of-the-art, for example. It also allows for recommendations.
Final DocumentationPhilip ABBETPhilip ABBEThttps://gitlab.idiap.ch/beat/beat.web/-/issues/435[experiments] Arrow icons not in the right direction2017-08-06T16:44:51ZAndré Anjos[experiments] Arrow icons not in the right directionThere is something wrong with the current logic:
![image](/uploads/6917f0a8a9e6382a3964868089b05326/image.png)
There is something wrong with the current logic:
![image](/uploads/6917f0a8a9e6382a3964868089b05326/image.png)
Open-source ReleaseAndré AnjosAndré Anjoshttps://gitlab.idiap.ch/beat/beat.web/-/issues/428[reports] chevrons tables2017-08-06T16:44:52ZFlavio TARSETTI[reports] chevrons tablesAdd gray up/down chevrons for all th in tables
Chevron need to be with white-space:nowrap next to thAdd gray up/down chevrons for all th in tables
Chevron need to be with white-space:nowrap next to thOpen-source ReleaseFlavio TARSETTIFlavio TARSETTIhttps://gitlab.idiap.ch/beat/beat.web/-/issues/427[reports] patch for sorting items + total time sort + experiment names2017-08-06T16:44:52ZFlavio TARSETTI[reports] patch for sorting items + total time sort + experiment namesPatch required for sorting itemsPatch required for sorting itemsOpen-source ReleaseFlavio TARSETTIFlavio TARSETTIhttps://gitlab.idiap.ch/beat/beat.web/-/issues/426[reports] tables sorting and saving2017-08-06T16:44:52ZFlavio TARSETTI[reports] tables sorting and savingThe tables in reports are currently not sortable.
We would like to enforce a mechanism to sort and save this sorting preferences for each tables of the report.
Others viewers should also be able to sort the table if they wish too.The tables in reports are currently not sortable.
We would like to enforce a mechanism to sort and save this sorting preferences for each tables of the report.
Others viewers should also be able to sort the table if they wish too.Open-source ReleaseFlavio TARSETTIFlavio TARSETTIhttps://gitlab.idiap.ch/beat/beat.web/-/issues/425[algorithms] If algorithms are only usable, users can still see it through th...2017-08-06T16:44:52ZAndré Anjos[algorithms] If algorithms are only usable, users can still see it through the "diff" panelExample:
https://www.beat-eu.org/platform/algorithms/diff/anjithgeorge/KBOC_IITKGP_ANJITH_APRIL7_1/10/anjithgeorge/KBOC_IITKGP_ANJITH_APRIL7_1/14/
This option should be removed.Example:
https://www.beat-eu.org/platform/algorithms/diff/anjithgeorge/KBOC_IITKGP_ANJITH_APRIL7_1/10/anjithgeorge/KBOC_IITKGP_ANJITH_APRIL7_1/14/
This option should be removed.Open-source ReleaseAndré AnjosAndré Anjos