beat issueshttps://gitlab.idiap.ch/groups/beat/-/issues2018-09-20T08:12:49Zhttps://gitlab.idiap.ch/beat/beat.editor/-/issues/156[conda] issue unable to install beat.editor2018-09-20T08:12:49ZFlavio TARSETTI[conda] issue unable to install beat.editor@tiago.pereira @amohammadi :
We are having an issue with `conda install` of `beat` packages:
Our partner on `allies` is having the same issue on his `osx` machine. I first thought it was a problem with his system,
but I am having the s...@tiago.pereira @amohammadi :
We are having an issue with `conda install` of `beat` packages:
Our partner on `allies` is having the same issue on his `osx` machine. I first thought it was a problem with his system,
but I am having the same issue here on my `linux` machine
Have a look at this strange output:
the search finds the package `beat.editor` from the channel `https://www.idiap.ch/software/beat/conda/label/beta` but unable to install it.
Could you please help ? Thanks.
```
conda create -n beat_editor_new_test
Solving environment: done
## Package Plan ##
environment location: <path>/miniconda3/envs/beat_editor_new_test
Proceed ([y]/n)? y
Preparing transaction: done
Verifying transaction: done
Executing transaction: done
#
# To activate this environment, use
#
# $ conda activate beat_editor_new_test
#
# To deactivate an active environment, use
#
# $ conda deactivate
~$ conda activate beat_editor_new_test
(beat_editor_new_test) ~$ conda search beat.editor
Loading channels: done
PackagesNotFoundError: The following packages are not available from current channels:
- beat.editor
Current channels:
- https://repo.anaconda.com/pkgs/main/linux-64
- https://repo.anaconda.com/pkgs/main/noarch
- https://repo.anaconda.com/pkgs/free/linux-64
- https://repo.anaconda.com/pkgs/free/noarch
- https://repo.anaconda.com/pkgs/r/linux-64
- https://repo.anaconda.com/pkgs/r/noarch
- https://repo.anaconda.com/pkgs/pro/linux-64
- https://repo.anaconda.com/pkgs/pro/noarch
To search for alternate channels that may provide the conda package you're
looking for, navigate to
https://anaconda.org
and use the search bar at the top of the page.
(beat_editor_new_test) ~$ conda config --env --add channels https://www.idiap.ch/software/beat/conda/label/beta
(beat_editor_new_test) ~$ conda search beat.editor
Loading channels: done
# Name Version Build Channel
beat.editor 1.0.0b0 py27h551ffd9_13 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h551ffd9_14 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h551ffd9_15 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h551ffd9_16 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h551ffd9_17 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h551ffd9_18 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h9b23768_19 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h9b23768_20 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h9b23768_21 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h9b23768_22 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h9b23768_23 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27h9b23768_24 software/beat/conda/label/beta
beat.editor 1.0.0b0 py27hdf2c03f_12 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h551ffd9_13 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h551ffd9_14 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h551ffd9_15 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h551ffd9_16 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h551ffd9_17 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h551ffd9_18 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h9b23768_19 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h9b23768_20 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h9b23768_21 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h9b23768_22 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h9b23768_23 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36h9b23768_24 software/beat/conda/label/beta
beat.editor 1.0.0b0 py36hdf2c03f_12 software/beat/conda/label/beta
(beat_editor_new_test) ~$ conda install beat.editor
Solving environment: failed
PackagesNotFoundError: The following packages are not available from current channels:
- beat.editor
- beat.cmdline
- beat.core
- docker-py
- beat.editor
- beat.cmdline
- oset
- beat.editor
- flask-restful
Current channels:
- https://www.idiap.ch/software/beat/conda/label/beta/linux-64
- https://www.idiap.ch/software/beat/conda/label/beta/noarch
- https://repo.anaconda.com/pkgs/main/linux-64
- https://repo.anaconda.com/pkgs/main/noarch
- https://repo.anaconda.com/pkgs/free/linux-64
- https://repo.anaconda.com/pkgs/free/noarch
- https://repo.anaconda.com/pkgs/r/linux-64
- https://repo.anaconda.com/pkgs/r/noarch
- https://repo.anaconda.com/pkgs/pro/linux-64
- https://repo.anaconda.com/pkgs/pro/noarch
To search for alternate channels that may provide the conda package you're
looking for, navigate to
https://anaconda.org
and use the search bar at the top of the page.
```
ping @andre.anjos @zmostaani @jdiefenbaugh @samuel.gaist to keep you updatedFlavio TARSETTIFlavio TARSETTIhttps://gitlab.idiap.ch/beat/beat.editor/-/issues/155Change license to BSD 3-clause2020-02-13T09:52:10ZAndré AnjosChange license to BSD 3-clauseAs discussed in earlier weekly meetings (see action points from week02/2018 till week09/2018), the prospective re-use of BEAT core components to re-implement some Bob frameworks would require a shift in the licensing terms for some core ...As discussed in earlier weekly meetings (see action points from week02/2018 till week09/2018), the prospective re-use of BEAT core components to re-implement some Bob frameworks would require a shift in the licensing terms for some core BEAT packages.
@jdiefenbaugh: could you please do that *before* the release? The idea is to go from AGPLv3 to BSD 3-clause (as the rest of Bob) for all packages up to beat.cmdline. We exclude beat.web from this, leaving it as AGPLv3.
We should do the following, in details:
1. Remove the file `LICENSE.AGPL`
2. Create a new file called `LICENSE` the contains the stock BSD 3-clause text (see info here: https://gitlab.idiap.ch/bob/bob.admin/tree/master/templates)
3. Modify the file `MANIFEST.in` to include `LICENSE` instead of `LICENSE.AGPL`
4. Modify the file `setup.py` to list the license correctly on the classifier list (replace license key by `License :: OSI Approved :: BSD License`)
5. Modify the file `conda/meta.yaml` to set the license appropriately. Should read `license: BSD 3-Clause` and `license_family: BSD` on the `about` section, replacing the current AGPL stuff.
6. Double-check all headers from source files - @flavio.tarsetti may have a script to do this.
Please consult with @philip.abbet since I think rechecked, by then, the feasibility of this move and found no issues against doing this.https://gitlab.idiap.ch/beat/beat.cmdline/-/issues/47Change license to BSD 3-clause2019-03-13T14:44:15ZAndré AnjosChange license to BSD 3-clauseAs discussed in earlier weekly meetings (see action points from week02/2018 till week09/2018), the prospective re-use of BEAT core components to re-implement some Bob frameworks would require a shift in the licensing terms for some core ...As discussed in earlier weekly meetings (see action points from week02/2018 till week09/2018), the prospective re-use of BEAT core components to re-implement some Bob frameworks would require a shift in the licensing terms for some core BEAT packages.
@samuel.gaist: could you please do that *before* the release? The idea is to go from AGPLv3 to BSD 3-clause (as the rest of Bob) for all packages up to beat.cmdline. We exclude beat.web from this, leaving it as AGPLv3.
We should do the following, in details:
1. Remove the file `LICENSE.AGPL`
2. Create a new file called `LICENSE` the contains the stock BSD 3-clause text (see info here: https://gitlab.idiap.ch/bob/bob.admin/tree/master/templates)
3. Modify the file `MANIFEST.in` to include `LICENSE` instead of `LICENSE.AGPL`
4. Modify the file `setup.py` to list the license correctly on the classifier list (replace license key by `License :: OSI Approved :: BSD License`)
5. Modify the file `conda/meta.yaml` to set the license appropriately. Should read `license: BSD 3-Clause` and `license_family: BSD` on the `about` section, replacing the current AGPL stuff.
6. Double-check all headers from `.py` files - @flavio.tarsetti may have a script to do this.
Please consult with @philip.abbet since I think rechecked, by then, the feasibility of this move and found no issues against doing this.Samuel GAISTSamuel GAISThttps://gitlab.idiap.ch/beat/beat.core/-/issues/60Change license to BSD 3-clause2019-03-13T15:00:24ZAndré AnjosChange license to BSD 3-clauseAs discussed in earlier weekly meetings (see action points from week02/2018 till week09/2018), the prospective re-use of BEAT core components to re-implement some Bob frameworks would require a shift in the licensing terms for some core ...As discussed in earlier weekly meetings (see action points from week02/2018 till week09/2018), the prospective re-use of BEAT core components to re-implement some Bob frameworks would require a shift in the licensing terms for some core BEAT packages.
@samuel.gaist: could you please do that *before* the release? The idea is to go from AGPLv3 to BSD 3-clause (as the rest of Bob) for all packages up to beat.cmdline. We exclude beat.web from this, leaving it as AGPLv3.
We should do the following, in details:
1. Remove the file `LICENSE.AGPL`
2. Create a new file called `LICENSE` the contains the stock BSD 3-clause text (see info here: https://gitlab.idiap.ch/bob/bob.admin/tree/master/templates)
3. Modify the file `MANIFEST.in` to include `LICENSE` instead of `LICENSE.AGPL`
4. Modify the file `setup.py` to list the license correctly on the classifier list (replace license key by `License :: OSI Approved :: BSD License`)
5. Modify the file `conda/meta.yaml` to set the license appropriately. Should read `license: BSD 3-Clause` and `license_family: BSD` on the `about` section, replacing the current AGPL stuff.
Please consult with @philip.abbet since I think rechecked, by then, the feasibility of this move and found no issues against doing this.Samuel GAISTSamuel GAISThttps://gitlab.idiap.ch/beat/beat.backend.python/-/issues/17Change license to BSD 3-clause2019-03-13T08:19:09ZAndré AnjosChange license to BSD 3-clauseAs discussed in earlier weekly meetings (see action points from week02/2018 till week09/2018), the prospective re-use of BEAT core components to re-implement some Bob frameworks would require a shift in the licensing terms for some core ...As discussed in earlier weekly meetings (see action points from week02/2018 till week09/2018), the prospective re-use of BEAT core components to re-implement some Bob frameworks would require a shift in the licensing terms for some core BEAT packages.
@samuel.gaist: could you please do that *before* the release? The idea is to go from AGPLv3 to BSD 3-clause (as the rest of Bob) for all packages up to beat.cmdline. We exclude beat.web from this, leaving it as AGPLv3.
We should do the following, in details:
1. Remove the file `LICENSE.AGPL`
2. Create a new file called `LICENSE` the contains the stock BSD 3-clause text (see info here: https://gitlab.idiap.ch/bob/bob.admin/tree/master/templates)
3. Modify the file `MANIFEST.in` to include `LICENSE` instead of `LICENSE.AGPL`
4. Modify the file `setup.py` to list the license correctly on the classifier list (replace license key by `License :: OSI Approved :: BSD License`)
5. Modify the file `conda/meta.yaml` to set the license appropriately. Should read `license: BSD 3-Clause` and `license_family: BSD` on the `about` section, replacing the current AGPL stuff.
Please consult with @philip.abbet since I think rechecked, by then, the feasibility of this move and found no issues against doing this.Samuel GAISTSamuel GAISThttps://gitlab.idiap.ch/beat/beat.editor/-/issues/154A few issues with the Plotter editor2018-10-31T13:05:05ZJaden DIEFENBAUGHA few issues with the Plotter editor- The dataformat select by default shows the first available plotter dataformat as selected, but it should really show a "blank" selection, prompting the user to choose a dataformat
- The default value for (only `bool`-type?) parameters ...- The dataformat select by default shows the first available plotter dataformat as selected, but it should really show a "blank" selection, prompting the user to choose a dataformat
- The default value for (only `bool`-type?) parameters arent saved to the metadata
- There's no way to add libraries to the plotter!Flavio TARSETTIFlavio TARSETTIhttps://gitlab.idiap.ch/beat/beat.editor/-/issues/153Add a way to delete multiple conns at once2020-02-13T09:55:42ZJaden DIEFENBAUGHAdd a way to delete multiple conns at onceSee #152 for the exact request.
This issue was fixed in https://gitlab.idiap.ch/beat/beat.editor/merge_requests/117See #152 for the exact request.
This issue was fixed in https://gitlab.idiap.ch/beat/beat.editor/merge_requests/117[v2] 2 - Toolchain editorhttps://gitlab.idiap.ch/beat/beat.editor/-/issues/152Some suggestions for beat editor2019-01-28T10:11:31ZZohreh MOSTAANISome suggestions for beat editorHi,
Here are some ideas that I got while using beat editor. You might find them useful.
* I think it would be good if the connections in a toolchain could break at some points and bypass blocks instead of directly connecting inputs and...Hi,
Here are some ideas that I got while using beat editor. You might find them useful.
* I think it would be good if the connections in a toolchain could break at some points and bypass blocks instead of directly connecting inputs and outputs. Something like what is available on the beat platform. The reason is that even though it is more convenient while making a toolchain to just connect an output to an input it is very difficult to understand what is going on when you look at an experiment using that toolchain, specially when the experiments get bigger and the connections are crossing each other. Even if there was a mechanism to highlight a chosen connection in the experiment page can be really helpful. This mechanism is already there when making a toolchain but not when you look at the experiment.
* It would be good if the user can assign a colour to a channel of data while making the toolchain. As far as I tried the channel colours are chosen by the system and in many times they are visually close to each other. I sometimes had difficulty distinguishing between them.
* It would be nice that user can change the name of groups while defining the algorithms. This is not the case as far as I experienced.
* It would be nice if the separation between different parameter categories in the experiment page such as Block, Global Settings, etc were more distinct.
* It would be nice if the user can choose several connections and delete them altogether. Even if there was a button like "delete all connections" might be useful. I had painful experiences trying to remove all the connections but not the blocks. :)
I might be wrong about any of these points and the features may already be there. Please let me know if that is the case. :)https://gitlab.idiap.ch/beat/beat.editor/-/issues/151need to manually access the web interface using `beat editor serve ` command2018-09-10T16:00:51ZZohreh MOSTAANIneed to manually access the web interface using `beat editor serve ` commandHi,
I got the new version of beat editor and now when I run the command `beat editor serve` the firfox window does not pop up. I have to access it manually.Hi,
I got the new version of beat editor and now when I run the command `beat editor serve` the firfox window does not pop up. I have to access it manually.https://gitlab.idiap.ch/beat/beat.editor/-/issues/150Specify the `june-allies` branch of `beat.tutorial` in documentation2018-12-19T19:02:44ZJaden DIEFENBAUGHSpecify the `june-allies` branch of `beat.tutorial` in documentationhttps://gitlab.idiap.ch/beat/beat.editor/-/issues/149beateditor is completely down for me2018-12-06T09:49:29ZZohreh MOSTAANIbeateditor is completely down for meHi,
Today when I ran beateditor command to edit some experiment, it did not show me anything on the Firefox. It only shows the page for split second and then it is a blank page. I updated beat.editor with conda in case there were some c...Hi,
Today when I ran beateditor command to edit some experiment, it did not show me anything on the Firefox. It only shows the page for split second and then it is a blank page. I updated beat.editor with conda in case there were some changes but the problem is still there.https://gitlab.idiap.ch/beat/beat.editor/-/issues/148packer_scripts does not exist, while it is mentioned in the documentation of ...2018-11-26T02:45:09ZZohreh MOSTAANIpacker_scripts does not exist, while it is mentioned in the documentation of Method 1 for installationThere is no packer_scripts in `https://gitlab.idiap.ch/beat/beat.tutorial.git`.There is no packer_scripts in `https://gitlab.idiap.ch/beat/beat.tutorial.git`.https://gitlab.idiap.ch/beat/beat.web/-/issues/505the experiment fails without any failure in any block2018-08-28T10:12:15ZZohreh MOSTAANIthe experiment fails without any failure in any blockI was running experiments on the beat platform. The experiments were big and I started them 4 days ago. Today I received notification that the experiments failed. When I checked them there were no error anywhere.
The state of the exper...I was running experiments on the beat platform. The experiments were big and I started them 4 days ago. Today I received notification that the experiments failed. When I checked them there were no error anywhere.
The state of the experiment is failed however the blocks are either passed or cancelled. This is strange and I do not know where the problem might be.
https://gitlab.idiap.ch/beat/beat.web/-/issues/504strange behaviour regarding an analyzer in beat platform2018-11-30T16:03:45ZZohreh MOSTAANIstrange behaviour regarding an analyzer in beat platformI ran two different experiment on beat platform and both of them used the same database (replay/3) and the same analyser (Kanma/iqm_spoof_eer_analyzer/1). The number of positive and negative scores regardless of what the experiment is sh...I ran two different experiment on beat platform and both of them used the same database (replay/3) and the same analyser (Kanma/iqm_spoof_eer_analyzer/1). The number of positive and negative scores regardless of what the experiment is should be the same, since it only depends on the database. However for one of the experiments these numbers were wrong.
| | exp1(correct) | exp2(wrong) |
| ------------------- |:--------------:| -----------:|
| dev_numNegatives | 300 | 297 |
| dev_numPositives | 60 | 60 |
| test_numNegatives | 400 | 260 |
| test_numPositives | 80 | 80 |
|Scientific Python 2.7| 0.0.4 | 1.0.0 |
Then I ran the experiment with the wrong numbers in my local beat using beat.cmdline and beat.editor ( python 3.6 with the latest version of bob and beat). This time the numbers were correct!
I was wondering if you have any idea what might be the problem and how can I debug it further.https://gitlab.idiap.ch/beat/beat.editor/-/issues/147database name does not check validity if the name is not matching the pattern...2018-08-24T22:52:16ZZohreh MOSTAANIdatabase name does not check validity if the name is not matching the pattern ruleI made a new database with name "replay-me" and it was valid. I assumed that if the name has a problem the editor will tell me. So I made an experiment using this database and it was invalid no matter what I did. Finally I ran the invali...I made a new database with name "replay-me" and it was valid. I assumed that if the name has a problem the editor will tell me. So I made an experiment using this database and it was invalid no matter what I did. Finally I ran the invalid experiment and I get:
```
Failed to load the experiment `zmostaani/zmostaani/sbhatta-iqm-face-antispoofing-test/2/newantispoofing':
* /datasets/dev_probes/database: 'replay-me/1' does not match '^[a-zA-Z_][a-zA-Z0-9_]*/[0-9]+$' (rule: /properties/datasets/patternProperties/^[a-zA-Z_][a-zA-Z0-9_-]*$/properties/database/
pattern)
* /datasets/test_probes/database: 'replay-me/1' does not match '^[a-zA-Z_][a-zA-Z0-9_]*/[0-9]+$' (rule: /properties/datasets/patternProperties/^[a-zA-Z_][a-zA-Z0-9_-]*$/properties/database
/pattern)
* /datasets/train/database: 'replay-me/1' does not match '^[a-zA-Z_][a-zA-Z0-9_]*/[0-9]+$' (rule: /properties/datasets/patternProperties/^[a-zA-Z_][a-zA-Z0-9_-]*$/properties/database/patte
rn)
Error: Error occured
```
I changed the name of the database to "replayme" and the experiment was valid afterwards. I think these patterns should be considered when checking the name chosen for the database.https://gitlab.idiap.ch/beat/beat.web/-/issues/503The experiment stays in state cancelling when stopping the experiment2018-08-31T09:02:11ZZohreh MOSTAANIThe experiment stays in state cancelling when stopping the experimentWhen I stop an experiment the blocks that were running stays in the 'cancelling' state and they never change to 'cancelled' plus there is nothing to resume a stopped experiment. I can only fork it again to be able to run it.When I stop an experiment the blocks that were running stays in the 'cancelling' state and they never change to 'cancelled' plus there is nothing to resume a stopped experiment. I can only fork it again to be able to run it.https://gitlab.idiap.ch/beat/beat.backend.python/-/issues/16The input to a block gets the object from index method of View while the name...2018-09-18T12:27:44ZZohreh MOSTAANIThe input to a block gets the object from index method of View while the name from get method of ViewI am running an experiment locally using beat.editor. I defined the replay-attack database (using both my own code and the one in beat.examples (https://gitlab.idiap.ch/beat/beat.examples/blob/master/advanced/databases/replay/4.py) (usin...I am running an experiment locally using beat.editor. I defined the replay-attack database (using both my own code and the one in beat.examples (https://gitlab.idiap.ch/beat/beat.examples/blob/master/advanced/databases/replay/4.py) (using the grandtest protocol). I get the following error in both cases.
```
(beat_py3) zmostaani@helvetix02:/remote/idiap.svm/user.active/zmostaani/experiments/beat$ beat exp run zmostaani/zmostaani/sbhatta-iqm-face-antispoofing-test/1/replay2-antispoofing-iqm-lda-2
Index for database replay/2 not found, building it
Index for database replay/2 not found, building it
Index for database replay/2 not found, building it
Running `zmostaani/iqm_features/1' for block `Feats_IQM_Train'
Start the execution of 'zmostaani/iqm_features/1'
Traceback (most recent call last):
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/bin/beat", line 11, in <module>
sys.exit(main())
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/click/core.py", line 722, in __call__
return self.main(*args, **kwargs)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/click/core.py", line 697, in main
rv = self.invoke(ctx)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/click/core.py", line 1066, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/click/core.py", line 1066, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/click/core.py", line 895, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/click/core.py", line 535, in invoke
return callback(*args, **kwargs)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/click/decorators.py", line 17, in new_func
return f(get_current_context(), *args, **kwargs)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/cmdline/decorators.py", line 75, in _decorator
value = view_func(*args, **kwargs)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/cmdline/experiments.py", line 569, in run
return run_experiment(config, name, force, docker, local, quiet)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/cmdline/experiments.py", line 212, in run_experiment
with executor:
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/core/execution/local.py", line 182, in __enter__
self._prepare_inputs()
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/core/execution/local.py", line 194, in _prepare_inputs
databases=self.databases
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/backend/python/helpers.py", line 195, in create_inputs_from_configuration
start_index=start_index, end_index=end_index)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/backend/python/database.py", line 174, in setup
start_index=start_index, end_index=end_index, pack=pack)
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/backend/python/data.py", line 569, in setup
previous_value = getattr(obj, output_name)
AttributeError: 'Entry' object has no attribute 'class'
```
In get method there is 'class' but in the index there is not, because beat is using namedtuple for making the Entry points and this function does not allow to have name 'class' as an Entry. Even if I define a dictionary instead of namedtuple in my database the python backend is chainging it to namedtuple.
```
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/site-packages/beat/backend/python/database.py", line 157, in setup
Entry = namedtuple('Entry', sorted(objs[0].keys()))
File "/idiap/user/zmostaani/miniconda3/envs/beat_py3/lib/python3.6/collections/__init__.py", line 404, in namedtuple
'keyword: %r' % name)
ValueError: Type names and field names cannot be a keyword: 'class'
```
I have two questions
- why not use simple dictionaries in python backend instead of namedtupe
- why the input of block is trying to access objects in index instead of get.
Please let me know if you need more information.https://gitlab.idiap.ch/beat/beat.web/-/issues/502GPU tasks on infrastructure2018-10-30T15:27:37ZSamuel GAISTGPU tasks on infrastructureIn infrastructure:
* [x] Setup compute node to use GPU
* [x] Test Docker with GPU
* [x] Document setup procedure
* [x] Validate functionality on compute node
* [x] Test with experiment from the platform
* [x] Deploy in productionIn infrastructure:
* [x] Setup compute node to use GPU
* [x] Test Docker with GPU
* [x] Document setup procedure
* [x] Validate functionality on compute node
* [x] Test with experiment from the platform
* [x] Deploy in productionEnable GPUSamuel GAISTSamuel GAIST2018-09-21https://gitlab.idiap.ch/beat/beat.web/-/issues/501GPU tasks on local machine2018-08-21T14:15:54ZSamuel GAISTGPU tasks on local machineOn local machine:
* [x] Create a new docker environment with GPU enabled packages like pytorch
* [x] Validate starting of GPU enabled conda containers with nvidia-docker2
* [x] Write sample algorithms that makes use of GPU
* [x] Vali...On local machine:
* [x] Create a new docker environment with GPU enabled packages like pytorch
* [x] Validate starting of GPU enabled conda containers with nvidia-docker2
* [x] Write sample algorithms that makes use of GPU
* [x] Validate running an experience from beat.cmdline
* [x] Validate starting and running an experience from beat.webEnable GPUSamuel GAISTSamuel GAIST2018-08-31https://gitlab.idiap.ch/beat/beat.cmdline/-/issues/45The script `beat.py` should be gone2018-08-27T15:37:08ZAndré AnjosThe script `beat.py` should be goneAfter replacing it by `main_cli.py`, I think this is now duplicated code which serves no purpose. @flavio.tarsetti, please check and make an MR which removes this file (https://gitlab.idiap.ch/beat/beat.cmdline/blob/1.4.x/beat/cmdline/sc...After replacing it by `main_cli.py`, I think this is now duplicated code which serves no purpose. @flavio.tarsetti, please check and make an MR which removes this file (https://gitlab.idiap.ch/beat/beat.cmdline/blob/1.4.x/beat/cmdline/scripts/beat.py).Flavio TARSETTIFlavio TARSETTI