installation Error: No such file or directory: '/data/model.toml'

[MaryElna]

Hi,
I got this error: ( FileNotFoundError: [Errno 2] No such file or directory: '/data/model.toml') during of built Docker images installation using GPU, I run it on ubuntu v20.04.01
Here is what was appeared:
Lia@Lia-GL553VD:~$ sudo docker run --runtime=nvidia -it --rm -v $PWD:/data --ipc=host mapbox/robosat:latest-gpu train --model /data/model.toml --dataset /data/dataset.toml --workers 4
Unable to find image 'mapbox/robosat:latest-gpu' locally
latest-gpu: Pulling from mapbox/robosat
35c102085707: Pull complete
251f5509d51d: Pull complete
8e829fe70a46: Pull complete
6001e1789921: Pull complete
9f0a21d58e5d: Pull complete
47b91ac70c27: Pull complete
a0529eb74f28: Pull complete
23bff6dcced5: Pull complete
6d10e7155e41: Pull complete
1077cd91e25b: Pull complete
17187e0cb362: Pull complete
59f16297861a: Pull complete
a01b17d3e344: Pull complete
832b14d21211: Pull complete
ec4844dbe142: Pull complete
5a77bff3a5ba: Pull complete
0c570f863b22: Pull complete
bb1abf747bd6: Pull complete
Digest: sha256:f4d0234287a9d7175554859cc70450cd3ef198c48b78dd4fd583d0d768a1db04
Status: Downloaded newer image for mapbox/robosat:latest-gpu
Traceback (most recent call last):
File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main
"main", mod_spec)
File "/usr/lib/python3.6/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/usr/src/app/robosat/tools/main.py", line 58, in
args.func(args)
File "/usr/src/app/robosat/tools/train.py", line 57, in main
model = load_config(args.model)
File "/usr/src/app/robosat/config.py", line 21, in load_config
return toml.load(path)
File "/opt/venv/lib/python3.6/site-packages/toml/decoder.py", line 111, in load
with io.open(_getpath(f), encoding='utf-8') as ffile:
FileNotFoundError: [Errno 2] No such file or directory: '/data/model.toml'

Thanks for any commands,
Best Regards
Mary

[daniel-j-h]

You are mounting your current directory to /data inside the container, and then you are accessing /data/model.toml.

Can you make you you have a model.toml file in your current directory? Check the config dir.

[MaryElna]

Dear Daniel After some tries I couldn't find model.toml file in my system, would you please guide me more or send me that file? Best Mary

…

On Thu, Dec 17, 2020 at 7:53 AM Daniel J. H. ***@***.***> wrote: You are mounting your current directory to /data inside the container, and then you are accessing /data/model.toml. Can you make you you have a model.toml file in your current directory? Check the config dir. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#215 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOOPU2QOJHPAAQH46ZWR7CTSVH5LDANCNFSM4U7H3GCQ> .

[daniel-j-h]

It is the model configuration file in the robosat/config directory. You will need to modify the model and tge dataset config to your specific use case, because you are running the train command.

…

On December 17, 2020 3:59:15 PM UTC, MaryElna ***@***.***> wrote: Dear Daniel> After some tries I couldn't find model.toml file in my system, would you> please guide me more or send me that file?> Best> Mary> > On Thu, Dec 17, 2020 at 7:53 AM Daniel J. H. ***@***.***>> wrote:> > > You are mounting your current directory to /data inside the container,> > and then you are accessing /data/model.toml.> >> > Can you make you you have a model.toml file in your current directory?> > Check the config dir.> >> > —> > You are receiving this because you authored the thread.> > Reply to this email directly, view it on GitHub> > <#215 (comment)>, or> > unsubscribe> > <https://github.com/notifications/unsubscribe-auth/AOOPU2QOJHPAAQH46ZWR7CTSVH5LDANCNFSM4U7H3GCQ>> > .> >> > > -- > You are receiving this because you commented.> Reply to this email directly or view it on GitHub:> #215 (comment)

[MaryElna]

Dear Daniel I want to use your trained data model on my drone images to extract building footprints, how can I do it? I read the" RoboSat Tanzania" tutorial, but it seems to work with specific downloaded input data! Best Regards Mary <https://www.openstreetmap.org/user/daniel-j-h/diary/44321> On Thu, Dec 17, 2020 at 8:49 PM Daniel J. H. <[email protected]> wrote:

…

It is the model configuration file in the robosat/config directory. You will need to modify the model and tge dataset config to your specific use case, because you are running the train command. On December 17, 2020 3:59:15 PM UTC, MaryElna ***@***.***> wrote: >Dear Daniel> >After some tries I couldn't find model.toml file in my system, would >you> >please guide me more or send me that file?> >Best> >Mary> >> >On Thu, Dec 17, 2020 at 7:53 AM Daniel J. H. ***@***.***>> >wrote:> >> >> You are mounting your current directory to /data inside the >container,> >> and then you are accessing /data/model.toml.> >>> >> Can you make you you have a model.toml file in your current >directory?> >> Check the config dir.> >>> >> —> >> You are receiving this because you authored the thread.> >> Reply to this email directly, view it on GitHub> >> ><#215 (comment)>, >or> >> unsubscribe> >> >< https://github.com/notifications/unsubscribe-auth/AOOPU2QOJHPAAQH46ZWR7CTSVH5LDANCNFSM4U7H3GCQ >> >> .> >>> >> >> >-- > >You are receiving this because you commented.> >Reply to this email directly or view it on GitHub:> >#215 (comment) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#215 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOOPU2UUFGOBSMR2JQYCW4DSVI4SBANCNFSM4U7H3GCQ> .

[daniel-j-h]

There is a pre-trained model for aerial imagery from Bavaria. You could use that and then fine tune on your own dataset (given you have labels). What you have to do is - get your dataset into slippy map format - either run inference with the trained model, or fine tune if you have labels

…

On December 23, 2020 8:54:15 AM UTC, MaryElna ***@***.***> wrote: Dear Daniel I want to use your trained data model on my drone images to> extract building footprints, how can I do it? I read the" RoboSat> Tanzania" tutorial, but it seems to work with specific downloaded input> data!> Best Regards> Mary <https://www.openstreetmap.org/user/daniel-j-h/diary/44321>> > On Thu, Dec 17, 2020 at 8:49 PM Daniel J. H. ***@***.***>> wrote:> > > It is the model configuration file in the robosat/config directory. You> > will need to modify the model and tge dataset config to your specific use> > case, because you are running the train command.> >> > On December 17, 2020 3:59:15 PM UTC, MaryElna ***@***.***>> > wrote:> > >Dear Daniel>> > >After some tries I couldn't find model.toml file in my system, would> > >you>> > >please guide me more or send me that file?>> > >Best>> > >Mary>> > >>> > >On Thu, Dec 17, 2020 at 7:53 AM Daniel J. H.> > ***@***.***>>> > >wrote:>> > >>> > >> You are mounting your current directory to /data inside the> > >container,>> > >> and then you are accessing /data/model.toml.>> > >>>> > >> Can you make you you have a model.toml file in your current> > >directory?>> > >> Check the config dir.>> > >>>> > >> —>> > >> You are receiving this because you authored the thread.>> > >> Reply to this email directly, view it on GitHub>> > >>> > ><#215 (comment)>,> > >or>> > >> unsubscribe>> > >>> > ><> > https://github.com/notifications/unsubscribe-auth/AOOPU2QOJHPAAQH46ZWR7CTSVH5LDANCNFSM4U7H3GCQ> > >>> > >> .>> > >>>> > >>> > >>> > >-- >> > >You are receiving this because you commented.>> > >Reply to this email directly or view it on GitHub:>> > >#215 (comment)> >> > —> > You are receiving this because you authored the thread.> > Reply to this email directly, view it on GitHub> > <#215 (comment)>, or> > unsubscribe> > <https://github.com/notifications/unsubscribe-auth/AOOPU2UUFGOBSMR2JQYCW4DSVI4SBANCNFSM4U7H3GCQ>> > .> >> > > -- > You are receiving this because you commented.> Reply to this email directly or view it on GitHub:> #215 (comment)

[MaryElna]

Dear Daniel I cant find the mentioned pre-trained model for aerial imagery from Bavarial, could you please explain me more, how can I find it? On Thu, Dec 24, 2020 at 4:22 PM Daniel J. H. <[email protected]> wrote:

…

There is a pre-trained model for aerial imagery from Bavaria. You could use that and then fine tune on your own dataset (given you have labels). What you have to do is - get your dataset into slippy map format - either run inference with the trained model, or fine tune if you have labels On December 23, 2020 8:54:15 AM UTC, MaryElna ***@***.***> wrote: >Dear Daniel I want to use your trained data model on my drone images >to> >extract building footprints, how can I do it? I read the" RoboSat> >Tanzania" tutorial, but it seems to work with specific downloaded >input> >data!> >Best Regards> >Mary <https://www.openstreetmap.org/user/daniel-j-h/diary/44321>> >> >On Thu, Dec 17, 2020 at 8:49 PM Daniel J. H. ***@***.***>> >wrote:> >> >> It is the model configuration file in the robosat/config directory. >You> >> will need to modify the model and tge dataset config to your specific >use> >> case, because you are running the train command.> >>> >> On December 17, 2020 3:59:15 PM UTC, MaryElna ***@***.***>> >> wrote:> >> >Dear Daniel>> >> >After some tries I couldn't find model.toml file in my system, >would> >> >you>> >> >please guide me more or send me that file?>> >> >Best>> >> >Mary>> >> >>> >> >On Thu, Dec 17, 2020 at 7:53 AM Daniel J. H.> >> ***@***.***>>> >> >wrote:>> >> >>> >> >> You are mounting your current directory to /data inside the> >> >container,>> >> >> and then you are accessing /data/model.toml.>> >> >>>> >> >> Can you make you you have a model.toml file in your current> >> >directory?>> >> >> Check the config dir.>> >> >>>> >> >> —>> >> >> You are receiving this because you authored the thread.>> >> >> Reply to this email directly, view it on GitHub>> >> >>> >> >><#215 (comment)>,> >> >or>> >> >> unsubscribe>> >> >>> >> ><> >> > https://github.com/notifications/unsubscribe-auth/AOOPU2QOJHPAAQH46ZWR7CTSVH5LDANCNFSM4U7H3GCQ > >> >>> >> >> .>> >> >>>> >> >>> >> >>> >> >-- >> >> >You are receiving this because you commented.>> >> >Reply to this email directly or view it on GitHub:>> >> >#215 (comment)> >>> >> —> >> You are receiving this because you authored the thread.> >> Reply to this email directly, view it on GitHub> >> ><#215 (comment)>, >or> >> unsubscribe> >> >< https://github.com/notifications/unsubscribe-auth/AOOPU2UUFGOBSMR2JQYCW4DSVI4SBANCNFSM4U7H3GCQ >> >> .> >>> >> >> >-- > >You are receiving this because you commented.> >Reply to this email directly or view it on GitHub:> >#215 (comment) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#215 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOOPU2TOFVVCJPTKUYCK643SWM2QTANCNFSM4U7H3GCQ> .

[daniel-j-h]

You can find it in the releases here https://github.com/mapbox/robosat/releases

…

On December 25, 2020 6:21:02 PM UTC, MaryElna ***@***.***> wrote: Dear Daniel> I cant find the mentioned pre-trained model for aerial imagery from> Bavarial, could you please explain me more, how can I find it?> > > On Thu, Dec 24, 2020 at 4:22 PM Daniel J. H. ***@***.***>> wrote:> > > There is a pre-trained model for aerial imagery from Bavaria. You could> > use that and then fine tune on your own dataset (given you have labels).> >> > What you have to do is> > - get your dataset into slippy map format> > - either run inference with the trained model, or fine tune if you have> > labels> >> >> >> >> > On December 23, 2020 8:54:15 AM UTC, MaryElna ***@***.***>> > wrote:> > >Dear Daniel I want to use your trained data model on my drone images> > >to>> > >extract building footprints, how can I do it? I read the" RoboSat>> > >Tanzania" tutorial, but it seems to work with specific downloaded> > >input>> > >data!>> > >Best Regards>> > >Mary <https://www.openstreetmap.org/user/daniel-j-h/diary/44321>>> > >>> > >On Thu, Dec 17, 2020 at 8:49 PM Daniel J. H.> > ***@***.***>>> > >wrote:>> > >>> > >> It is the model configuration file in the robosat/config directory.> > >You>> > >> will need to modify the model and tge dataset config to your specific> > >use>> > >> case, because you are running the train command.>> > >>>> > >> On December 17, 2020 3:59:15 PM UTC, MaryElna> > ***@***.***>>> > >> wrote:>> > >> >Dear Daniel>>> > >> >After some tries I couldn't find model.toml file in my system,> > >would>> > >> >you>>> > >> >please guide me more or send me that file?>>> > >> >Best>>> > >> >Mary>>> > >> >>>> > >> >On Thu, Dec 17, 2020 at 7:53 AM Daniel J. H.>> > >> ***@***.***>>>> > >> >wrote:>>> > >> >>>> > >> >> You are mounting your current directory to /data inside the>> > >> >container,>>> > >> >> and then you are accessing /data/model.toml.>>> > >> >>>>> > >> >> Can you make you you have a model.toml file in your current>> > >> >directory?>>> > >> >> Check the config dir.>>> > >> >>>>> > >> >> —>>> > >> >> You are receiving this because you authored the thread.>>> > >> >> Reply to this email directly, view it on GitHub>>> > >> >>>> > >>> > >><#215 (comment)>,>> > >> >or>>> > >> >> unsubscribe>>> > >> >>>> > >> ><>> > >>> > >> > https://github.com/notifications/unsubscribe-auth/AOOPU2QOJHPAAQH46ZWR7CTSVH5LDANCNFSM4U7H3GCQ> > >> > >> >>>> > >> >> .>>> > >> >>>>> > >> >>>> > >> >>>> > >> >-- >>> > >> >You are receiving this because you commented.>>> > >> >Reply to this email directly or view it on GitHub:>>> > >> >#215 (comment)>> > >>>> > >> —>> > >> You are receiving this because you authored the thread.>> > >> Reply to this email directly, view it on GitHub>> > >>> > ><#215 (comment)>,> > >or>> > >> unsubscribe>> > >>> > ><> > https://github.com/notifications/unsubscribe-auth/AOOPU2UUFGOBSMR2JQYCW4DSVI4SBANCNFSM4U7H3GCQ> > >>> > >> .>> > >>>> > >>> > >>> > >-- >> > >You are receiving this because you commented.>> > >Reply to this email directly or view it on GitHub:>> > >#215 (comment)> >> > —> > You are receiving this because you authored the thread.> > Reply to this email directly, view it on GitHub> > <#215 (comment)>, or> > unsubscribe> > <https://github.com/notifications/unsubscribe-auth/AOOPU2TOFVVCJPTKUYCK643SWM2QTANCNFSM4U7H3GCQ>> > .> >> > > -- > You are receiving this because you commented.> Reply to this email directly or view it on GitHub:> #215 (comment)

[MaryElna]

Dear Daniel please excuse me if I trouble you, I want to extract building footprints on raw drone images neither on orthophotos nor mosaic of them, is it possible to reach this goal with your trained model, if it is yes how can I prepare the slippy format of small cover drone images ? Best regards Mary On Sat, Dec 26, 2020 at 1:23 AM Daniel J. H. <[email protected]> wrote:

…

You can find it in the releases here https://github.com/mapbox/robosat/releases On December 25, 2020 6:21:02 PM UTC, MaryElna ***@***.***> wrote: >Dear Daniel> >I cant find the mentioned pre-trained model for aerial imagery from> >Bavarial, could you please explain me more, how can I find it?> >> >> >On Thu, Dec 24, 2020 at 4:22 PM Daniel J. H. ***@***.***>> >wrote:> >> >> There is a pre-trained model for aerial imagery from Bavaria. You >could> >> use that and then fine tune on your own dataset (given you have >labels).> >>> >> What you have to do is> >> - get your dataset into slippy map format> >> - either run inference with the trained model, or fine tune if you >have> >> labels> >>> >>> >>> >>> >> On December 23, 2020 8:54:15 AM UTC, MaryElna ***@***.***>> >> wrote:> >> >Dear Daniel I want to use your trained data model on my drone >images> >> >to>> >> >extract building footprints, how can I do it? I read the" RoboSat>> >> >Tanzania" tutorial, but it seems to work with specific downloaded> >> >input>> >> >data!>> >> >Best Regards>> >> >Mary <https://www.openstreetmap.org/user/daniel-j-h/diary/44321>>> >> >>> >> >On Thu, Dec 17, 2020 at 8:49 PM Daniel J. H.> >> ***@***.***>>> >> >wrote:>> >> >>> >> >> It is the model configuration file in the robosat/config >directory.> >> >You>> >> >> will need to modify the model and tge dataset config to your >specific> >> >use>> >> >> case, because you are running the train command.>> >> >>>> >> >> On December 17, 2020 3:59:15 PM UTC, MaryElna> >> ***@***.***>>> >> >> wrote:>> >> >> >Dear Daniel>>> >> >> >After some tries I couldn't find model.toml file in my system,> >> >would>> >> >> >you>>> >> >> >please guide me more or send me that file?>>> >> >> >Best>>> >> >> >Mary>>> >> >> >>>> >> >> >On Thu, Dec 17, 2020 at 7:53 AM Daniel J. H.>> >> >> ***@***.***>>>> >> >> >wrote:>>> >> >> >>>> >> >> >> You are mounting your current directory to /data inside the>> >> >> >container,>>> >> >> >> and then you are accessing /data/model.toml.>>> >> >> >>>>> >> >> >> Can you make you you have a model.toml file in your current>> >> >> >directory?>>> >> >> >> Check the config dir.>>> >> >> >>>>> >> >> >> —>>> >> >> >> You are receiving this because you authored the thread.>>> >> >> >> Reply to this email directly, view it on GitHub>>> >> >> >>>> >> >>> >> >>><#215 (comment) >,>> >> >> >or>>> >> >> >> unsubscribe>>> >> >> >>>> >> >> ><>> >> >>> >> >> >> > https://github.com/notifications/unsubscribe-auth/AOOPU2QOJHPAAQH46ZWR7CTSVH5LDANCNFSM4U7H3GCQ > >> >> >> >> >>>> >> >> >> .>>> >> >> >>>>> >> >> >>>> >> >> >>>> >> >> >-- >>> >> >> >You are receiving this because you commented.>>> >> >> >Reply to this email directly or view it on GitHub:>>> >> >> >>#215 (comment)>> >> >>>> >> >> —>> >> >> You are receiving this because you authored the thread.>> >> >> Reply to this email directly, view it on GitHub>> >> >>> >> >><#215 (comment)>,> >> >or>> >> >> unsubscribe>> >> >>> >> ><> >> > https://github.com/notifications/unsubscribe-auth/AOOPU2UUFGOBSMR2JQYCW4DSVI4SBANCNFSM4U7H3GCQ > >> >>> >> >> .>> >> >>>> >> >>> >> >>> >> >-- >> >> >You are receiving this because you commented.>> >> >Reply to this email directly or view it on GitHub:>> >> >#215 (comment)> >>> >> —> >> You are receiving this because you authored the thread.> >> Reply to this email directly, view it on GitHub> >> ><#215 (comment)>, >or> >> unsubscribe> >> >< https://github.com/notifications/unsubscribe-auth/AOOPU2TOFVVCJPTKUYCK643SWM2QTANCNFSM4U7H3GCQ >> >> .> >>> >> >> >-- > >You are receiving this because you commented.> >Reply to this email directly or view it on GitHub:> >#215 (comment) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#215 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOOPU2XDKYGNGUAFVYSI67TSWUCWRANCNFSM4U7H3GCQ> .

[daniel-j-h]

Without geo-referenced imagery, you will not be able to get vectorized shapes in world coordinates. You can still run the model on the images, tho, that will work. Unformtunately I really do not have the time to run this through with you step by step - I'm sorry.

[MaryElna]

thanks Daniel I found much easier deep learning platform and achieved desirable results!

…

On Tue, Jan 5, 2021, 3:59 PM Daniel J. H. ***@***.***> wrote: Without geo-referenced imagery, you will not be able to get vectorized shapes in world coordinates. You can still run the model on the images, tho, that will work. Unformtunately I really do not have the time to run this through with you step by step - I'm sorry. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#215 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AOOPU2XOMAEYJWDUEHOB4NLSYMA3RANCNFSM4U7H3GCQ> .