Groups
Sign in
Groups
marian-nmt
Conversations
About
Send feedback
Help
marian-nmt
Contact owners and managers
1–30 of 178
Welcome to the Marian NMT discussion group. Here you can post questions concerning usage, examples etc.
Mark all as read
Report group
0 selected
Jash Dubal
May 25
Automatic hyperparameter tuning on Marian
Hello, I've been working with Opus-Marian for the past few weeks, re-training en-fr base models
unread,
Automatic hyperparameter tuning on Marian
Hello, I've been working with Opus-Marian for the past few weeks, re-training en-fr base models
May 25
Jash Dubal
May 25
Adding input layers to NMT models
Hello, I've been working with Opus-Marian for the past few weeks, performing model re-training
unread,
Adding input layers to NMT models
Hello, I've been working with Opus-Marian for the past few weeks, performing model re-training
May 25
unhammer
Apr 18
Use transformer as lightweight classifier
I find the translateLocally models quite amazing (a 14MB CPU-decoded model giving nynorsk→english
unread,
Use transformer as lightweight classifier
I find the translateLocally models quite amazing (a 14MB CPU-decoded model giving nynorsk→english
Apr 18
Kai Piontek
,
Adrian Yalda
2
Mar 29
Retrain a non factor pretrained model with factors
Hello, You may have already seen this, but the only documentation I found on using Marian with
unread,
Retrain a non factor pretrained model with factors
Hello, You may have already seen this, but the only documentation I found on using Marian with
Mar 29
Marcin
12/5/23
Locking down this group to members only
Hi due to the incoming spam I have locked down this group to be member only. Users can still ask to
unread,
Locking down this group to members only
Hi due to the incoming spam I have locked down this group to be member only. Users can still ask to
12/5/23
Liling
7/18/23
[CFP] 3rd Workshop for NLP Open Source Software at EMNLP 2023
(apologies for cross-posting) ----- Workshop for NLP Open Source Software (NLP-OSS) 06 Dec 2023, Co-
unread,
[CFP] 3rd Workshop for NLP Open Source Software at EMNLP 2023
(apologies for cross-posting) ----- Workshop for NLP Open Source Software (NLP-OSS) 06 Dec 2023, Co-
7/18/23
Matt Russell (clojurian)
4/27/23
ShapeSizeException - looking for pointers!
Hi, I'm getting the following error (Marin NMT v1.12.0): [2023-04-26 12:13:17] Error: Caught std:
unread,
ShapeSizeException - looking for pointers!
Hi, I'm getting the following error (Marin NMT v1.12.0): [2023-04-26 12:13:17] Error: Caught std:
4/27/23
oravec...@gmail.com
4/11/23
training big transformer with mixed precision
We've been training a few big transformer models with marian 1.12.0 with about 25m segments on 4
unread,
training big transformer with mixed precision
We've been training a few big transformer models with marian 1.12.0 with about 25m segments on 4
4/11/23
oravec...@gmail.com
2
1/18/23
multi-node support in marian
In the meantime I've found these notes using 'mpirun' with some specific arguments: https
unread,
multi-node support in marian
In the meantime I've found these notes using 'mpirun' with some specific arguments: https
1/18/23
yvan Zhu
11/12/22
enc-depth and dec-depth
Hi Marian Team, Just want to write something i found when I use Marian and I would be appreciate it
unread,
enc-depth and dec-depth
Hi Marian Team, Just want to write something i found when I use Marian and I would be appreciate it
11/12/22
Sunil Kumar
,
Marcin Junczys-Dowmunt
12
8/16/22
Model behaviour changes drastically with change in Marian Version
Here, if I am not wrong: marian-dev/src/models/transformer.h at master · marian-nmt/marian-dev ·
unread,
Model behaviour changes drastically with change in Marian Version
Here, if I am not wrong: marian-dev/src/models/transformer.h at master · marian-nmt/marian-dev ·
8/16/22
Jindra Helcl
8/15/22
Fwd: MT Marathon 2022 in Prague: Last call for participation
(Apologies for cross-posting) This is the last call for participation on the 15th MT Marathon that
unread,
Fwd: MT Marathon 2022 in Prague: Last call for participation
(Apologies for cross-posting) This is the last call for participation on the 15th MT Marathon that
8/15/22
Jindra Helcl
7/11/22
15th MT Marathon: registration is now open!
(Apologies for cross-posting) This is a reminder that the 15th MT Marathon will take place in Prague
unread,
15th MT Marathon: registration is now open!
(Apologies for cross-posting) This is a reminder that the 15th MT Marathon will take place in Prague
7/11/22
sjbr...@gmail.com
,
Marcin Junczys-Dowmunt
9
6/10/22
different translations w/ different batch parameters?
Ah, that makes sense. Best, Marcin From: sjbr...@gmail.com Sent: Friday, June 10, 2022 11:54 PM To:
unread,
different translations w/ different batch parameters?
Ah, that makes sense. Best, Marcin From: sjbr...@gmail.com Sent: Friday, June 10, 2022 11:54 PM To:
6/10/22
Felipe Gasper
6/3/22
ICU MessageFormat?
Hello, Can this project translate ICU MessageFormat strings? For example: I have {0, plural, one {#
unread,
ICU MessageFormat?
Hello, Can this project translate ICU MessageFormat strings? For example: I have {0, plural, one {#
6/3/22
Sunil Kumar
,
Marcin Junczys-Dowmunt
2
5/26/22
Query:: Credits to Marian ToolKit
Hi, That's fine. Thanks for mentioning it in the FAQ. Even that would not be required, but
unread,
Query:: Credits to Marian ToolKit
Hi, That's fine. Thanks for mentioning it in the FAQ. Even that would not be required, but
5/26/22
yvan Zhu
5/26/22
pre-trained model from Marian
Hi, I wonder if there are some pre-trained model from offical Marian can be used? Best, Yvan
unread,
pre-trained model from Marian
Hi, I wonder if there are some pre-trained model from offical Marian can be used? Best, Yvan
5/26/22
Jindra Helcl
5/19/22
MT Marathon 2022 in Prague: Call for Participation
(Apologies for cross-posting) It is with great pleasure that we announce the 15th MT Marathon will
unread,
MT Marathon 2022 in Prague: Call for Participation
(Apologies for cross-posting) It is with great pleasure that we announce the 15th MT Marathon will
5/19/22
Jeremiah Chow
,
Marcin Junczys-Dowmunt
15
4/16/22
Error using guided-alignment and fp16 together
I have sent to your email, thanks! On Sunday, April 17, 2022 at 1:09:30 AM UTC+8 marc...@microsoft.
unread,
Error using guided-alignment and fp16 together
I have sent to your email, thanks! On Sunday, April 17, 2022 at 1:09:30 AM UTC+8 marc...@microsoft.
4/16/22
sjbr...@gmail.com
,
Jindra Helcl
3
10/15/21
Forming averaged marian models?
Thanks! perfect. On Thursday, October 14, 2021 at 12:01:03 PM UTC-4 jindra...@gmail.com wrote: Hello
unread,
Forming averaged marian models?
Thanks! perfect. On Thursday, October 14, 2021 at 12:01:03 PM UTC-4 jindra...@gmail.com wrote: Hello
10/15/21
sjbr...@gmail.com
9/28/21
How to measure Marian model size
I'd like to compare the size of my Marian models to models built w/ other tools. How can I find
unread,
How to measure Marian model size
I'd like to compare the size of my Marian models to models built w/ other tools. How can I find
9/28/21
Charlfaro mall
,
Roman Grundkiewicz
3
6/30/21
What is Marian supported language pairs?
Thank you for your reply and info. ในวันที่ วันพุธที่ 30 มิถุนายน ค.ศ. 2021 เวลา 15 นาฬิกา 08 นาที 38
unread,
What is Marian supported language pairs?
Thank you for your reply and info. ในวันที่ วันพุธที่ 30 มิถุนายน ค.ศ. 2021 เวลา 15 นาฬิกา 08 นาที 38
6/30/21
Eleftherios Avramidis
, …
Marcin Junczys-Dowmunt
7
6/23/21
How to disable Sentencepiece integration when pre-compiled
Hi Marcin, exactly, our source is sign language glosses, which is a symbolic language with no
unread,
How to disable Sentencepiece integration when pre-compiled
Hi Marcin, exactly, our source is sign language glosses, which is a symbolic language with no
6/23/21
Transcrobes Project
, …
Roman Grundkiewicz
8
5/26/21
CPU speed optimisations
As a side note, if you look for fast CPU-optimized NMT models trained with Marian, some are available
unread,
CPU speed optimisations
As a side note, if you look for fast CPU-optimized NMT models trained with Marian, some are available
5/26/21
Transcrobes Project
,
Roman Grundkiewicz
2
5/26/21
Translation truncation
Unless you set `--max-length` and `--max-length-crop`, that's most likely model related as it
unread,
Translation truncation
Unless you set `--max-length` and `--max-length-crop`, that's most likely model related as it
5/26/21
Eleftherios Avramidis
, …
Fateme Imani
9
5/15/21
Cannot compile Marian in docker
thank you . On Thu, May 13, 2021 at 2:48 AM Eleftherios Avramidis <eleftherio...@gmail.com
unread,
Cannot compile Marian in docker
thank you . On Thu, May 13, 2021 at 2:48 AM Eleftherios Avramidis <eleftherio...@gmail.com
5/15/21
Chatzitheodorou Konstantinos
,
Jörg Tiedemann
3
4/22/21
Pre-trained models
Great! Thanks a lot, Jörg Konstantinos On Wed, 21 Apr 2021 at 22:04, Jörg Tiedemann <tiedeman@
unread,
Pre-trained models
Great! Thanks a lot, Jörg Konstantinos On Wed, 21 Apr 2021 at 22:04, Jörg Tiedemann <tiedeman@
4/22/21
sjbr...@gmail.com
,
cep...@gmail.com
3
4/19/21
source and/or target factors
Thanks! I'll check this out. Steve On Saturday, April 17, 2021 at 6:37:58 AM UTC-4 cep...@gmail.
unread,
source and/or target factors
Thanks! I'll check this out. Steve On Saturday, April 17, 2021 at 6:37:58 AM UTC-4 cep...@gmail.
4/19/21
Clara GINOVART
,
Roman Grundkiewicz
3
3/30/21
Basic question
Thank you, Roman 😊 Clara De : maria...@googlegroups.com <maria...@googlegroups.com> De la
unread,
Basic question
Thank you, Roman 😊 Clara De : maria...@googlegroups.com <maria...@googlegroups.com> De la
3/30/21
Chatzitheodorou Konstantinos
, …
Chatzitheodorou Konstantinos
3
3/11/21
Training crashes in valiadation
Hi Roman I am using a smaller `--valid-mini-batch` now and it works fine. Best Konstantinos On Wed,
unread,
Training crashes in valiadation
Hi Roman I am using a smaller `--valid-mini-batch` now and it works fine. Best Konstantinos On Wed,
3/11/21