stanza icon indicating copy to clipboard operation
stanza copied to clipboard

How to train our own models in sentimental analysis

Open kalpa277 opened this issue 3 years ago • 5 comments

python -m stanza.utils.datasets.prepare_${module}_treebank ${corpus} ${other_args} what we have to pass in module,corpus,other_args

C:\Users\Kalpataru\Desktop\stanza-main\stanza\utils\datasets\ner>python -m stanza.utils.datasets.ner.prepare_ner_dataset fi_turku

FileNotFoundError: Cannot find train component of fi_turku in extern_data/ner\fi_turku\train.tsv

I want to train models as per our every time updated data

kalpa277 avatar Jun 07 '22 18:06 kalpa277

Have you downloaded the data? If so, where did you put it?

In general, please include the entire stack trace when making an issue.

On Tue, Jun 7, 2022 at 11:11 AM kalpa277 @.***> wrote:

python -m stanza.utils.datasets.prepare_${module}_treebank ${corpus} ${other_args} what we have to pass in module,corpus,other_args

C:\Users\Kalpataru\Desktop\stanza-main\stanza\utils\datasets\ner>python -m stanza.utils.datasets.ner.prepare_ner_dataset fi_turku

FileNotFoundError: Cannot find train component of fi_turku in extern_data/ner\fi_turku\train.tsv

I want to train models as per our every time updated data

— Reply to this email directly, view it on GitHub https://github.com/stanfordnlp/stanza/issues/1045, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2AYWIFXFX6XFXHP6VN4TLVN6GDHANCNFSM5YD2NAVQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

AngledLuffa avatar Jun 07 '22 19:06 AngledLuffa

Yes I have downloaded the data.I am not aware about how to put it can you share some blog or tutorial that will help me to figure it out how to train our own model.

I have downloaded https://github.com/stanfordnlp/stanza.

kalpa277 avatar Jun 08 '22 03:06 kalpa277

You're already looking through the page on how to retrain models, it seems. However, the error message is pretty much telling you where it's looking for the data.

You can change the root directory for where to put the fi_turku data by setting NERBASE as an environment variable. There is a bit of description on the environment variables here:

https://stanfordnlp.github.io/stanza/training.html#setting-environment-variables

I will add a bit more to it about Windows in a moment.

AngledLuffa avatar Jun 08 '22 04:06 AngledLuffa

I added some more doc - lmk if it helps

also here

https://stanfordnlp.github.io/stanza/training.html#ner-data

and the title of the thread is "sentiment", but you're asking about NER. which are you more interested in?

AngledLuffa avatar Jun 08 '22 07:06 AngledLuffa

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

stale[bot] avatar Aug 10 '22 00:08 stale[bot]

This issue has been automatically closed due to inactivity.

stale[bot] avatar Aug 31 '22 11:08 stale[bot]

For the record:

https://stanfordnlp.github.io/stanza/new_language_sentiment.html

https://stanfordnlp.github.io/stanza/new_language_ner.html

AngledLuffa avatar Aug 31 '22 11:08 AngledLuffa