Hugging Face's logo

Datasets:
jamalinu
/
amazigh-berber-nlp

Tasks:
Token Classification
Translation
Languages:
Standard Moroccan Tamazight
Catalan
Size:
n<1K
Tags:
amazigh
tamazight
berber
low-resource-nlp
morphology
grammar
License:
Dataset card Data Studio Files
xet
 Community
Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 8 new columns ({'syntax', 'phonology', 'territory', 'language', 'speakers', 'comparison_catalan', 'family', 'morphology'}) and 3 missing columns ({'translation', 'berber', 'type'}).

This happened while the json dataset builder was generating data using

hf://datasets/jamalinu/amazigh-berber-nlp/grammar_summary.json (at revision dd41969d00bdda9e563243170a0258d9ccde87f0), [/tmp/hf-datasets-cache/medium/datasets/20342831577343-config-parquet-and-info-jamalinu-amazigh-berber-n-89d1a9fb/hub/datasets--jamalinu--amazigh-berber-nlp/snapshots/dd41969d00bdda9e563243170a0258d9ccde87f0/example_sentences.json (origin=hf://datasets/jamalinu/amazigh-berber-nlp@dd41969d00bdda9e563243170a0258d9ccde87f0/example_sentences.json), /tmp/hf-datasets-cache/medium/datasets/20342831577343-config-parquet-and-info-jamalinu-amazigh-berber-n-89d1a9fb/hub/datasets--jamalinu--amazigh-berber-nlp/snapshots/dd41969d00bdda9e563243170a0258d9ccde87f0/grammar_summary.json (origin=hf://datasets/jamalinu/amazigh-berber-nlp@dd41969d00bdda9e563243170a0258d9ccde87f0/grammar_summary.json), /tmp/hf-datasets-cache/medium/datasets/20342831577343-config-parquet-and-info-jamalinu-amazigh-berber-n-89d1a9fb/hub/datasets--jamalinu--amazigh-berber-nlp/snapshots/dd41969d00bdda9e563243170a0258d9ccde87f0/phoneme_table.json (origin=hf://datasets/jamalinu/amazigh-berber-nlp@dd41969d00bdda9e563243170a0258d9ccde87f0/phoneme_table.json)]

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1890, in _prepare_split_single
                  writer.write_table(table)
                File "/usr/local/lib/python3.12/site-packages/datasets/arrow_writer.py", line 760, in write_table
                  pa_table = table_cast(pa_table, self._schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2272, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              language: string
              family: string
              speakers: string
              territory: string
              source: string
              phonology: struct<n_vowels: int64, vowels: list<item: string>, n_consonants: string, features: list<item: strin (... 3 chars omitted)
                child 0, n_vowels: int64
                child 1, vowels: list<item: string>
                    child 0, item: string
                child 2, n_consonants: string
                child 3, features: list<item: string>
                    child 0, item: string
              morphology: struct<type: string, features: list<item: string>, verb_aspects: struct<perfectiu: string, imperfect (... 30 chars omitted)
                child 0, type: string
                child 1, features: list<item: string>
                    child 0, item: string
                child 2, verb_aspects: struct<perfectiu: string, imperfectiu: string, habitual: string>
                    child 0, perfectiu: string
                    child 1, imperfectiu: string
                    child 2, habitual: string
              syntax: struct<basic_order: string, features: list<item: string>>
                child 0, basic_order: string
                child 1, features: list<item: string>
                    child 0, item: string
              comparison_catalan: struct<similar: list<item: string>, different: list<item: string>>
                child 0, similar: list<item: string>
                    child 0, item: string
                child 1, different: list<item: string>
                    child 0, item: string
              to
              {'berber': Value('string'), 'translation': Value('string'), 'type': Value('string'), 'source': Value('string')}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1347, in compute_config_parquet_and_info_response
                  parquet_operations = convert_to_parquet(builder)
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 980, in convert_to_parquet
                  builder.download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 884, in download_and_prepare
                  self._download_and_prepare(
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 947, in _download_and_prepare
                  self._prepare_split(split_generator, **prepare_split_kwargs)
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1739, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1892, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 8 new columns ({'syntax', 'phonology', 'territory', 'language', 'speakers', 'comparison_catalan', 'family', 'morphology'}) and 3 missing columns ({'translation', 'berber', 'type'}).
              
              This happened while the json dataset builder was generating data using
              
              hf://datasets/jamalinu/amazigh-berber-nlp/grammar_summary.json (at revision dd41969d00bdda9e563243170a0258d9ccde87f0), [/tmp/hf-datasets-cache/medium/datasets/20342831577343-config-parquet-and-info-jamalinu-amazigh-berber-n-89d1a9fb/hub/datasets--jamalinu--amazigh-berber-nlp/snapshots/dd41969d00bdda9e563243170a0258d9ccde87f0/example_sentences.json (origin=hf://datasets/jamalinu/amazigh-berber-nlp@dd41969d00bdda9e563243170a0258d9ccde87f0/example_sentences.json), /tmp/hf-datasets-cache/medium/datasets/20342831577343-config-parquet-and-info-jamalinu-amazigh-berber-n-89d1a9fb/hub/datasets--jamalinu--amazigh-berber-nlp/snapshots/dd41969d00bdda9e563243170a0258d9ccde87f0/grammar_summary.json (origin=hf://datasets/jamalinu/amazigh-berber-nlp@dd41969d00bdda9e563243170a0258d9ccde87f0/grammar_summary.json), /tmp/hf-datasets-cache/medium/datasets/20342831577343-config-parquet-and-info-jamalinu-amazigh-berber-n-89d1a9fb/hub/datasets--jamalinu--amazigh-berber-nlp/snapshots/dd41969d00bdda9e563243170a0258d9ccde87f0/phoneme_table.json (origin=hf://datasets/jamalinu/amazigh-berber-nlp@dd41969d00bdda9e563243170a0258d9ccde87f0/phoneme_table.json)]
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

berber
string
translation
string
type
string
source
string
ikerz
ha llaurat¶ ikerrez
lexical_example
Lamuela (2005)
yerzu
ha buscat¶ irezzu
lexical_example
Lamuela (2005)
em a la frase
en Joan, no l¶he vist¶, en qu0 en Joan /s presentat com a refer0ncia del que es dir) a continuaci8 vg. † ... .  === PAGE BREAK === === PAGE BREAK === . DESCRIPCIÏ GRAMATICAL .. FONÊTICA I FONOLOGIA ... L¶INVENTARI DE LES CONSONANTS BERBERS .... El conjunt de les consonants de les varietats berbe...
lexical_example
Lamuela (2005)
ssekkwar w gg ed
sucre¶
lexical_example
Lamuela (2005)
ameqqran
coure¶ verb
lexical_example
Lamuela (2005)
sers
posar¶ izi
lexical_example
Lamuela (2005)
muc
gat¶ raja 
lexical_example
Lamuela (2005)
ecc
menjar¶ i÷  ijj
lexical_example
Lamuela (2005)
aqnenni
conill¶ aembub
lexical_example
Lamuela (2005)
axxam
habitaci8¶ La fricativa faringal sorda, hŽ [Ò
lexical_example
Lamuela (2005)
wah
s3¶  === PAGE BREAK === L¶amazig t/ les dues semivocals que trobem en catal), y [j
lexical_example
Lamuela (2005)
baw awal
paraula¶
lexical_example
Lamuela (2005)
amen
creure¶ mmet
lexical_example
Lamuela (2005)
ineqq
mata, est) matant¶ raja
lexical_example
Lamuela (2005)
ejj
deixar¶ seqsa
lexical_example
Lamuela (2005)
qqed
cremar¶ dŽew
lexical_example
Lamuela (2005)
w izegg a
travessa, est) travessant¶ En moltes varietats, sobretot del nord, les consonants oclusives laxes coneixen realitzacions fricatives, segons un fenomen semblant al de l¶articulaci8 de b, d i g del catal) i de l¶espanyol en paraules com acabar, poder i pagar. La difer0ncia /s que en berber la fricativitzaci8...
lexical_example
Lamuela (2005)
ibawen
faves¶ bael
lexical_example
Lamuela (2005)
tanyart
front¶ tamart
lexical_example
Lamuela (2005)
adef
entrar¶ fud
lexical_example
Lamuela (2005)
sken
ensenyar¶ amkan
lexical_example
Lamuela (2005)
aker
robar¶ g w [w:
lexical_example
Lamuela (2005)
tagwurt
porta¶ La t i la d mantenen l¶articulaci8 oclusiva quan entren en contacte entre elles o darrere d¶una n. La m davant de t o de d es pronuncia com una n:  ad tased [at:asè
lexical_example
Lamuela (2005)
tandint
Tandint
lexical_example
Lamuela (2005)
amti
¸anti
lexical_example
Lamuela (2005)
ittazzel
corre, est) corrent¶ tidett
lexical_example
Lamuela (2005)
awal
paraula¶ twala
lexical_example
Lamuela (2005)
yelli
filla¶ tamellalt
lexical_example
Lamuela (2005)
tamellalt
ou¶ weltma
lexical_example
Lamuela (2005)
ajer
superar, /sser m/s gran¶ ahŽenjir
lexical_example
Lamuela (2005)
tamurt
terra¶ Algunes caracter3stiques del consonantisme rifeny poden influenciar la pronunciaci8 del catal). $s el cas de l¶abs0ncia o de la distribuci8 limitada dels sons de ll i de ny. Pel que fa al primer, el fet que no hi sigui en berber es combina amb la influ0ncia de l¶espanyol perqu0 pu...
lexical_example
Lamuela (2005)
cula
p[
lexical_example
Lamuela (2005)
dat
hem qu[
lexical_example
Lamuela (2005)
rtida
div[
lexical_example
Lamuela (2005)
nic
mec[a
lexical_example
Lamuela (2005)
goti
b[i
lexical_example
Lamuela (2005)
ntre
d[i
lexical_example
Lamuela (2005)
njo
m[e
lexical_example
Lamuela (2005)
lla
par[E
lexical_example
Lamuela (2005)
d r D e
f g gw ÷  w [f
lexical_example
Lamuela (2005)
tarwa
fills¶ ibawen
lexical_example
Lamuela (2005)
ecc
menjar¶ da
lexical_example
Lamuela (2005)
aqemmum
boca¶ fus
lexical_example
Lamuela (2005)
egg
fer¶ ggwed
lexical_example
Lamuela (2005)
ijj
un¶ aembub
lexical_example
Lamuela (2005)
raja
esperar¶ kkes
lexical_example
Lamuela (2005)
ssekkwar
sucre¶ ili
lexical_example
Lamuela (2005)
marra
tot¶ ini
lexical_example
Lamuela (2005)
pisu
pis¶ aqnenni
lexical_example
Lamuela (2005)
ameqqran
gran¶ ssad
lexical_example
Lamuela (2005)
irar
jugar¶ rŽrŽuz
lexical_example
Lamuela (2005)
sers
posar¶ nqesŽ
lexical_example
Lamuela (2005)
tabratt
carta¶ atŽtŽas
lexical_example
Lamuela (2005)
tidett
veritat¶ fud
lexical_example
Lamuela (2005)
hwa
baixar¶ axxam
lexical_example
Lamuela (2005)
taxzant
armari¶ yazŽidŽ
lexical_example
Lamuela (2005)
izi
mosca¶ zŽer
lexical_example
Lamuela (2005)
agezzar
carnisser¶ .. MORFOLOGIA ... L¶ESTRUCTURA DELS MOTS La primera caracter3stica a destacar de la morfologia de la llengua berber /s el conjunt de recursos que utilitza per a la flexi8 i la derivaci8 els derivats i les formes de flexi8 es formen per prefixaci8 i sufixaci8, i tamb/ modificant la dotaci8 voc)lica del...
lexical_example
Lamuela (2005)
aserdun
mula¶ iCCCaC PL. iserdan
lexical_example
Lamuela (2005)
argaz
home¶ b. aCCCaC SG. iCCaCn PL. irgazen
lexical_example
Lamuela (2005)
anegmar
ca.ador¶ iCCCaCn PL. inegmaren
lexical_example
Lamuela (2005)
s al seu torn un nom derivat del verb gmer
ca.ar¶ ±es tracta del radical sol, que coincideix amb la forma de l¶imperatiu singular vg. † ... ± format per l¶adjunci8 del prefix an i la inserci8 de la vocal a davant de l¶=ltima consonant per un procediment semblant podem formar un nom d¶acci8 adjuntant el prefix a i inser...
lexical_example
Lamuela (2005)
anegmar
ca.ador¶ CCC RAD. aCCaC nom d¶acci8 . gmer radical de
lexical_example
Lamuela (2005)
agmar
ca.a¶ Si prenem el verb que significa
lexical_example
Lamuela (2005)
rzu radical de
buscar¶ yerzu
lexical_example
Lamuela (2005)
rzu radical de
buscar¶  yCCu  MASC. PERF. irezzu
lexical_example
Lamuela (2005)
argaz
home¶ ils
lexical_example
Lamuela (2005)
uzzal
ferro¶ S8n femenins:  tamart
lexical_example
Lamuela (2005)
tsa
fetge¶ tili
lexical_example
Lamuela (2005)
atbir
colom¶ aserdun
lexical_example
Lamuela (2005)
ifri
cova¶ iger
lexical_example
Lamuela (2005)
udem
cara¶ uccen
lexical_example
Lamuela (2005)
fad
set¶ En rifeny, d¶altra banda, falta sovint la a inicial present en d¶altres varietats berbers:  dŽar adŽar
lexical_example
Lamuela (2005)
baw abaw
fava¶ yazŽidŽ ayazŽidŽ
lexical_example
Lamuela (2005)
lmuallim
mestre¶ lxemis
lexical_example
Lamuela (2005)
lkursi
cadira¶ ssuq
lexical_example
Lamuela (2005)
ttarix
data¶ ccarr
lexical_example
Lamuela (2005)
tabratt
carta¶ taddart
lexical_example
Lamuela (2005)
enjayt
cullera¶  === PAGE BREAK === b. tisit
lexical_example
Lamuela (2005)
tudart
vida¶ tcuqqet
lexical_example
Lamuela (2005)
timessi
foc¶ timesna
lexical_example
Lamuela (2005)
takecca
cuc¶ Tal com ocorre amb la a inicial dels masculins, tamb/ la a del prefix femen3 ta falta sovint en rifeny dins de paraules que la tenen en d¶altres varietats:  tmart tamart
lexical_example
Lamuela (2005)
tsa tasa
fetge¶ tyazŽidŽtŽ tayazŽidŽtŽ
lexical_example
Lamuela (2005)
tamenzut
filla gran¶ ajjar
lexical_example
Lamuela (2005)
tajjart
ve6na¶ afunas
lexical_example
Lamuela (2005)
tafunast
vaca¶ alem
lexical_example
Lamuela (2005)
emt
camella¶ yazŽidŽ
lexical_example
Lamuela (2005)
aqemmum
boca¶ taqemmumt
lexical_example
Lamuela (2005)
afer
fulla¶ tafert
lexical_example
Lamuela (2005)
izi
mosca¶ tizit
lexical_example
Lamuela (2005)
taddart
casa¶ addar
lexical_example
Lamuela (2005)
enjayt
cullera¶ aenja
lexical_example
Lamuela (2005)
takecca
cuc¶ akeccaw
lexical_example
Lamuela (2005)
argaz
home¶ tamart  tametŽtŽut
lexical_example
Lamuela (2005)
baba
pare¶ yemma
lexical_example
Lamuela (2005)
mmi
fill¶ yelli
lexical_example
Lamuela (2005)
ikerri
molt8¶ tixsi
lexical_example
Lamuela (2005)
atrus
boc¶ tatŽ
lexical_example
Lamuela (2005)
yis tagmart
egua¶
lexical_example
Lamuela (2005)
argaz
home¶ irgazen afunas
lexical_example
Lamuela (2005)
End of preview.

Amazigh/Berber NLP Dataset

Ghosts in the Font: Reviving a Low-Resource Language for AI

A structured NLP dataset and toolkit for the Amazigh/Berber language (Tarifit/Riffian dialect focus), extracted and digitized from:

Lamuela, X. (2005). El Berber: Estudi Comparatiu entre la Gramàtica del Català i la del Berber o Amazig. Universitat de Girona / Generalitat de Catalunya. ISBN 84-393-6956-5.

What's inside

File Content
phoneme_table.json Full consonant inventory (32 consonants) with IPA, place/manner of articulation, pharyngealization features
grammar_summary.json Structured grammar: phonology, morphology, syntax, comparison with Catalan
example_sentences.json ~200 example sentences extracted from the original grammar book

Key linguistic features covered

  • 3-vowel system /a, i, u/ vs. 7 vowels in Catalan
  • Pharyngealized (emphatic) consonants: ṭ ḍ ṣ ẓ ṛ ḷ
  • Aspect-based verb system (not tense-based): perfective / imperfective / habitual-durative
  • Free state vs. Annexed state morphology (a- → u- prefix alternation)
  • VSO word order (Verb-Subject-Object)
  • Circumfixal negation: ur...ara
  • Bilingual vocabulary: Catalan ↔ Berber

Interactive demo

🚀 Try the live NLP toolkit on Hugging Face Spaces: Berber NLP Toolkit

Features: tokenizer, morphological analyzer, phonological analyzer, syntax annotator, bilingual dictionary, learner error analyzer.

Language

  • ISO 639-3: zgh (Standard Moroccan Tamazight)
  • Dialect focus: Tarifit / Riffian (northern Morocco)
  • Family: Afroasiatic > Berber
  • Estimated speakers: 15–20 million

Citation

If you use this dataset, please cite the original source:

```bibtex @book{lamuela2005berber, author = {Lamuela Garcia, F. Xavier}, title = {El Berber: Estudi Comparatiu entre la Gramàtica del Català i la del Berber o Amazig}, publisher = {Generalitat de Catalunya, Departament de Benestar i Família}, year = {2005}, edition = {2a edició}, series = {Llengua, immigració i ensenyament del català, núm. 5}, isbn = {84-393-6956-5}, } ```
Downloads last month
19
Total file size:
56.9 kB

Space using jamalinu/amazigh-berber-nlp 1