Wi " d dl Zd dlZ d dlmZ dZd dl m
Z
m Z d dl m
Z
d dlmZ d dlmZ d dlmZ dde
e dz d edz d
ed e eef d ed
ededej2 fdZdej6 de
e d ed
edef
dZ y# eef$ r dZY {w xY w) N)UMAPTF)ListUnion) MinMaxScaler)select_topic_representationtopics top_n_topics
use_ctfidf
custom_labelstitlewidthheightreturnc J | j }|j |j dk7 ddf }| t | }nL|'t |j j d| }n#t |j j }t | } | D
cg c] }
| j |
} }
t |t r| D
cg c]# }
t |
dgg| j | |
z % } }
| D
cg c]' }
dj |
dd D cg c] }|d c} ) } }
}| D cg c] }t | dk r|n|dd dz } }n{|r6| j *| D
cg c] }
| j |
| j z } }
nC| D
cg c]6 }
d j | j |
dd
D cg c] }|d c} 8 } }
}t t | j j! }t# j$ |D
cg c] }
|j' |
c}
}t) | j* | j, |d \ }}|| }t. r[|r9t1 j3 | }t5 d
d
ddd j3 | }n+t5 d
d
ddd j3 | }n t7 d t9 j: |dddf |dddf | | | d }t= || ||| S c c}
w c c}
w c c}w c c}}
w c c}w c c}
w c c}w c c}}
w c c}
w )a Visualize topics, their sizes, and their corresponding words.
This visualization is highly inspired by LDAvis, a great visualization
technique typically reserved for LDA.
Arguments:
topic_model: A fitted BERTopic instance.
topics: A selection of topics to visualize
top_n_topics: Only select the top n most frequent topics
use_ctfidf: Whether to use c-TF-IDF representations instead of the embeddings from the embedding model.
custom_labels: If bool, whether to use custom topic labels that were defined using
`topic_model.set_topic_labels`.
If `str`, it uses labels from other aspects, e.g., "Aspect1".
title: Title of the plot.
width: The width of the figure.
height: The height of the figure.
Examples:
To visualize the topics simply run:
```python
topic_model.visualize_topics()
```
Or if you want to save the resulting figure:
```python
fig = topic_model.visualize_topics()
fig.write_html("path/to/file.html")
```
N_ r z...z | T)r
output_ndarray hellingerrandom* ) n_neighbors n_componentsmetricinit random_statecosinez]UMAP is required to reduce the embeddings.. Please install it using `pip install umap-learn`. )xyTopicWordsSize) get_topic_freqlocr% listsortedto_list topic_sizes_
isinstancestrtopic_aspects_joinlencustom_labels_ _outliers get_topic
get_topicskeysnparrayindexr c_tf_idf_topic_embeddings_HAS_UMAPr
fit_transformr ModuleNotFoundErrorpd DataFrame_plotly_topic_visualization) topic_modelr r r
r r r
r freq_df
topic_listtopic frequencieswordslabelslabelword
all_topicsindices
embeddings c_tfidf_useddfs e/home/sietch6/trending-topics-pipeline/venv/lib/python3.12/site-packages/bertopic/plotting/_topics.pyvisualize_topicsrR s X ((*Gkk'--2-q01G
f !
--/
>?
--/0 J@JKu;++E2KKK-%fpq]b3u:t$% (B(B=(QRW(XXqqLQR&<58<=RROTUe#e*r/uSbzE/AAUU ;55AXbcu++EK4I4I,IJccaklX]1F1Fu1Mbq1Q RT!W RSll [ 3 3 5 : : <=>JhhVDE
((/DEG : %% J G$J % 55jAJ Ak_amJ'
AhX\^mJ'
" k
AqD!AqD!
B 'r:ueV LL_ Lq<RUc Rl EsN K6(K;5L L L( L #L&L- L9 LL LLrP rE c fd}| j j t | j j dz z
| j j t | j j dz z f}| j j t | j j dz z
| j j t | j j dz z f}t j | dddddddd d
d
d
d d d
}|j t dt dd |j dj g d D cg c] } t d| d || }
} t ddd i|
g} |j | d d d d!t d"d#$ d%||t d&d'd() d*d id*d i| + |j |, |j |, |j d-t | dz |d t | dz |d. t d/d0 1 |j d-|d t | dz |d. t | dz t d2d0 1 |j |d t | dz d3d d45 |j |d. t | dz d6d d47 |j d8d8d9 |_ |S c c} w ):zNCreate plotly-based visualization of topics with a slider for topic selection.c z | dk( rD cg c] }d }}nD cg c] }|| k( rdnd
}}d|gigS c c}w c c}w )Nr #B0BEC5redz marker.color )topic_selectedr marker_colorrF rE s rQ get_colorz._plotly_topic_visualization..get_color~ sV R /9:!I:L:YcdPUUn%)z Topic %{customdata[0]}z%{customdata[1]}zSize: %{customdata[2]})
hovertemplatezTopic update)rJ methodargsr t2 )activepadstepsgffffff?g ?centertop Black)r^ rc )textr$ r# xanchoryanchorfontwhite Rockwell)bgcolor font_size font_familyvisible)r r
r
hoverlabelxaxisyaxissliders)rangerd r" z#CFD8DC)rc r
)typex0y0x1y1rd z#9E9E9ED1
)r# r$ rs showarrowyshiftD2)r$ r# rs r xshiftNr )r# minabsmaxr$ pxscatter
update_tracesdictr1
update_layout update_xaxes update_yaxes add_shapesumadd_annotationdata) rP rE r r
r rZ x_rangey_rangefigrF rn r s ` rQ rB rB { s 2
S"$$((*,--
S"$$((*,--G
S"$$((*,--
S"$$((*,--G
**
b! !D$UQVW C T 1O8\ ]^ kk
`j jV[T&( %@P
Q jE j13)59:G g b0
2:N% % $ 7#7# MM
wIntertopic Distance Map r ) numpyr8 pandasr@ umapr r= ImportErrorr? typingr r sklearn.preprocessingr bertopic._utilsr plotly.expressexpressr plotly.graph_objects
graph_objectsgointboolr/ FigurerR rA rB rW rQ r s H . 7 !
$ # &+1fM I fM *fM fM
s# fM
fM fM
fM YYfMR[BLL [d3i [PS [\_ [il [g () Hs B B
B