使用BERTopic进行一些研究任务时，我得到了：
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (2,) + inhomogeneous part
是否有计划支持 BERTopic 和 numpy >1.23.5 以及 numba > 0.56.4 的 BERTopic ?
这将使得在没有手动降级(!pip install numba==0.56.4)的情况下更容易在Colab中使用BERTopic,尤其是对于那些没有Colab.pro的人。
谢谢。
[注意]
相关问题：closed | [ #1697 , #1602 , #1309 ]
相关问题：open | [ #1814 , #1799 , #1684 , #1584 , #1421 , #1269 ]

SO: https://stackoverflow.com/a/76504825

PS:我收集到的一个解决方法是降级numpy!
在问题 #1421 中：@aaron-imani 在 _guided_topic_modelling() 中暗示了 np.average() 。@MaartenGr指出，将 numba 设置为 0.56.4 或更早的时间应该理想地解决了这个问题。
感谢 @MaartenGr 在之前的问题的参与和建议。

环境: colab.research.google.com

注：(截至2024年4月26日，Colab运行的是 python: 3.10.12, bertopic: 0.16.1, numpy: 1.25.2, numba: 0.58.1 )

BERTopic:

# Install BERTopic
!pip install bertopic
# Load libraries
from bertopic import BERTopic
from bertopic.representation import KeyBERTInspired
from bertopic.vectorizers import ClassTfidfTransformer
from umap import UMAP
from sentence_transformers import SentenceTransformer

[代码片段]

# Train BERTopic
topic_model_03 = BERTopic(
    verbose=True,
    min_topic_size= 5, #10,  #12,  #15,  ## the higher, the lower the clusters/topics
    nr_topics = 4, #5, ## reduce the initial number of topics
    #seed_topic_list=seed_topic_list,  ##ValueError: ...
    n_gram_range = (1,3), ## n-gram range for the CountVectorizer
    zeroshot_topic_list=zeroshot_topic_list,
    zeroshot_min_similarity=.35,  #.55,  #.75,  #.85,
    embedding_model="thenlper/gte-small",  ## pass string directly to sbert sentence-transformers models
    #umap_model = umap_model, ## dimensionality
    ctfidf_model = ctfidf_model,
    representation_model=KeyBERTInspired(),
    )

种子主题看起来像这样(部分被屏蔽)

### Try with Guided Representation
seed_topic_list = [["sustainability", "...", "sustain", "...", "..."],
                   ["climate change", "climate", "...", "...", "...", "...", "ozone"],
                   ["social justice", "social", "...", "...", "...", "...", "...", "..."],
                   ["net zero", "..."]
                   ]
## fit and transform
#topics_03, probs_03 = topic_model_03.fit_transform(docs)
## visualise documents' topics spread
#topic_model_03.visualize_documents(docs)