transformers_issues_topics

This is a BERTopic model. BERTopic is a flexible and modular topic modeling framework that allows for the generation of easily interpretable topics from large datasets.

Usage

To use this model, please install BERTopic:

pip install -U bertopic

You can use the model as follows:

from bertopic import BERTopic
topic_model = BERTopic.load("u571/transformers_issues_topics")

topic_model.get_topic_info()

Topic overview

• Number of topics: 30
• Number of training documents: 9000

Click here for an overview of all topics.

Topic ID	Topic Keywords	Topic Frequency	Label
-1	tensorflow - pytorch - tokenizers - tokenizer - bert	10	-1_tensorflow_pytorch_tokenizers_tokenizer
0	tokenizer - tokenizers - tokenization - tokenize - token	2107	0_tokenizer_tokenizers_tokenization_tokenize
1	cuda - memory - gpu - gpus - tensorflow	1271	1_cuda_memory_gpu_gpus
2	tf - trainer - tf2 - tpu - trainertrain	901	2_tf_trainer_tf2_tpu
3	summarization - summaries - summary - sentences - sentencepiece	543	3_summarization_summaries_summary_sentences
4	modelcard - modelcards - card - model - cards	483	4_modelcard_modelcards_card_model
5	gpt2 - gpt2tokenizer - gpt2tokenizerfast - gpt2xl - gpt	431	5_gpt2_gpt2tokenizer_gpt2tokenizerfast_gpt2xl
6	xlnet - xlnetlmheadmodel - xlm - xlmr - xla	423	6_xlnet_xlnetlmheadmodel_xlm_xlmr
7	typos - typo - fix - fixed - correction	334	7_typos_typo_fix_fixed
8	s2s - exampless2s - seq2seqtrainer - seq2seq - runseq2seq	324	8_s2s_exampless2s_seq2seqtrainer_seq2seq
9	testing - tests - test - slow - ci	316	9_testing_tests_test_slow
10	readmemd - readmetxt - readme - modelcard - file	296	10_readmemd_readmetxt_readme_modelcard
11	transformerscli - transformers - transformer - transformerxl - importerror	262	11_transformerscli_transformers_transformer_transformerxl
12	ner - pipeline - pipelines - nerpipeline - fillmaskpipeline	223	12_ner_pipeline_pipelines_nerpipeline
13	rag - ragtokenforgeneration - ragmodel - ragsequenceforgeneration - tokenizer	166	13_rag_ragtokenforgeneration_ragmodel_ragsequenceforgeneration
14	trainertrain - checkpoint - checkpoints - trainer - training	146	14_trainertrain_checkpoint_checkpoints_trainer
15	datacollatorforlanguagemodeling - datacollatorforpermutationlanguagemodeling - datacollatorforlanguagemodelling - labelsmoothingfactor - maskedlmlabels	128	15_datacollatorforlanguagemodeling_datacollatorforpermutationlanguagemodeling_datacollatorforlanguagemodelling_labelsmoothingfactor
16	onnx - onnxonnxruntime - onnxexport - 04onnxexport - 04onnxexportipynb	99	16_onnx_onnxonnxruntime_onnxexport_04onnxexport
17	longformer - longformers - longform - longformerforqa - longformerlayer	84	17_longformer_longformers_longform_longformerforqa
18	benchmark - benchmarks - results - datasets - v100a100	78	18_benchmark_benchmarks_results_datasets
19	generationbeamsearchpy - generatebeamsearch - beamsearch - nonbeamsearch - beam	75	19_generationbeamsearchpy_generatebeamsearch_beamsearch_nonbeamsearch
20	wav2vec2 - wav2vec - wav2vec20 - wav2vec2forctc - wav2vec2xlrswav2vec2	71	20_wav2vec2_wav2vec_wav2vec20_wav2vec2forctc
21	flax - flaxelectraformaskedlm - flaxelectraforpretraining - flaxjax - flaxelectramodel	52	21_flax_flaxelectraformaskedlm_flaxelectraforpretraining_flaxjax
22	wandbproject - wandb - wandbcallback - wandbdisabled - wandbdisabledtrue	47	22_wandbproject_wandb_wandbcallback_wandbdisabled
23	cachedir - cache - cachedpath - caching - cached	37	23_cachedir_cache_cachedpath_caching
24	layoutlm - layout - layoutlmtokenizer - layoutlmbaseuncased - tf	33	24_layoutlm_layout_layoutlmtokenizer_layoutlmbaseuncased
25	dict - dictstr - returndict - parse - arguments	18	25_dict_dictstr_returndict_parse
26	pplm - pr - deprecated - variable - ppl	16	26_pplm_pr_deprecated_variable
27	colab - cola - crashes - crash - tcmalloc	14	27_colab_cola_crashes_crash
28	ctrl - ctrlsum - shortcuts - model - navigate	12	28_ctrl_ctrlsum_shortcuts_model

Training hyperparameters

• calculate_probabilities: False
• language: english
• low_memory: False
• min_topic_size: 10
• n_gram_range: (1, 1)
• nr_topics: 30
• seed_topic_list: None
• top_n_words: 10
• verbose: True

Framework versions

• Numpy: 1.23.5
• HDBSCAN: 0.8.33
• UMAP: 0.5.3
• Pandas: 1.5.3
• Scikit-Learn: 1.2.2
• Sentence-transformers: 2.2.2
• Transformers: 4.32.0
• Numba: 0.56.4
• Plotly: 5.15.0
• Python: 3.10.12