Owner

deeppavlov

1 indexed repository · View on GitHub

AutoIntent
Automated machine learning for text classification

Python · 54 stars

Beginner friendly Everything

Open All

Extract `BaseAPIDescriptionScorer` shared by `LLMDescriptionScorer` and `TypeSafeDescriptionScorer` Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 48/100

deeppavlov/AutoIntent#357 · Opened Sep 19, 2026
Advisor treats `description_llm` / `description_typesafe` as embedder consumers and estimates no API cost for them Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 52/100

deeppavlov/AutoIntent#356 · Opened Sep 19, 2026
`LLMDescriptionScorer` prompt puts the utterance before the descriptions, defeating provider prefix caching Open

enhancement

Difficulty 3/5 1-2 days Newbie friendliness 72/100

deeppavlov/AutoIntent#355 · Opened Sep 19, 2026
`LLMDescriptionScorer` runs cache hits through the `max_per_second` limiter — 10 s per 100 cached utterances Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 58/100

deeppavlov/AutoIntent#354 · Opened Sep 19, 2026
API scorers: `predict()` fails inside a running event loop; `LLMDescriptionScorer` leaks an unclosed loop per instance on Python 3.14 Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 45/100

deeppavlov/AutoIntent#353 · Opened Sep 19, 2026
`OptimizationConfig.seed` is `PositiveInt` — `seed=0` is rejected while `Pipeline(seed=0)` accepts it Open

bug good first issue

Difficulty 2/5 1-3 hours Newbie friendliness 85/100

deeppavlov/AutoIntent#352 · Opened Sep 19, 2026
`Generator` lets `openai.APIStatusError` escape → one HTTP 4xx/5xx call aborts the whole `description_llm` run Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 42/100

deeppavlov/AutoIntent#351 · Opened Sep 19, 2026
Cache writes are not atomic (embeddings + structured outputs) → an interrupted/concurrent write poisons future reads Open

Difficulty 4/5 3-5 days Newbie friendliness 55/100

deeppavlov/AutoIntent#335 · Opened Jun 22, 2026
BertScorer multilabel fine-tuning collapses on sparse data (no pos_weight/focal, no warmup, brittle default LR) Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

deeppavlov/AutoIntent#324 · Opened Jun 20, 2026
BertScorer early-stopping default (scoring_f1@0.5) collapses on sparse multilabel Open

Difficulty 4/5 3-5 days Newbie friendliness 48/100

deeppavlov/AutoIntent#323 · Opened Jun 20, 2026
Upgrade transformers to 5.x (drops conftest mistral monkey-patch) Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 55/100

deeppavlov/AutoIntent#295 · Opened Jun 5, 2026
Устройство dump и load у Embedder и VectorIndex Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 15/100

deeppavlov/AutoIntent#265 · Opened Sep 20, 2025
Early stopping does work with lora. Probably Open

enhancement

Difficulty 1/5 Under an hour Newbie friendliness 25/100

deeppavlov/AutoIntent#264 · 1 comment · Opened Sep 13, 2025
Проблема с удалением артефактов (?) в _optimization_info.py Open

bug

Difficulty 4/5 3-5 days Newbie friendliness 35/100

deeppavlov/AutoIntent#262 · 1 comment · Opened Aug 28, 2025
Newer sentence-transformers support Open

enhancement

Difficulty 1/5 Under an hour Newbie friendliness 45/100

deeppavlov/AutoIntent#245 · Opened Jul 2, 2025
Add micro, weighted metrics Open

enhancement

Difficulty 3/5 1-2 days Newbie friendliness 30/100

deeppavlov/AutoIntent#217 · Opened May 22, 2025
Add logger message for setting EmbedderConfig Open

enhancement

Difficulty 2/5 1-3 hours Newbie friendliness 45/100

deeppavlov/AutoIntent#213 · Opened May 22, 2025
sampler option for each module Open

enhancement

Difficulty 4/5 3-5 days Newbie friendliness 42/100

deeppavlov/AutoIntent#193 · 1 comment · Opened May 2, 2025
Args name in signature and documentation mismatch Open

documentation

Difficulty 3/5 1-2 days Newbie friendliness 35/100

deeppavlov/AutoIntent#99 · Opened Jan 20, 2025
Pooling methods for cross-encoder Open

enhancement

Difficulty 5/5 Over a week Newbie friendliness 30/100

deeppavlov/AutoIntent#92 · Opened Jan 11, 2025

deeppavlov

Get new issues in your inbox