tessl install tessl/pypi-gensim@4.3.0Python library for topic modelling, document indexing and similarity retrieval with large corpora
Agent Success
Agent success rate when using this tile
78%
Improvement
Agent success rate improvement when using this tile compared to baseline
1.03x
Baseline
Agent success rate without this tile
76%
{
"context": "Evaluation checks how well the solution uses gensim's cross-lingual mapping tools to align embedding spaces, translate tokens, and score top-k accuracy for the provided spec. Emphasis is on using TranslationMatrix/KeyedVectors APIs rather than hand-rolled linear algebra or search.",
"type": "weighted_checklist",
"checklist": [
{
"name": "TranslationMatrix fit",
"description": "Creates and trains `gensim.models.translation_matrix.TranslationMatrix` (or `BackMappingTranslationMatrix` when appropriate) with provided source/target `KeyedVectors` and seed pairs instead of manual regression.",
"max_score": 35
},
{
"name": "Seed filtering",
"description": "Leverages `KeyedVectors` vocabulary checks or TranslationMatrix preprocessing to drop seed pairs missing from either model while still fitting when enough valid pairs remain.",
"max_score": 15
},
{
"name": "Translate topn",
"description": "Uses `TranslationMatrix.translate(..., topn=...)` to return ordered candidate translations per source token and handles out-of-vocabulary queries by returning an empty list before applying further logic.",
"max_score": 25
},
{
"name": "Accuracy from translate",
"description": "Computes top-k accuracy by reusing `TranslationMatrix.translate` outputs (or BackMapping translate) instead of manual similarity loops, counting hits where the gold target appears within `k`.",
"max_score": 15
},
{
"name": "KeyedVectors usage",
"description": "Retrieves vectors and vocabulary membership via `gensim.models.KeyedVectors` APIs (indexing, `__contains__`, `get_vector`) rather than custom storage when preparing seeds or translations.",
"max_score": 10
}
]
}