h1alexbel/samples-filter

Issues

• Collect 500k repositories dataset

  #125 opened 5 months ago by h1alexbel
  14

• Split collected dataset into train/evaluation

  #128 opened 5 months ago by h1alexbel
  1

• Analyze 100 random ratings from test set

  #130 opened 5 months ago by h1alexbel
  1

• what makes SR so specific among other repositories?

  #157 opened 5 months ago by h1alexbel
  0

• Dockerize models training

  #143 opened 6 months ago by h1alexbel
  2

• embeddings.py:42-46: We generate embeddings for each...

  #153 opened 6 months ago by 0pdd
  0

• Investigate models based on unsupervised learning

  #129 opened 6 months ago by h1alexbel
  5

• t_bert.py:24-27: Train BERT transformer model. We need to...

  #154 opened 6 months ago by 0pdd
  0

• Update models README

  #145 opened 6 months ago by h1alexbel
  0

• full pipeline from repository to vector

  #149 opened 6 months ago by h1alexbel
  0

• preprocessing objects

  #146 opened 6 months ago by h1alexbel
  0

• Generate embeddings

  #142 opened 6 months ago by h1alexbel
  1

• Example of repository as encoded vector

  #140 opened 6 months ago by h1alexbel
  0

• Put total number of `commits` for each repo in `model/data/train.csv` and `pipeline/input.csv`

  #114 opened 6 months ago by h1alexbel
  1

• transformer.py:36-40: Compose train.csv data as input...

  #104 opened 6 months ago by 0pdd
  3

• Investigate learning algorithm for real/sample repo classification

  #75 opened 7 months ago by h1alexbel
  11

• compose.py:32-39: Split README.md into important...

  #119 opened 6 months ago by 0pdd
  1

• matches_keywords.py:47-50: Remove code duplication in...

  #82 opened 6 months ago by 0pdd
  1

• last_commit.py:25-29: Find a way to fetch date of last...

  #81 opened 6 months ago by 0pdd
  1

• integrate `ghminer` into dataset build pipeline

  #137 opened 6 months ago by h1alexbel
  2

• Makefile:32-35: Train isolation forest model. We need to...

  #139 opened 6 months ago by 0pdd
  0

• Unsupervised

  #126 opened 6 months ago by h1alexbel
  0

• Measure `F-score` and other metrics for predicted classifications

  #74 opened 6 months ago by h1alexbel
  0

• train.py:44-49: Train model on target columns instead...

  #52 opened 7 months ago by 0pdd
  2

• feed.py:33-36: Feed `readme`, `last_commit`,...

  #113 opened 7 months ago by 0pdd
  0

• Train classification model on hugging face

  #30 opened 7 months ago by h1alexbel
  7

• train.py:31-34: Prepare labels and encodings for model...

  #34 opened 7 months ago by 0pdd
  3

• dataset.py:51-57: CSV column 'readme' is too big to read...

  #39 opened 6 months ago by 0pdd
  2

• feed.py:33-37: Process all fields required as inputs. We...

  #109 opened 7 months ago by 0pdd
  1

• Integrate multiple models in CLI

  #105 opened 7 months ago by h1alexbel
  6

• cli.py:58-64: Find effective way for processing readme....

  #62 opened 7 months ago by 0pdd
  1

• Makefile:45-48: Bump code coverage to 55. After massive...

  #107 opened 7 months ago by 0pdd
  0

• Preprocess readme in 4 steps

  #71 opened 7 months ago by h1alexbel
  1

• predict.py:25-28: Resolve issue with all-time negative...

  #45 opened 7 months ago by 0pdd
  4

• random-forest.py:80-83: Upload model and vectorizer as...

  #90 opened 7 months ago by 0pdd
  2

• random-forest.py:35-37: Read CSV file from HuggingFace....

  #89 opened 7 months ago by 0pdd
  2

• Makefile:46-51: Commands `trainrf` and `transformer`...

  #88 opened 7 months ago by 0pdd
  0

• Pick random pack of repos from all collected

  #73 opened 7 months ago by h1alexbel
  0

• Pipeline for dataset building

  #41 opened 7 months ago by h1alexbel
  2

• combine.py:34-37: Combine all the csv files into...

  #44 opened 7 months ago by 0pdd
  0

• Document filtering method

  #66 opened 7 months ago by h1alexbel
  0

• Setup codecov.io

  #58 opened 7 months ago by h1alexbel
  0

• filter_pipe.py:39-42: Create integration test case for...

  #63 opened 7 months ago by 0pdd
  0

• predict.py:33-37: Fetch pretrained model saved in...

  #46 opened 7 months ago by 0pdd
  0

• predict.py:45-48: Give the ability to provide input_text...

  #43 opened 7 months ago by 0pdd
  1

• Include trained model files (tokenizer, weights and so) in cli package

  #31 opened 7 months ago by h1alexbel
  1

• load.py:29-32: Load dataset file from huggingface dataset...

  #38 opened 7 months ago by 0pdd
  1

• train.py:35-38: Document model training in...

  #35 opened 7 months ago by 0pdd
  0

• readme.py:35-38: Use raw.githubusercontent.com instead of...

  #24 opened 7 months ago by 0pdd
  0

• Escape readme in csv

  #25 opened 7 months ago by h1alexbel
  0