README.md

metadata

language:
  - en
license: mit
size_categories:
  - 10M<n<100M
task_categories:
  - text-classification
  - table-question-answering
  - fill-mask
  - sentence-similarity
pretty_name: Movies Data with Embeddings
tags:
  - movies
  - embeddings
  - sentiment
  - vectors
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
dataset_info:
  features:
    - name: rated
      dtype: string
    - name: writers
      sequence: string
    - name: runtime
      dtype: float64
    - name: num_mflix_comments
      dtype: int64
    - name: title
      dtype: string
    - name: cast
      sequence: string
    - name: plot
      dtype: string
    - name: directors
      sequence: string
    - name: type
      dtype: string
    - name: fullplot
      dtype: string
    - name: languages
      sequence: string
    - name: awards
      struct:
        - name: nominations
          dtype: int64
        - name: text
          dtype: string
        - name: wins
          dtype: int64
    - name: imdb
      struct:
        - name: id
          dtype: int64
        - name: rating
          dtype: float64
        - name: votes
          dtype: int64
    - name: plot_embedding
      sequence: float64
    - name: metacritic
      dtype: float64
    - name: countries
      sequence: string
    - name: genres
      sequence: string
    - name: poster
      dtype: string
    - name: __index_level_0__
      dtype: int64
  splits:
    - name: train
      num_bytes: 13739333
      num_examples: 1017
    - name: test
      num_bytes: 5863663
      num_examples: 434
  download_size: 19321684
  dataset_size: 19602996

This dataset was created from the HuggingFace dataset AIatMongoDB/embedded_movies

Why was it needed?

1. The original dataset is close to 25 GB, for learning and experiments it is an overkill
2. Data in the dataset needs to be cleaned up e.g., some features are Null that requires extra care
3. Some of the embeddings are missing

How to use?

• Use for sentiment analysis
• Text similarity (plot)
• Embeddings : ready to use with vector DB & search libraries

---

dataset_info: features: - name: rated dtype: string - name: writers sequence: string - name: runtime dtype: float64 - name: num_mflix_comments dtype: int64 - name: title dtype: string - name: cast sequence: string - name: plot dtype: string - name: directors sequence: string - name: type dtype: string - name: fullplot dtype: string - name: languages sequence: string - name: awards struct: - name: nominations dtype: int64 - name: text dtype: string - name: wins dtype: int64 - name: imdb struct: - name: id dtype: int64 - name: rating dtype: float64 - name: votes dtype: int64 - name: plot_embedding sequence: float64 - name: metacritic dtype: float64 - name: countries sequence: string - name: genres sequence: string - name: poster dtype: string - name: index_level_0 dtype: int64 splits: - name: train num_bytes: 13791171 num_examples: 1021 - name: test num_bytes: 5811892 num_examples: 430 download_size: 19323013 dataset_size: 19603063 configs: - config_name: default data_files: - split: train path: data/train-* - split: test path: data/test-* license: mit task_categories: - text-classification - question-answering - zero-shot-classification - sentence-similarity - fill-mask - text-to-speech language: - en tags: - movies - embeddings - sentiment analysis pretty_name: Movies data with plot-embeddings size_categories: - 10M<n<100M