Delete README.md

README.md

@@ -1,74 +0,0 @@
-# Parallel Corpus: Romansh - German
-
-This repository contains the code and methodology used to create the Romansh–German parallel corpus published on [Hugging Face](https://huggingface.co/datasets/Sudehsna/parallel_dataset/settings). The project was developed as part of a programming course at the University of Zurich.
-
-## Description
-
-This project performs document-level alignment between Romansh and German web texts, which were extracted from the [Fineweb2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-2) dataset. It uses [OpenAI](https://platform.openai.com/docs/models/text-embedding-3-small) embeddings and cosine similarity to identify potential parallel texts.
-The full dataset is available on [Hugging Face](https://huggingface.co/datasets/Sudehsna/parallel_dataset/settings).
-
-## Dataset
-
-This project uses the Romansh and German partitions of the  [Fineweb2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-2) dataset. Both the original and the removed  versions of the dataset were used to improve alignment coverage.
-
-Since the German dataset contained significantly more entries than the Romansh dataset, it was filtered by retaining only those entries whose domains matched those found in the Romansh dataset.
-
-## Method
-
-1. **Preprocessing**:
-    - German dataset filtered to only contain domains also appearing in Romansh dataset
-2. **Manual Evaluation**:
-   -   A sample of 50 documents was manually checked to estimate alignability (see `analysis/manual_alignment`)
-3. **Embedding**:
-   - Used [OpenAI](https://platform.openai.com/docs/models/text-embedding-3-small) `text-embedding-3-small` (with truncation at 8192 tokens)
-4. **Similarity Calculation**:
-   - Cosine similarity computed between Romansh and German document embeddings
-   - With penalization for sentence length difference
-5. **Evaluation**:
-   - A Negative gold standard was used to ensure quality: a set of Romansh documents known to have no valid German alignment. 
-
-
-
-## Getting Started
-
-### Set up environment
-```bash
-python3 -m venv venv
-source venv/bin/activate
-pip install -r requirements.txt
-```
-
-### Run embedding
-```bash
-python3 scripts/get_embeddings.py
-```
->**NOTE**: This project uses OpenAI's embedding models. To run the embedding script, you need to provide your own OpenAI API key.
-
-The output is a `.csv` file with the following structure:
-
-```
-"text","id","metadata","embedding"
-```
-
-
-### Run similarity
-```bash
-python3 scripts/cosine_sim.py
-```
-
-The output is a `.jsonl` file containing entries in the following format:
-
-```json
-{
-  "romansh_text": " ... ",
-  "german_text": " ... ",
-  "similarity": float,
-  "original_similarity": float
-}
-```
-To determine a suitable _similarity threshold_ for aligning sentence pairs, we tested multiple values and observed that a threshold of **?** provided the best balance for this dataset (refer to `./scripts/threshold_comparisons.py`).
-
-
-## License
-
-?