Upload all models and assets for ban (20251201)

.gitattributes

@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+visualizations/embedding_similarity.png filter=lfs diff=lfs merge=lfs -text
+visualizations/performance_dashboard.png filter=lfs diff=lfs merge=lfs -text
+visualizations/position_encoding_comparison.png filter=lfs diff=lfs merge=lfs -text
+visualizations/tsne_sentences.png filter=lfs diff=lfs merge=lfs -text
+visualizations/tsne_words.png filter=lfs diff=lfs merge=lfs -text
+visualizations/zipf_law.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,584 @@
+---
+language: ban
+language_name: BAN
+language_family: austronesian_other
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-austronesian_other
+license: mit
+library_name: wikilangs
+pipeline_tag: feature-extraction
+datasets:
+  - omarkamali/wikipedia-monthly
+dataset_info:
+  name: wikipedia-monthly
+  description: Monthly snapshots of Wikipedia articles across 300+ languages
+metrics:
+  - name: best_compression_ratio
+    type: compression
+    value: 4.782
+  - name: best_isotropy
+    type: isotropy
+    value: 0.8612
+  - name: vocabulary_size
+    type: vocab
+    value: 109825
+generated: 2025-12-27
+---
+
+# BAN - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **BAN** Wikipedia data.
+We analyze tokenizers, n-gram models, Markov chains, vocabulary statistics, and word embeddings.
+
+## 📋 Repository Contents
+
+### Models & Assets
+
+- Tokenizers (8k, 16k, 32k, 64k)
+- N-gram models (2, 3, 4-gram)
+- Markov chains (context of 1, 2, 3 and 4)
+- Subword N-gram and Markov chains
+- Embeddings in various sizes and dimensions
+- Language Vocabulary
+- Language Statistics
+![Performance Dashboard](visualizations/performance_dashboard.png)
+
+### Analysis and Evaluation
+
+- [1. Tokenizer Evaluation](#1-tokenizer-evaluation)
+- [2. N-gram Model Evaluation](#2-n-gram-model-evaluation)
+- [3. Markov Chain Evaluation](#3-markov-chain-evaluation)
+- [4. Vocabulary Analysis](#4-vocabulary-analysis)
+- [5. Word Embeddings Evaluation](#5-word-embeddings-evaluation)
+- [6. Summary & Recommendations](#6-summary--recommendations)
+- [Metrics Glossary](#appendix-metrics-glossary--interpretation-guide)
+- [Visualizations Index](#visualizations-index)
+
+---
+## 1. Tokenizer Evaluation
+
+![Tokenizer Compression](visualizations/tokenizer_compression.png)
+
+### Results
+
+| Vocab Size | Compression | Avg Token Len | UNK Rate | Total Tokens |
+|------------|-------------|---------------|----------|--------------|
+| **8k** | 3.889x | 3.84 | 0.1469% | 269,485 |
+| **16k** | 4.255x | 4.21 | 0.1608% | 246,312 |
+| **32k** | 4.547x | 4.49 | 0.1718% | 230,479 |
+| **64k** | 4.782x 🏆 | 4.73 | 0.1807% | 219,125 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `1020
+
+1021
+
+1022
+
+1023
+
+1024
+
+1025
+
+1026
+
+1027
+
+1028
+
+1029
+
+Jadma
+
+Embas
+
+Seda
+
+...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁ 1 0 2 0 ▁ 1 0 2 1 ... (+53 more)` | 63 |
+| 16k | `▁ 1 0 2 0 ▁ 1 0 2 1 ... (+53 more)` | 63 |
+| 32k | `▁ 1 0 2 0 ▁ 1 0 2 1 ... (+53 more)` | 63 |
+| 64k | `▁ 1 0 2 0 ▁ 1 0 2 1 ... (+53 more)` | 63 |
+
+**Sample 2:** `Pustaka
+
+Pranala liyané
+
+Kategori:Abad ka-17`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁pustaka ▁pranala ▁liyané ▁kategori : abad ▁ka - 1 7` | 10 |
+| 16k | `▁pustaka ▁pranala ▁liyané ▁kategori : abad ▁ka - 1 7` | 10 |
+| 32k | `▁pustaka ▁pranala ▁liyané ▁kategori : abad ▁ka - 1 7` | 10 |
+| 64k | `▁pustaka ▁pranala ▁liyané ▁kategori : abad ▁ka - 1 7` | 10 |
+
+**Sample 3:** `Siung Sri Lanka (Gracula ptilogenys), inggih punika satunggil curik, anggota kul...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁si ung ▁sri ▁lan ka ▁( gr ac ula ▁p ... (+24 more)` | 34 |
+| 16k | `▁si ung ▁sri ▁lanka ▁( gr ac ula ▁p til ... (+21 more)` | 31 |
+| 32k | `▁siung ▁sri ▁lanka ▁( gr ac ula ▁p til ogen ... (+19 more)` | 29 |
+| 64k | `▁siung ▁sri ▁lanka ▁( gracula ▁ptil ogen ys ), ▁inggih ... (+15 more)` | 25 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.782x compression
+- **Lowest UNK Rate:** 8k with 0.1469% unknown tokens
+- **Trade-off:** Larger vocabularies improve compression but increase model size
+- **Recommendation:** 32k vocabulary provides optimal balance for production use
+
+---
+## 2. N-gram Model Evaluation
+
+![N-gram Perplexity](visualizations/ngram_perplexity.png)
+
+![N-gram Coverage](visualizations/ngram_coverage.png)
+
+### Results
+
+| N-gram | Perplexity | Entropy | Unique N-grams | Top-100 Coverage | Top-1000 Coverage |
+|--------|------------|---------|----------------|------------------|-------------------|
+| **2-gram** | 6,772 🏆 | 12.73 | 86,017 | 32.0% | 53.5% |
+| **2-gram** | 287 🏆 | 8.17 | 6,739 | 67.5% | 98.5% |
+| **3-gram** | 9,433 | 13.20 | 132,180 | 30.5% | 50.3% |
+| **3-gram** | 2,255 | 11.14 | 56,338 | 28.2% | 73.8% |
+| **4-gram** | 14,846 | 13.86 | 212,984 | 26.8% | 45.5% |
+| **4-gram** | 10,513 | 13.36 | 295,874 | 17.0% | 50.1% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `kategori :` | 56,343 |
+| 2 | `situs resmi` | 43,670 |
+| 3 | `inggih punika` | 39,156 |
+| 4 | `pusat statistik` | 24,773 |
+| 5 | `badan pusat` | 24,763 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `badan pusat statistik` | 24,761 |
+| 2 | `pustaka pranala jaba` | 21,699 |
+| 3 | `) inggih punika` | 21,548 |
+| 4 | `inggih punika silih` | 20,523 |
+| 5 | `punika silih tunggil` | 20,157 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `inggih punika silih tunggil` | 20,047 |
+| 2 | `pranala jaba situs resmi` | 19,038 |
+| 3 | `pustaka pranala jaba situs` | 18,670 |
+| 4 | `) inggih punika silih` | 18,246 |
+| 5 | `( aksara bali :` | 17,893 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 287
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~50% of corpus
+- **Recommendation:** 4-gram or 5-gram for best predictive performance
+
+---
+## 3. Markov Chain Evaluation
+
+![Markov Entropy](visualizations/markov_entropy.png)
+
+![Markov Branching](visualizations/markov_branching.png)
+
+### Results
+
+| Context | Avg Entropy | Perplexity | Branching Factor | Unique Contexts | Predictability |
+|---------|-------------|------------|------------------|-----------------|----------------|
+| **1** | 0.5448 | 1.459 | 4.25 | 354,734 | 45.5% |
+| **1** | 1.0973 | 2.140 | 6.84 | 3,094 | 0.0% |
+| **2** | 0.2539 | 1.192 | 1.68 | 1,504,653 | 74.6% |
+| **2** | 0.8467 | 1.798 | 5.47 | 21,162 | 15.3% |
+| **3** | 0.0992 | 1.071 | 1.20 | 2,520,521 | 90.1% |
+| **3** | 0.8884 | 1.851 | 4.45 | 115,750 | 11.2% |
+| **4** | 0.0443 🏆 | 1.031 | 1.08 | 3,008,321 | 95.6% |
+| **4** | 0.7379 🏆 | 1.668 | 3.16 | 515,141 | 26.2% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `, seperti kota binjai kategori : désa dinas sané mangkin madué 10 désa pakraman buléléng .`
+2. `. iklan di pulo kyushu . akéhnyané 1 . kategori : ᬓ ᭂ ᬕᬮ ᭄ ᬤ`
+3. `ring warsa 2019 , definisi definisi asli riantara 24 / ilang . there is defined hypnosis`
+
+**Context Size 2:**
+
+1. `kategori : kota kendari wali kota ngawit jabatan saking pinanggal 22 pébruari 1857 – 1 al -`
+2. `situs resmi pamréntahan kabupatén tuban cutetan : pranala dados kauahin / ilang . yening url nenten ...`
+3. `inggih punika silih tunggil sanganan sané nénten pastika sakéwanten sumber akéh saking cina , itsĕrl...`
+
+**Context Size 3:**
+
+1. `badan pusat statistik kepulauan bangka belitung badan pusat statistik kota surabaya cutetan : url da...`
+2. `pustaka pranala jaba taman pahlawan margarana , ring pamahbah nyané , kain sasirangan kapercaya pras...`
+3. `) inggih punika silih tunggil kecamatan ring kabupatén bungo , propinsi jambi , ring panegara indoné...`
+
+**Context Size 4:**
+
+1. `inggih punika silih tunggil kecamatan ring kabupatén gowa , propinsi sulawesi selatan tanjung batu ,...`
+2. `pranala jaba situs resmi pemerintah kota tangerang situs resmi bps kota tangerang cutetan : url dado...`
+3. `pustaka pranala jaba situs resmi pamréntahan nusa tenggara barat badan pusat statistik sumatra utara...`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 95.6% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (515,141 contexts)
+- **Recommendation:** Context-3 or Context-4 for text generation
+
+---
+## 4. Vocabulary Analysis
+
+![Zipf's Law](visualizations/zipf_law.png)
+
+![Top Words](visualizations/top20_words.png)
+
+![Coverage Curve](visualizations/vocab_coverage.png)
+
+### Statistics
+
+| Metric | Value |
+|--------|-------|
+| Vocabulary Size | 109,825 |
+| Total Tokens | 4,059,826 |
+| Mean Frequency | 36.97 |
+| Median Frequency | 3 |
+| Frequency Std Dev | 763.95 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | ring | 133,380 |
+| 2 | kabupatén | 67,955 |
+| 3 | kategori | 56,442 |
+| 4 | punika | 52,655 |
+| 5 | situs | 48,035 |
+| 6 | sané | 47,128 |
+| 7 | resmi | 44,824 |
+| 8 | kecamatan | 42,212 |
+| 9 | inggih | 39,593 |
+| 10 | saking | 39,394 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | padaido | 2 |
+| 2 | inswambesi | 2 |
+| 3 | asaryendi | 2 |
+| 4 | sopendo | 2 |
+| 5 | pomdori | 2 |
+| 6 | yawosi | 2 |
+| 7 | ᬧᬓᬓ | 2 |
+| 8 | potrekwastanngawit | 2 |
+| 9 | patonangi | 2 |
+| 10 | ᬢᬢᬓᬦ | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 1.1336 |
+| R² (Goodness of Fit) | 0.997567 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 43.3% |
+| Top 1,000 | 67.9% |
+| Top 5,000 | 82.1% |
+| Top 10,000 | 87.0% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9976 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 43.3% of corpus
+- **Long Tail:** 99,825 words needed for remaining 13.0% coverage
+
+---
+## 5. Word Embeddings Evaluation
+
+![Embedding Isotropy](visualizations/embedding_isotropy.png)
+
+![Similarity Matrix](visualizations/embedding_similarity.png)
+
+![t-SNE Words](visualizations/tsne_words.png)
+
+![t-SNE Sentences](visualizations/tsne_sentences.png)
+
+### Model Comparison
+
+| Model | Vocab Size | Dimension | Avg Norm | Std Norm | Isotropy |
+|-------|------------|-----------|----------|----------|----------|
+| **mono_32d** | 50,333 | 32 | 4.290 | 1.041 | 0.8612 🏆 |
+| **mono_64d** | 50,333 | 64 | 4.879 | 1.017 | 0.8485 |
+| **mono_128d** | 50,333 | 128 | 5.532 | 0.920 | 0.8053 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.8612 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 50,333 words
+- **Recommendation:** 100d for balanced semantic capture and efficiency
+
+---
+## 6. Summary & Recommendations
+
+![Performance Dashboard](visualizations/performance_dashboard.png)
+
+### Production Recommendations
+
+| Component | Recommended | Rationale |
+|-----------|-------------|-----------|
+| Tokenizer | **32k BPE** | Best compression (4.78x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (287) |
+| Markov | **Context-4** | Highest predictability (95.6%) |
+| Embeddings | **100d** | Balanced semantic capture and isotropy |
+
+---
+## Appendix: Metrics Glossary & Interpretation Guide
+
+This section provides definitions, intuitions, and guidance for interpreting the metrics used throughout this report.
+
+### Tokenizer Metrics
+
+**Compression Ratio**
+> *Definition:* The ratio of characters to tokens (chars/token). Measures how efficiently the tokenizer represents text.
+>
+> *Intuition:* Higher compression means fewer tokens needed to represent the same text, reducing sequence lengths for downstream models. A 3x compression means ~3 characters per token on average.
+>
+> *What to seek:* Higher is generally better for efficiency, but extremely high compression may indicate overly aggressive merging that loses morphological information.
+
+**Average Token Length (Fertility)**
+> *Definition:* Mean number of characters per token produced by the tokenizer.
+>
+> *Intuition:* Reflects the granularity of tokenization. Longer tokens capture more context but may struggle with rare words; shorter tokens are more flexible but increase sequence length.
+>
+> *What to seek:* Balance between 2-5 characters for most languages. Arabic/morphologically-rich languages may benefit from slightly longer tokens.
+
+**Unknown Token Rate (OOV Rate)**
+> *Definition:* Percentage of tokens that map to the unknown/UNK token, indicating words the tokenizer cannot represent.
+>
+> *Intuition:* Lower OOV means better vocabulary coverage. High OOV indicates the tokenizer encounters many unseen character sequences.
+>
+> *What to seek:* Below 1% is excellent; below 5% is acceptable. BPE tokenizers typically achieve very low OOV due to subword fallback.
+
+### N-gram Model Metrics
+
+**Perplexity**
+> *Definition:* Measures how "surprised" the model is by test data. Mathematically: 2^(cross-entropy). Lower values indicate better prediction.
+>
+> *Intuition:* If perplexity is 100, the model is as uncertain as if choosing uniformly among 100 options at each step. A perplexity of 10 means effectively choosing among 10 equally likely options.
+>
+> *What to seek:* Lower is better. Perplexity decreases with larger n-grams (more context). Values vary widely by language and corpus size.
+
+**Entropy**
+> *Definition:* Average information content (in bits) needed to encode the next token given the context. Related to perplexity: perplexity = 2^entropy.
+>
+> *Intuition:* High entropy means high uncertainty/randomness; low entropy means predictable patterns. Natural language typically has entropy between 1-4 bits per character.
+>
+> *What to seek:* Lower entropy indicates more predictable text patterns. Entropy should decrease as n-gram size increases.
+
+**Coverage (Top-K)**
+> *Definition:* Percentage of corpus occurrences explained by the top K most frequent n-grams.
+>
+> *Intuition:* High coverage with few patterns indicates repetitive/formulaic text; low coverage suggests diverse vocabulary usage.
+>
+> *What to seek:* Depends on use case. For language modeling, moderate coverage (40-60% with top-1000) is typical for natural text.
+
+### Markov Chain Metrics
+
+**Average Entropy**
+> *Definition:* Mean entropy across all contexts, measuring average uncertainty in next-word prediction.
+>
+> *Intuition:* Lower entropy means the model is more confident about what comes next. Context-1 has high entropy (many possible next words); Context-4 has low entropy (few likely continuations).
+>
+> *What to seek:* Decreasing entropy with larger context sizes. Very low entropy (<0.1) indicates highly deterministic transitions.
+
+**Branching Factor**
+> *Definition:* Average number of unique next tokens observed for each context.
+>
+> *Intuition:* High branching = many possible continuations (flexible but uncertain); low branching = few options (predictable but potentially repetitive).
+>
+> *What to seek:* Branching factor should decrease with context size. Values near 1.0 indicate nearly deterministic chains.
+
+**Predictability**
+> *Definition:* Derived metric: (1 - normalized_entropy) × 100%. Indicates how deterministic the model's predictions are.
+>
+> *Intuition:* 100% predictability means the next word is always certain; 0% means completely random. Real text falls between these extremes.
+>
+> *What to seek:* Higher predictability for text generation quality, but too high (>98%) may produce repetitive output.
+
+### Vocabulary & Zipf's Law Metrics
+
+**Zipf's Coefficient**
+> *Definition:* The slope of the log-log plot of word frequency vs. rank. Zipf's law predicts this should be approximately -1.
+>
+> *Intuition:* A coefficient near -1 indicates the corpus follows natural language patterns where a few words are very common and most words are rare.
+>
+> *What to seek:* Values between -0.8 and -1.2 indicate healthy natural language distribution. Deviations may suggest domain-specific or artificial text.
+
+**R² (Coefficient of Determination)**
+> *Definition:* Measures how well the linear fit explains the frequency-rank relationship. Ranges from 0 to 1.
+>
+> *Intuition:* R² near 1.0 means the data closely follows Zipf's law; lower values indicate deviation from expected word frequency patterns.
+>
+> *What to seek:* R² > 0.95 is excellent; > 0.99 indicates near-perfect Zipf adherence typical of large natural corpora.
+
+**Vocabulary Coverage**
+> *Definition:* Cumulative percentage of corpus tokens accounted for by the top N words.
+>
+> *Intuition:* Shows how concentrated word usage is. If top-100 words cover 50% of text, the corpus relies heavily on common words.
+>
+> *What to seek:* Top-100 covering 30-50% is typical. Higher coverage indicates more repetitive text; lower suggests richer vocabulary.
+
+### Word Embedding Metrics
+
+**Isotropy**
+> *Definition:* Measures how uniformly distributed vectors are in the embedding space. Computed as the ratio of minimum to maximum singular values.
+>
+> *Intuition:* High isotropy (near 1.0) means vectors spread evenly in all directions; low isotropy means vectors cluster in certain directions, reducing expressiveness.
+>
+> *What to seek:* Higher isotropy generally indicates better-quality embeddings. Values > 0.1 are reasonable; > 0.3 is good. Lower-dimensional embeddings tend to have higher isotropy.
+
+**Average Norm**
+> *Definition:* Mean magnitude (L2 norm) of word vectors in the embedding space.
+>
+> *Intuition:* Indicates the typical "length" of vectors. Consistent norms suggest stable training; high variance may indicate some words are undertrained.
+>
+> *What to seek:* Relatively consistent norms across models. The absolute value matters less than consistency (low std deviation).
+
+**Cosine Similarity**
+> *Definition:* Measures angular similarity between vectors, ranging from -1 (opposite) to 1 (identical direction).
+>
+> *Intuition:* Words with similar meanings should have high cosine similarity. This is the standard metric for semantic relatedness in embeddings.
+>
+> *What to seek:* Semantically related words should score > 0.5; unrelated words should be near 0. Synonyms often score > 0.7.
+
+**t-SNE Visualization**
+> *Definition:* t-Distributed Stochastic Neighbor Embedding - a dimensionality reduction technique that preserves local structure for visualization.
+>
+> *Intuition:* Clusters in t-SNE plots indicate groups of semantically related words. Spread indicates vocabulary diversity; tight clusters suggest semantic coherence.
+>
+> *What to seek:* Meaningful clusters (e.g., numbers together, verbs together). Avoid over-interpreting distances - t-SNE preserves local, not global, structure.
+
+### General Interpretation Guidelines
+
+1. **Compare within model families:** Metrics are most meaningful when comparing models of the same type (e.g., 8k vs 64k tokenizer).
+2. **Consider trade-offs:** Better performance on one metric often comes at the cost of another (e.g., compression vs. OOV rate).
+3. **Context matters:** Optimal values depend on downstream tasks. Text generation may prioritize different metrics than classification.
+4. **Corpus influence:** All metrics are influenced by corpus characteristics. Wikipedia text differs from social media or literature.
+5. **Language-specific patterns:** Morphologically rich languages (like Arabic) may show different optimal ranges than analytic languages.
+
+
+### Visualizations Index
+
+| Visualization | Description |
+|---------------|-------------|
+| Tokenizer Compression | Compression ratios by vocabulary size |
+| Tokenizer Fertility | Average token length by vocabulary |
+| Tokenizer OOV | Unknown token rates |
+| Tokenizer Total Tokens | Total tokens by vocabulary |
+| N-gram Perplexity | Perplexity by n-gram size |
+| N-gram Entropy | Entropy by n-gram size |
+| N-gram Coverage | Top pattern coverage |
+| N-gram Unique | Unique n-gram counts |
+| Markov Entropy | Entropy by context size |
+| Markov Branching | Branching factor by context |
+| Markov Contexts | Unique context counts |
+| Zipf's Law | Frequency-rank distribution with fit |
+| Vocab Frequency | Word frequency distribution |
+| Top 20 Words | Most frequent words |
+| Vocab Coverage | Cumulative coverage curve |
+| Embedding Isotropy | Vector space uniformity |
+| Embedding Norms | Vector magnitude distribution |
+| Embedding Similarity | Word similarity heatmap |
+| Nearest Neighbors | Similar words for key terms |
+| t-SNE Words | 2D word embedding visualization |
+| t-SNE Sentences | 2D sentence embedding visualization |
+| Position Encoding | Encoding method comparison |
+| Model Sizes | Storage requirements |
+| Performance Dashboard | Comprehensive performance overview |
+
+---
+## About This Project
+
+### Data Source
+
+Models trained on [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly) - a monthly snapshot of Wikipedia articles across 300+ languages.
+
+### Project
+
+A project by **[Wikilangs](https://wikilangs.org)** - Open-source NLP models for every Wikipedia language.
+
+### Maintainer
+
+[Omar Kamali](https://omarkamali.com) - [Omneity Labs](https://omneitylabs.com)
+
+### Citation
+
+If you use these models in your research, please cite:
+
+```bibtex
+@misc{wikilangs2025,
+  author = {Kamali, Omar},
+  title = {Wikilangs: Open NLP Models for Wikipedia Languages},
+  year = {2025},
+  publisher = {HuggingFace},
+  url = {https://huggingface.co/wikilangs}
+  institution = {Omneity Labs}
+}
+```
+
+### License
+
+MIT License - Free for academic and commercial use.
+
+### Links
+
+- 🌐 Website: [wikilangs.org](https://wikilangs.org)
+- 🤗 Models: [huggingface.co/wikilangs](https://huggingface.co/wikilangs)
+- 📊 Data: [wikipedia-monthly](https://huggingface.co/datasets/omarkamali/wikipedia-monthly)
+- 👤 Author: [Omar Kamali](https://huggingface.co/omarkamali)
+---
+*Generated by Wikilangs Models Pipeline*
+
+*Report Date: 2025-12-27 23:53:08*

models/embeddings/monolingual/ban_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea87c1477ff7362a05ee5aecc04514daeeb0751d9258d36096220826e2071195
+size 1076400925

models/embeddings/monolingual/ban_128d.meta.json

@@ -0,0 +1 @@
+{"lang": "ban", "dim": 128, "max_seq_len": 512, "is_aligned": false}

models/embeddings/monolingual/ban_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ban",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 50333
+}

models/embeddings/monolingual/ban_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b78c8beda9adf946cfb61ebe5d7345f1a46cfbada20e008894b25148ee2b7fb6
+size 269745181

models/embeddings/monolingual/ban_32d.meta.json

@@ -0,0 +1 @@
+{"lang": "ban", "dim": 32, "max_seq_len": 512, "is_aligned": false}

models/embeddings/monolingual/ban_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ban",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 50333
+}

models/embeddings/monolingual/ban_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac247ae04637df7c57b93803868d52ed47fb91b34dd674de9b86612a1856543d
+size 538630429

models/embeddings/monolingual/ban_64d.meta.json

@@ -0,0 +1 @@
+{"lang": "ban", "dim": 64, "max_seq_len": 512, "is_aligned": false}

models/embeddings/monolingual/ban_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ban",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 50333
+}

models/subword_markov/ban_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07d34933ba1ea350752ae85b877a8c59815d819177bf4c06cc1c33c52a23a4ab
+size 164464

models/subword_markov/ban_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "ban",
+  "unique_contexts": 3094,
+  "total_transitions": 31281701
+}

models/subword_markov/ban_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a094b1a8d92237ba3ee5d4bb02c77daf556d4e57c92818f8594ee12d0d9dba7
+size 930375

models/subword_markov/ban_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "ban",
+  "unique_contexts": 21162,
+  "total_transitions": 31247141
+}

models/subword_markov/ban_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c14ef4a97de8adc59b7a39da2f9d5cdcbdc15c1339b56d4fe8f54919ee7676cc
+size 3747847

models/subword_markov/ban_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "ban",
+  "unique_contexts": 115750,
+  "total_transitions": 31212581
+}

models/subword_markov/ban_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83b89243df0423d04c3f14b54a3854b87b3df8860b6528722a527ca63138bedf
+size 12242821

models/subword_markov/ban_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "ban",
+  "unique_contexts": 515141,
+  "total_transitions": 31178021
+}

models/subword_ngram/ban_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d331dea0ac3863010c7f44bc8af6db64163208b39797a90da3a3db2cb5068c2
+size 89432

models/subword_ngram/ban_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "ban",
+  "unique_ngrams": 6739,
+  "total_ngrams": 31281701
+}

models/subword_ngram/ban_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e87314f3eb227a47badf79f1a2964e00acef96da0cc33f99259523ddadbe0267
+size 689305

models/subword_ngram/ban_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "ban",
+  "unique_ngrams": 56338,
+  "total_ngrams": 31247141
+}

models/subword_ngram/ban_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fe02f497267ca92b94af3687493cf750310e35ab85b736752b26ff56e51a295
+size 3233720

models/subword_ngram/ban_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "ban",
+  "unique_ngrams": 295874,
+  "total_ngrams": 31212581
+}

models/tokenizer/ban_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bedf8563a1afe2051346993a2e31dc66fa70f8ddf79e172628e2963e0a99dd47
+size 503307

models/tokenizer/ban_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c710ee000231372d734483b9d28128e8359cb0283ee8585cfee7e121ff3e088c
+size 776699

models/tokenizer/ban_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7f6893ba9c5195f098265006c68b68d4f854f4f3aef663fccea594bd3ecddbb
+size 1338022

models/tokenizer/ban_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e665b6cb1dccbf5b7fa71b5c8e3e9ea156603fcf9f94aeeaa2d783a6631da368
+size 371066

models/vocabulary/ban_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33bac4c04f637c4a7c7b840996ea84729668359a0bba018a9c721c27ccd525bc
+size 1847029

models/vocabulary/ban_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "ban",
+  "vocabulary_size": 109825,
+  "statistics": {
+    "type_token_ratio": 0.08234307722115898,
+    "coverage": {
+      "top_100": 0.4083867316599922,
+      "top_1000": 0.6399472916582452,
+      "top_5000": 0.7739293501921968,
+      "top_10000": 0.8207767696718877
+    },
+    "hapax_count": 244616,
+    "hapax_ratio": 0.690145891699888,
+    "total_documents": 34560
+  }
+}

models/word_markov/ban_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:490f92d81b31f52d83fd0f2cd5b002bd546e2be9ec34f436bbd4b503863a38fd
+size 14573406

models/word_markov/ban_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "ban",
+  "unique_contexts": 354734,
+  "total_transitions": 5300849
+}

models/word_markov/ban_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7933f21fe3dc7737031daf59c6f0b75d5f0ec0760ea5fbe33e643b28f9377775
+size 30285330

models/word_markov/ban_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "ban",
+  "unique_contexts": 1504653,
+  "total_transitions": 5266289
+}

models/word_markov/ban_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7d248fbf95bb7c4c07d3ad6a40e7491d998158dd768f5df49fb1a1fb786ad4d
+size 44547695

models/word_markov/ban_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "ban",
+  "unique_contexts": 2520521,
+  "total_transitions": 5231730
+}

models/word_markov/ban_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af85b8acafe95f6c6598d6c3d9f95ab7ddbfe52fec36eb64d4e73ed213f6ba54
+size 53030226

models/word_markov/ban_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "ban",
+  "unique_contexts": 3008321,
+  "total_transitions": 5197173
+}

models/word_ngram/ban_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a7e05025863931724c16c54d53b71d59185644881045f57e68714b7edff8d95
+size 1171748

models/word_ngram/ban_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "ban",
+  "unique_ngrams": 86017,
+  "total_ngrams": 5300849
+}

models/word_ngram/ban_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6382c65c522ccb8d948baed426584d1da5bc0135ba2f1a160c62c3a280a15c16
+size 1910455

models/word_ngram/ban_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "ban",
+  "unique_ngrams": 132180,
+  "total_ngrams": 5266289
+}

models/word_ngram/ban_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4076c7de92dd17498060b87078339b4ccf91574d719a5e0d84560c92179b24b1
+size 3227946

models/word_ngram/ban_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "ban",
+  "unique_ngrams": 212984,
+  "total_ngrams": 5231730
+}