Upload all models and assets for ady (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 +1,561 @@
+---
+language: ady
+language_name: ADY
+language_family: caucasian_northwest
+tags:
+  - wikilangs
+  - nlp
+  - tokenizer
+  - embeddings
+  - n-gram
+  - markov
+  - wikipedia
+  - monolingual
+  - family-caucasian_northwest
+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: 4.453
+  - name: best_isotropy
+    type: isotropy
+    value: 0.6831
+  - name: vocabulary_size
+    type: vocab
+    value: 8988
+generated: 2025-12-27
+---
+
+# ADY - Wikilangs Models
+## Comprehensive Research Report & Full Ablation Study
+
+This repository contains NLP models trained and evaluated by Wikilangs, specifically on **ADY** 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** | 3.223x | 3.18 | 0.1016% | 189,909 |
+| **16k** | 3.621x | 3.57 | 0.1142% | 169,055 |
+| **32k** | 4.071x | 4.02 | 0.1284% | 150,370 |
+| **64k** | 4.453x 🏆 | 4.39 | 0.1404% | 137,476 |
+
+### Tokenization Examples
+
+Below are sample sentences tokenized with each vocabulary size:
+
+**Sample 1:** `Киото — Японием и къалэ.
+
+Category:Къалэхэр
+Category:Японие`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁к и от о ▁— ▁японием ▁и ▁къалэ . ▁category ... (+5 more)` | 15 |
+| 16k | `▁ки ото ▁— ▁японием ▁и ▁къалэ . ▁category : къалэхэр ... (+3 more)` | 13 |
+| 32k | `▁ки ото ▁— ▁японием ▁и ▁къалэ . ▁category : къалэхэр ... (+3 more)` | 13 |
+| 64k | `▁киото ▁— ▁японием ▁и ▁къалэ . ▁category : къалэхэр ▁category ... (+2 more)` | 12 |
+
+**Sample 2:** `Ереван () – Армение и къэлэшъхьаI. Нэбгырэ млн 1,06 фэдиз дэс. Къалэм и лIышъхьэ...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁е ре ван ▁() ▁– ▁армение ▁и ▁къэлэшъхьа i . ... (+28 more)` | 38 |
+| 16k | `▁ереван ▁() ▁– ▁армение ▁и ▁къэлэшъхьа i . ▁нэбгырэ ▁млн ... (+25 more)` | 35 |
+| 32k | `▁ереван ▁() ▁– ▁армение ▁и ▁къэлэшъхьа i . ▁нэбгырэ ▁млн ... (+25 more)` | 35 |
+| 64k | `▁ереван ▁() ▁– ▁армение ▁и ▁къэлэшъхьа i . ▁нэбгырэ ▁млн ... (+20 more)` | 30 |
+
+**Sample 3:** `thumb
+thumb
+Ишъхъэрэ Америкэ — континент.
+
+ЧIырэу млн 24,7 км² фэдиз еубыты. ЦIы...`
+
+| Vocab | Tokens | Count |
+|-------|--------|-------|
+| 8k | `▁thumb ▁thumb ▁ишъхъэрэ ▁америкэ ▁— ▁континент . ▁ч i ырэу ... (+27 more)` | 37 |
+| 16k | `▁thumb ▁thumb ▁ишъхъэрэ ▁америкэ ▁— ▁континент . ▁ч i ырэу ... (+27 more)` | 37 |
+| 32k | `▁thumb ▁thumb ▁ишъхъэрэ ▁америкэ ▁— ▁континент . ▁ч i ырэу ... (+27 more)` | 37 |
+| 64k | `▁thumb ▁thumb ▁ишъхъэрэ ▁америкэ ▁— ▁континент . ▁ч i ырэу ... (+27 more)` | 37 |
+
+
+### Key Findings
+
+- **Best Compression:** 64k achieves 4.453x compression
+- **Lowest UNK Rate:** 8k with 0.1016% 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** | 927 🏆 | 9.86 | 1,856 | 38.3% | 83.1% |
+| **2-gram** | 486 🏆 | 8.92 | 2,656 | 53.5% | 95.5% |
+| **3-gram** | 1,521 | 10.57 | 2,744 | 28.3% | 71.0% |
+| **3-gram** | 3,351 | 11.71 | 15,024 | 23.1% | 61.6% |
+| **4-gram** | 4,981 | 12.28 | 7,604 | 14.3% | 42.5% |
+| **4-gram** | 12,700 | 13.63 | 44,900 | 11.7% | 37.6% |
+
+### Top 5 N-grams by Size
+
+**2-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `category :` | 662 |
+| 2 | `- рэ` | 638 |
+| 3 | `- м` | 464 |
+| 4 | `рэ илъэсым` | 335 |
+| 5 | `. category` | 276 |
+
+**3-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `- рэ илъэсым` | 333 |
+| 2 | `. category :` | 276 |
+| 3 | `category : !` | 179 |
+| 4 | `: ! main` | 179 |
+| 5 | `! main category` | 179 |
+
+**4-grams:**
+
+| Rank | N-gram | Count |
+|------|--------|-------|
+| 1 | `category : ! main` | 179 |
+| 2 | `: ! main category` | 179 |
+| 3 | `. category : !` | 132 |
+| 4 | `. хэгэгум чiырэу иiэр` | 101 |
+| 5 | `. дло - м` | 87 |
+
+
+### Key Findings
+
+- **Best Perplexity:** 2-gram with 486
+- **Entropy Trend:** Decreases with larger n-grams (more predictable)
+- **Coverage:** Top-1000 patterns cover ~38% 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** | 0.3643 | 1.287 | 2.28 | 28,827 | 63.6% |
+| **1** | 1.5343 | 2.896 | 12.27 | 463 | 0.0% |
+| **2** | 0.1248 | 1.090 | 1.24 | 65,637 | 87.5% |
+| **2** | 1.1477 | 2.216 | 5.66 | 5,679 | 0.0% |
+| **3** | 0.0452 | 1.032 | 1.07 | 80,882 | 95.5% |
+| **3** | 0.7357 | 1.665 | 2.89 | 32,122 | 26.4% |
+| **4** | 0.0244 🏆 | 1.017 | 1.04 | 86,492 | 97.6% |
+| **4** | 0.4145 🏆 | 1.333 | 1.83 | 92,841 | 58.5% |
+
+### Generated Text Samples
+
+Below are text samples generated from each Markov chain model:
+
+**Context Size 1:**
+
+1. `. ахъщэр iэгурыхьэ - 14 . ы ↔ ӏей ; пшэс 1угжъу category : / даутэ`
+2. `, коц , " ids " " тэiошъ , батэ вгъэшым сэ силъэпкъ шъукъыхахьэу , къэрабгъэр`
+3. `- рэ лӏэшӏэгъуршапсыгъэхэршапсыгъэ ныпгорэ . географие гъунэгъухэр : " kaynar , каракас ) къо зиiэм ...`
+
+**Context Size 2:**
+
+1. `category : район category : ! main category зэпыщэхэр`
+2. `- рэ щэпсэу . гулъытэгъуэ техьэпӏэхэр къайсэр адыгэ хасэм ( дах - м хахьэ . хэгъэгу лiышъхьэр`
+3. `- м инароднэ тхакiу . илэжьэнхэр 1960 - рэ ислъэсхэм – адыгэ къэралыгъо университетым студентхэр щег...`
+
+**Context Size 3:**
+
+1. `- рэ илъэсым къыщегъэжьагъэу щэiэфэ гуманитар ушэтынхэмкiэ адыгэ республикэ институтым литературэмкi...`
+2. `. category : ! main category зэпыщэхэр`
+3. `category : ! main category зэпыщэхэр`
+
+**Context Size 4:**
+
+1. `category : ! main category зэпыщэхэр`
+2. `. category : ! main category зэпыщэхэр`
+3. `. хэгэгум чiырэу иiэр 9 984 670 км² ( дунаемкiэ я - 11 ) . хэгэгум чiырэу иiэр 283`
+
+
+### Key Findings
+
+- **Best Predictability:** Context-4 with 97.6% predictability
+- **Branching Factor:** Decreases with context size (more deterministic)
+- **Memory Trade-off:** Larger contexts require more storage (92,841 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 | 8,988 |
+| Total Tokens | 57,159 |
+| Mean Frequency | 6.36 |
+| Median Frequency | 3 |
+| Frequency Std Dev | 23.47 |
+
+### Most Common Words
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | и | 1,019 |
+| 2 | category | 841 |
+| 3 | адыгэ | 701 |
+| 4 | рэ | 641 |
+| 5 | м | 541 |
+| 6 | илъэсым | 407 |
+| 7 | ащ | 392 |
+| 8 | я | 349 |
+| 9 | ары | 276 |
+| 10 | а | 259 |
+
+### Least Common Words (from vocabulary)
+
+| Rank | Word | Frequency |
+|------|------|-----------|
+| 1 | britishpedia | 2 |
+| 2 | encyklopedia | 2 |
+| 3 | osobistości | 2 |
+| 4 | rzeczypospolitej | 2 |
+| 5 | polskiej | 2 |
+| 6 | bph | 2 |
+| 7 | british | 2 |
+| 8 | publishing | 2 |
+| 9 | ltd | 2 |
+| 10 | 912100 | 2 |
+
+### Zipf's Law Analysis
+
+| Metric | Value |
+|--------|-------|
+| Zipf Coefficient | 0.7855 |
+| R² (Goodness of Fit) | 0.976491 |
+| Adherence Quality | **excellent** |
+
+### Coverage Analysis
+
+| Top N Words | Coverage |
+|-------------|----------|
+| Top 100 | 26.7% |
+| Top 1,000 | 56.9% |
+| Top 5,000 | 86.0% |
+| Top 10,000 | 0.0% |
+
+### Key Findings
+
+- **Zipf Compliance:** R²=0.9765 indicates excellent adherence to Zipf's law
+- **High Frequency Dominance:** Top 100 words cover 26.7% of corpus
+- **Long Tail:** -1,012 words needed for remaining 100.0% 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** | 1,830 | 32 | 3.764 | 0.663 | 0.6831 🏆 |
+| **mono_64d** | 1,830 | 64 | 3.806 | 0.668 | 0.2517 |
+| **mono_128d** | 1,830 | 128 | 3.824 | 0.669 | 0.0484 |
+| **embeddings_enhanced** | 0 | 0 | 0.000 | 0.000 | 0.0000 |
+
+### Key Findings
+
+- **Best Isotropy:** mono_32d with 0.6831 (more uniform distribution)
+- **Dimension Trade-off:** Higher dimensions capture more semantics but reduce isotropy
+- **Vocabulary Coverage:** All models cover 1,830 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 (4.45x) with low UNK rate |
+| N-gram | **5-gram** | Lowest perplexity (486) |
+| Markov | **Context-4** | Highest predictability (97.6%) |
+| 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 04:34:00*

models/embeddings/monolingual/ady_128d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27f96cb22cce1dccf3f5a68afba58270c9a3d30c33c99e0b985a0160dc08be41
+size 1025913894

models/embeddings/monolingual/ady_128d.meta.json

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

models/embeddings/monolingual/ady_128d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ady",
+  "dimension": 128,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 128,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 1830
+}

models/embeddings/monolingual/ady_32d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4df91a4ae1b68f41d1ebc698b3b79863f82558368b99039be6879f7219c9dfab
+size 256508454

models/embeddings/monolingual/ady_32d.meta.json

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

models/embeddings/monolingual/ady_32d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ady",
+  "dimension": 32,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 32,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 1830
+}

models/embeddings/monolingual/ady_64d.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e39765ffd2c8eb6ef0e3b1854af129b7b0260b2e70a54156a81a9a099b1fa27b
+size 512976934

models/embeddings/monolingual/ady_64d.meta.json

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

models/embeddings/monolingual/ady_64d_metadata.json

@@ -0,0 +1,13 @@
+{
+  "language": "ady",
+  "dimension": 64,
+  "version": "monolingual",
+  "training_params": {
+    "dim": 64,
+    "min_count": 5,
+    "window": 5,
+    "negative": 5,
+    "epochs": 5
+  },
+  "vocab_size": 1830
+}

models/subword_markov/ady_markov_ctx1_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb658a5696f0194212546a4afabf40b1859c08800bbd6f1ffecda37fddfd3c58
+size 45335

models/subword_markov/ady_markov_ctx1_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "subword",
+  "language": "ady",
+  "unique_contexts": 463,
+  "total_transitions": 611427
+}

models/subword_markov/ady_markov_ctx2_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5ebd13fcb32a70f1eeeb3c7ec3df3945466bdb98d3539405ae08a602378e310
+size 217004

models/subword_markov/ady_markov_ctx2_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "subword",
+  "language": "ady",
+  "unique_contexts": 5679,
+  "total_transitions": 610662
+}

models/subword_markov/ady_markov_ctx3_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9df0d86e436385c2b2494f0af89b0ca67a6a9d8aa7e5706dadd31de6649a3d8
+size 714083

models/subword_markov/ady_markov_ctx3_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "subword",
+  "language": "ady",
+  "unique_contexts": 32122,
+  "total_transitions": 609897
+}

models/subword_markov/ady_markov_ctx4_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14192c7fe2413ebd6957d9d1e1bf08f20d2cf28008cf6b161e3c9ff26a079c9b
+size 1599436

models/subword_markov/ady_markov_ctx4_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "subword",
+  "language": "ady",
+  "unique_contexts": 92841,
+  "total_transitions": 609132
+}

models/subword_ngram/ady_2gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:016367258813db8565f69ab7cb29945baccaac7e94321afc25a4b6c3c9f79f58
+size 32965

models/subword_ngram/ady_2gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "subword",
+  "language": "ady",
+  "unique_ngrams": 2656,
+  "total_ngrams": 611427
+}

models/subword_ngram/ady_3gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19fa61eac0061901eb9506a6a556a9261919e5b319eaf1999aa663587ce22014
+size 187929

models/subword_ngram/ady_3gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "subword",
+  "language": "ady",
+  "unique_ngrams": 15024,
+  "total_ngrams": 610662
+}

models/subword_ngram/ady_4gram_subword.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c90862733e2c9b9b0d0d50264ea47e26c1bdce42f31aa0ab349bff62b10cc5fe
+size 580214

models/subword_ngram/ady_4gram_subword_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "subword",
+  "language": "ady",
+  "unique_ngrams": 44900,
+  "total_ngrams": 609897
+}

models/tokenizer/ady_tokenizer_16k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6120c55e21bf381032672a3067bcf46760e2c4d359e940e77b9f3f746421b69
+size 564362

models/tokenizer/ady_tokenizer_32k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e432f1beef1ea7d335d31918e5e88f5751d4fc577f2231b239c92518f39dc9c
+size 955536

models/tokenizer/ady_tokenizer_64k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94c5357e44e30fbb0a13d6b3824a83f80c45d7160db13b16aa0d683fee765fe9
+size 1660676

models/tokenizer/ady_tokenizer_8k.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc7334d75b921cca2c9e616b995ac20d0728c4f9b3c54f3c5cb0b2ba2b5267c
+size 398334

models/vocabulary/ady_vocabulary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:733b017356844758ce21db6ab8269f1bff9141c2d41430f547302391bc541d7e
+size 161889

models/vocabulary/ady_vocabulary_metadata.json

@@ -0,0 +1,16 @@
+{
+  "language": "ady",
+  "vocabulary_size": 8988,
+  "statistics": {
+    "type_token_ratio": 0.374012371348373,
+    "coverage": {
+      "top_100": 0.1982014762449319,
+      "top_1000": 0.42253612641646743,
+      "top_5000": 0.6391386838548706,
+      "top_10000": 0.7559387670235991
+    },
+    "hapax_count": 19793,
+    "hapax_ratio": 0.6877106424377193,
+    "total_documents": 765
+  }
+}

models/word_markov/ady_markov_ctx1_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0faaa6cb0c52d77ac3f86b442e03b6cdd0646165c5b5467afaa42799100c5c08
+size 1013255

models/word_markov/ady_markov_ctx1_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 1,
+  "variant": "word",
+  "language": "ady",
+  "unique_contexts": 28827,
+  "total_transitions": 105741
+}

models/word_markov/ady_markov_ctx2_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdf814da92430ed71f7261a45abbc27d6bc4f679afc5556ff21d158586b5709f
+size 1738893

models/word_markov/ady_markov_ctx2_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 2,
+  "variant": "word",
+  "language": "ady",
+  "unique_contexts": 65637,
+  "total_transitions": 104976
+}

models/word_markov/ady_markov_ctx3_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c430f624230776fb0bed9ceec3edfcc58d221b9f682aa377294e7faf952a7f33
+size 2203636

models/word_markov/ady_markov_ctx3_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 3,
+  "variant": "word",
+  "language": "ady",
+  "unique_contexts": 80882,
+  "total_transitions": 104212
+}

models/word_markov/ady_markov_ctx4_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f994f94aaa5faf04bea4da903f474a186f4d719203b671b4c8538fa7e3d6c323
+size 2577702

models/word_markov/ady_markov_ctx4_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "context_size": 4,
+  "variant": "word",
+  "language": "ady",
+  "unique_contexts": 86492,
+  "total_transitions": 103449
+}

models/word_ngram/ady_2gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e36a3e7022b130b28b923c0f0edfaff7cb6f8c269b793337bd6e999b68fc36cd
+size 35981

models/word_ngram/ady_2gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 2,
+  "variant": "word",
+  "language": "ady",
+  "unique_ngrams": 1856,
+  "total_ngrams": 105741
+}

models/word_ngram/ady_3gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2ce08e36db31a04439c365d753c2684b44924398908b6f7a2883305f54f1f0c
+size 61177

models/word_ngram/ady_3gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 3,
+  "variant": "word",
+  "language": "ady",
+  "unique_ngrams": 2744,
+  "total_ngrams": 104976
+}

models/word_ngram/ady_4gram_word.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43176b575286a7924ea5d8764dc0f6dfc283f757937cfa2d648424a65326ca9f
+size 203246

models/word_ngram/ady_4gram_word_metadata.json

@@ -0,0 +1,7 @@
+{
+  "n": 4,
+  "variant": "word",
+  "language": "ady",
+  "unique_ngrams": 7604,
+  "total_ngrams": 104212
+}