Datasets:
Upload models/commonlid_20pct/training_summary.json with huggingface_hub
Browse files
models/commonlid_20pct/training_summary.json
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"command": "train.py --corpus-dir data/cl20_group_4 --initial-vocab /workspace/hf_home/hub/models--CohereLabs--tiny-aya-global/snapshots/b2f40192d990d66a6f28a4b90d72817ab9c613da/tokenizer.json --per-lang-counts-method soft --em-iterations 10 --results-dir results_commonlid_20pct --lang-batch-size 10 --reuse-base",
|
| 3 |
+
"timestamp": "2026-03-24T06:25:41.252835+00:00",
|
| 4 |
+
"training_completed": true,
|
| 5 |
+
"source": {
|
| 6 |
+
"format": "corpus",
|
| 7 |
+
"path": "/workspace/runpod/tinyAyaLid/data/cl20_group_4",
|
| 8 |
+
"max_samples": null,
|
| 9 |
+
"total_samples": 8014,
|
| 10 |
+
"num_languages": 7,
|
| 11 |
+
"samples_per_language": {
|
| 12 |
+
"amh": 323,
|
| 13 |
+
"ces": 186,
|
| 14 |
+
"gaz": 214,
|
| 15 |
+
"ind": 6765,
|
| 16 |
+
"mar": 212,
|
| 17 |
+
"sot": 188,
|
| 18 |
+
"wuu": 126
|
| 19 |
+
}
|
| 20 |
+
},
|
| 21 |
+
"method": {
|
| 22 |
+
"vocab_size": 261000,
|
| 23 |
+
"base_training_method": "hf",
|
| 24 |
+
"per_lang_counts_method": "soft",
|
| 25 |
+
"byte_level": true,
|
| 26 |
+
"seed": 42,
|
| 27 |
+
"initial_vocab": "/workspace/hf_home/hub/models--CohereLabs--tiny-aya-global/snapshots/b2f40192d990d66a6f28a4b90d72817ab9c613da/tokenizer.json",
|
| 28 |
+
"lang_batch_size": 10,
|
| 29 |
+
"sampling": {
|
| 30 |
+
"mode": "base_only",
|
| 31 |
+
"max_base_samples_per_lang": 10000,
|
| 32 |
+
"max_lang_samples_per_lang": null,
|
| 33 |
+
"shared_samples_per_lang": null
|
| 34 |
+
},
|
| 35 |
+
"reuse": {
|
| 36 |
+
"reuse_corpus": true,
|
| 37 |
+
"reuse_base": true,
|
| 38 |
+
"skip_existing_langs": true
|
| 39 |
+
}
|
| 40 |
+
},
|
| 41 |
+
"timing": {
|
| 42 |
+
"total_seconds": 206.39,
|
| 43 |
+
"base_tokenizer_seconds": 0.0,
|
| 44 |
+
"language_tokenizers_seconds": 206.39,
|
| 45 |
+
"base_tokenizer_reused": true
|
| 46 |
+
},
|
| 47 |
+
"output": {
|
| 48 |
+
"results_dir": "/workspace/runpod/tinyAyaLid/results_commonlid_20pct",
|
| 49 |
+
"corpus_dir": "/workspace/runpod/tinyAyaLid/data/cl20_group_4",
|
| 50 |
+
"tokenizers_dir": "/workspace/runpod/tinyAyaLid/results_commonlid_20pct/tokenizers",
|
| 51 |
+
"base_tokenizer": "/workspace/runpod/tinyAyaLid/results_commonlid_20pct/tokenizers/langspec_base_tokenizer.json",
|
| 52 |
+
"num_languages_trained_this_run": 7,
|
| 53 |
+
"language_tokenizers": {
|
| 54 |
+
"base_path": "results_commonlid_20pct/tokenizers/langspec_base_tokenizer.json",
|
| 55 |
+
"language_paths": {
|
| 56 |
+
"amh": "results_commonlid_20pct/tokenizers/langspec_soft_amh.tokenizer.json",
|
| 57 |
+
"ces": "results_commonlid_20pct/tokenizers/langspec_soft_ces.tokenizer.json",
|
| 58 |
+
"gaz": "results_commonlid_20pct/tokenizers/langspec_soft_gaz.tokenizer.json",
|
| 59 |
+
"ind": "results_commonlid_20pct/tokenizers/langspec_soft_ind.tokenizer.json",
|
| 60 |
+
"mar": "results_commonlid_20pct/tokenizers/langspec_soft_mar.tokenizer.json",
|
| 61 |
+
"sot": "results_commonlid_20pct/tokenizers/langspec_soft_sot.tokenizer.json",
|
| 62 |
+
"wuu": "results_commonlid_20pct/tokenizers/langspec_soft_wuu.tokenizer.json"
|
| 63 |
+
}
|
| 64 |
+
}
|
| 65 |
+
},
|
| 66 |
+
"languages": [
|
| 67 |
+
"amh",
|
| 68 |
+
"ces",
|
| 69 |
+
"gaz",
|
| 70 |
+
"ind",
|
| 71 |
+
"mar",
|
| 72 |
+
"sot",
|
| 73 |
+
"wuu"
|
| 74 |
+
]
|
| 75 |
+
}
|