Upload Hindi embeddings model and all associated files

.gitattributes

@@ -33,3 +33,4 @@ 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
+hindi_corpus.txt filter=lfs diff=lfs merge=lfs -text

embedding_model.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:343abe4c8b9a9e8f88fca1c5ab8bfc129c1ab2b3852c598e56ae7b3c3dedc1f1
-size 1491671832
+oid sha256:540a0bf4ad54049eb94815984cad54d3b569bab3a02317e1f0ee306d84f2e1d8
+size 497221224

hindi_corpus.txt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2e1c771f83f7eca2fc1270939e14c12fd939b431be50308f07e52698111ba80
+size 57984810

special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "unk_token": "<unk>",
+  "sep_token": "<sep>",
+  "pad_token": "<pad>",
+  "cls_token": "<cls>",
+  "mask_token": "<mask>"
+}

tokenizer.model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcea7b417395ecf731062d1d241db13069e60e23ccc7cad4a77535900c38fe37
-size 1302086
+oid sha256:123070be6221922fc42f22e54064f9cb2261715e02dd586a464231064d327d34
+size 1290162

tokenizer_config.json

@@ -1,11 +1,33 @@
 {
-  "model_type": "hindi_embedding",
-  "tokenizer_class": "HindiSentencePieceTokenizer",
+  "name": "hindi-tokenizer",
+  "version": "1.0.0",
+  "model_type": "sentencepiece",
+  "sp_model_type": "unigram",
+  "tokenizer_class": "SentencePieceTokenizer",
   "bos_token": "<s>",
   "eos_token": "</s>",
   "pad_token": "<pad>",
   "unk_token": "<unk>",
   "mask_token": "<mask>",
+  "cls_token": "<cls>",
+  "sep_token": "<sep>",
   "model_max_length": 512,
-  "vocab_size": 50000
+  "vocab_size": 50000,
+  "do_lower_case": false,
+  "special_tokens_map": {
+    "pad_token": "<pad>",
+    "unk_token": "<unk>",
+    "bos_token": "<s>",
+    "eos_token": "</s>",
+    "mask_token": "<mask>",
+    "sep_token": "<sep>",
+    "cls_token": "<cls>"
+  },
+  "tokenizer_file": "tokenizer.model",
+  "auto_map": {
+    "AutoTokenizer": [
+      "PreTrainedTokenizerFast",
+      null
+    ]
+  }
 }

training_config.json

@@ -0,0 +1,12 @@
+{
+  "output_dir": "hindi_tokenizer_sota",
+  "vocab_size": 50000,
+  "model_type": "unigram",
+  "character_coverage": 0.9995,
+  "num_iterations": 2000000,
+  "max_examples": 300000,
+  "sample_factor": 1.0,
+  "create_hf_config": true,
+  "debug": false,
+  "seed": 42
+}