tokenizer unigram_bi_ar
Browse files
tokenizers/unigram_bi_ar/meta.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"flavor": "unigram",
|
| 3 |
+
"condition": "bi_ar",
|
| 4 |
+
"vocab_size": 65536,
|
| 5 |
+
"specials": [
|
| 6 |
+
"<unk>",
|
| 7 |
+
"<bos>",
|
| 8 |
+
"<eos>",
|
| 9 |
+
"<pad>"
|
| 10 |
+
],
|
| 11 |
+
"corpus_files": [
|
| 12 |
+
"/mnt/scratch/xscript/tok_corpora/bi_ar/ar.txt",
|
| 13 |
+
"/mnt/scratch/xscript/tok_corpora/bi_ar/en.txt"
|
| 14 |
+
]
|
| 15 |
+
}
|
tokenizers/unigram_bi_ar/sp.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d47e847ad652f7a85979ca0bcc3ca2fee14980093aac180b074ab32fe4ab2d5e
|
| 3 |
+
size 1503083
|