upload e5

.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
+e5/e5-finetuned/tokenizer.json filter=lfs diff=lfs merge=lfs -text

e5/__huggingface_repos__.json

@@ -0,0 +1 @@
+{"repos": [{"repoId": "intfloat/multilingual-e5-base", "repoType": "model", "commitHash": "835193815a3936a24a0ee7dc9e3d48c1fbb19c55", "filePaths": ["sentencepiece.bpe.model", "model.safetensors", "tokenizer_config.json", "tokenizer.json", "config.json", "special_tokens_map.json"]}]}

e5/e5-finetuned/config.json

@@ -0,0 +1,186 @@
+{
+  "architectures": [
+    "XLMRobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "balo",
+    "1": "b\u00e0n",
+    "2": "b\u00e0n ph\u00edm",
+    "3": "b\u00e0n ph\u00edm r\u1eddi",
+    "4": "b\u00e0n \u0103n",
+    "5": "b\u00ecnh gi\u1eef nhi\u1ec7t",
+    "6": "b\u00fat",
+    "7": "b\u01a1m n\u01b0\u1edbc",
+    "8": "b\u1ebfp t\u1eeb",
+    "9": "b\u1ebfp \u0111i\u1ec7n",
+    "10": "camera",
+    "11": "camera h\u00e0nh tr\u00ecnh",
+    "12": "chu\u1ed9t",
+    "13": "chu\u1ed9t m\u00e1y t\u00ednh",
+    "14": "c\u1ed1c gi\u1eef nhi\u1ec7t",
+    "15": "ebook reader",
+    "16": "gh\u1ebf",
+    "17": "gh\u1ebf gaming",
+    "18": "gh\u1ebf sofa",
+    "19": "gi\u00e0y",
+    "20": "gi\u00e0y th\u1ec3 thao",
+    "21": "k\u1ec7 gi\u00e0y",
+    "22": "laptop",
+    "23": "loa",
+    "24": "loa desktop",
+    "25": "loa soundbar",
+    "26": "m\u00e1y gi\u1eb7t",
+    "27": "m\u00e1y gi\u1eb7t c\u1eeda tr\u01b0\u1edbc",
+    "28": "m\u00e1y h\u00fat b\u1ee5i",
+    "29": "m\u00e1y h\u00fat b\u1ee5i mini",
+    "30": "m\u00e1y l\u1ea1nh",
+    "31": "m\u00e1y n\u01b0\u1edbc n\u00f3ng",
+    "32": "m\u00e1y pha c\u00e0 ph\u00ea",
+    "33": "m\u00e1y t\u00ednh x\u00e1ch tay",
+    "34": "m\u00e1y xay",
+    "35": "m\u00e1y \u00e9p",
+    "36": "m\u00e1y \u1ea3nh",
+    "37": "m\u0169 b\u1ea3o hi\u1ec3m",
+    "38": "m\u1ef9 ph\u1ea9m",
+    "39": "nh\u1ea1c c\u1ee5",
+    "40": "n\u01b0\u1edbc hoa",
+    "41": "n\u1ed3i chi\u00ean",
+    "42": "n\u1ed3i c\u01a1m",
+    "43": "n\u1ed3i c\u01a1m nh\u1ecf",
+    "44": "n\u1ed3i c\u01a1m \u0111i\u1ec7n",
+    "45": "n\u1ed9i th\u1ea5t",
+    "46": "other",
+    "47": "ph\u1ee5 ki\u1ec7n \u0111i\u1ec7n tho\u1ea1i",
+    "48": "qu\u1ea1t",
+    "49": "qu\u1ea7n",
+    "50": "sd card",
+    "51": "smart tivi",
+    "52": "son m\u00f4i",
+    "53": "s\u00e1ch",
+    "54": "s\u1ea1c d\u1ef1 ph\u00f2ng",
+    "55": "tablet",
+    "56": "tai nghe",
+    "57": "tivi",
+    "58": "trang s\u1ee9c",
+    "59": "t\u00fai x\u00e1ch",
+    "60": "t\u1ee7 l\u1ea1nh",
+    "61": "t\u1ee7 nh\u1ef1a",
+    "62": "t\u1ee7 \u0111\u00f4ng",
+    "63": "vali",
+    "64": "v\u00e1y",
+    "65": "v\u00e1y \u0111\u1ea7m",
+    "66": "watch automatic",
+    "67": "xe h\u01a1i",
+    "68": "xe m\u00e1y",
+    "69": "xe \u0111\u1ea1p",
+    "70": "\u00e1o",
+    "71": "\u00e1o thun",
+    "72": "\u0111i\u1ec1u h\u00f2a",
+    "73": "\u0111i\u1ec7n tho\u1ea1i",
+    "74": "\u0111\u1ed3ng h\u1ed3",
+    "75": "\u0111\u1ed3ng h\u1ed3 th\u00f4ng minh",
+    "76": "\u1ee7ng nh\u1ef1a"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "balo": 0,
+    "b\u00e0n": 1,
+    "b\u00e0n ph\u00edm": 2,
+    "b\u00e0n ph\u00edm r\u1eddi": 3,
+    "b\u00e0n \u0103n": 4,
+    "b\u00ecnh gi\u1eef nhi\u1ec7t": 5,
+    "b\u00fat": 6,
+    "b\u01a1m n\u01b0\u1edbc": 7,
+    "b\u1ebfp t\u1eeb": 8,
+    "b\u1ebfp \u0111i\u1ec7n": 9,
+    "camera": 10,
+    "camera h\u00e0nh tr\u00ecnh": 11,
+    "chu\u1ed9t": 12,
+    "chu\u1ed9t m\u00e1y t\u00ednh": 13,
+    "c\u1ed1c gi\u1eef nhi\u1ec7t": 14,
+    "ebook reader": 15,
+    "gh\u1ebf": 16,
+    "gh\u1ebf gaming": 17,
+    "gh\u1ebf sofa": 18,
+    "gi\u00e0y": 19,
+    "gi\u00e0y th\u1ec3 thao": 20,
+    "k\u1ec7 gi\u00e0y": 21,
+    "laptop": 22,
+    "loa": 23,
+    "loa desktop": 24,
+    "loa soundbar": 25,
+    "m\u00e1y gi\u1eb7t": 26,
+    "m\u00e1y gi\u1eb7t c\u1eeda tr\u01b0\u1edbc": 27,
+    "m\u00e1y h\u00fat b\u1ee5i": 28,
+    "m\u00e1y h\u00fat b\u1ee5i mini": 29,
+    "m\u00e1y l\u1ea1nh": 30,
+    "m\u00e1y n\u01b0\u1edbc n\u00f3ng": 31,
+    "m\u00e1y pha c\u00e0 ph\u00ea": 32,
+    "m\u00e1y t\u00ednh x\u00e1ch tay": 33,
+    "m\u00e1y xay": 34,
+    "m\u00e1y \u00e9p": 35,
+    "m\u00e1y \u1ea3nh": 36,
+    "m\u0169 b\u1ea3o hi\u1ec3m": 37,
+    "m\u1ef9 ph\u1ea9m": 38,
+    "nh\u1ea1c c\u1ee5": 39,
+    "n\u01b0\u1edbc hoa": 40,
+    "n\u1ed3i chi\u00ean": 41,
+    "n\u1ed3i c\u01a1m": 42,
+    "n\u1ed3i c\u01a1m nh\u1ecf": 43,
+    "n\u1ed3i c\u01a1m \u0111i\u1ec7n": 44,
+    "n\u1ed9i th\u1ea5t": 45,
+    "other": 46,
+    "ph\u1ee5 ki\u1ec7n \u0111i\u1ec7n tho\u1ea1i": 47,
+    "qu\u1ea1t": 48,
+    "qu\u1ea7n": 49,
+    "sd card": 50,
+    "smart tivi": 51,
+    "son m\u00f4i": 52,
+    "s\u00e1ch": 53,
+    "s\u1ea1c d\u1ef1 ph\u00f2ng": 54,
+    "tablet": 55,
+    "tai nghe": 56,
+    "tivi": 57,
+    "trang s\u1ee9c": 58,
+    "t\u00fai x\u00e1ch": 59,
+    "t\u1ee7 l\u1ea1nh": 60,
+    "t\u1ee7 nh\u1ef1a": 61,
+    "t\u1ee7 \u0111\u00f4ng": 62,
+    "vali": 63,
+    "v\u00e1y": 64,
+    "v\u00e1y \u0111\u1ea7m": 65,
+    "watch automatic": 66,
+    "xe h\u01a1i": 67,
+    "xe m\u00e1y": 68,
+    "xe \u0111\u1ea1p": 69,
+    "\u00e1o": 70,
+    "\u00e1o thun": 71,
+    "\u0111i\u1ec1u h\u00f2a": 72,
+    "\u0111i\u1ec7n tho\u1ea1i": 73,
+    "\u0111\u1ed3ng h\u1ed3": 74,
+    "\u0111\u1ed3ng h\u1ed3 th\u00f4ng minh": 75,
+    "\u1ee7ng nh\u1ef1a": 76
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.52.4",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

e5/e5-finetuned/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:867fc47cb99f0842db972b3e58517eb8c36473946ca80a603e8b49d3335fb691
+size 1112435708

e5/e5-finetuned/sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

e5/e5-finetuned/special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

e5/e5-finetuned/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c088c06cf975b7097e469bd69630cdb0d675c6db1ce3af1042b6e19c6d01f22
+size 17082999

e5/e5-finetuned/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

e5/e5-finetuned/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:edbd5f1c185a19a9f2b36fa5321e463d2415cf28b99960bbf120e1c8dcd1a65d
+size 5240