Sentence Similarity
sentence-transformers
Safetensors
Chinese
English
text-embeddings-inference
feature-extraction
semantic-search
retrieval
traditional-chinese
lora
Instructions to use BluePlanetAI/BPVELA-G300M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use BluePlanetAI/BPVELA-G300M with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("BluePlanetAI/BPVELA-G300M") sentences = [ "那是 個快樂的人", "那是 條快樂的狗", "那是 個非常幸福的人", "今天是晴天" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
Upload BPVELA-G300M-v1.0.0
Browse files
README.md
CHANGED
|
@@ -37,6 +37,14 @@ base_model:
|
|
| 37 |
|
| 38 |
這個 repository 釋出的是 LoRA adapter,不是 merged full checkpoint。使用時需要以 base model 為底,再載入這個 adapter。
|
| 39 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
### 驗證摘要
|
| 41 |
|
| 42 |
- Taiwan-md pair benchmark:Spearman `0.8319`、Pearson `0.8953`
|
|
@@ -86,6 +94,14 @@ BPVELA-G300M 的訓練與優化過程使用了 Taiwan-MD 內容;依目前資
|
|
| 86 |
|
| 87 |
This repository contains a LoRA adapter release, not a merged full checkpoint. Load it on top of the base model.
|
| 88 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 89 |
## Validation Snapshot
|
| 90 |
|
| 91 |
- Taiwan-md pair benchmark: Spearman `0.8319`, Pearson `0.8953`
|
|
|
|
| 37 |
|
| 38 |
這個 repository 釋出的是 LoRA adapter,不是 merged full checkpoint。使用時需要以 base model 為底,再載入這個 adapter。
|
| 39 |
|
| 40 |
+
### 存取前置條件
|
| 41 |
+
|
| 42 |
+
`BPVELA-G300M` 建立於 `google/embeddinggemma-300m` 之上,因此除了本 adapter repository 之外,使用者也必須能夠存取上游 Gemma base model。
|
| 43 |
+
|
| 44 |
+
- 請先在 Hugging Face 上完成 `google/embeddinggemma-300m` 的 gated access 申請與條款同意
|
| 45 |
+
- 若使用 fine-grained token,請確認 token 已開啟 public gated repositories 的讀取權限
|
| 46 |
+
- 若載入時出現 `401 Unauthorized` 或 `403 Forbidden`,且訊息指向 `google/embeddinggemma-300m/resolve/...`,通常表示缺少上游 Gemma 存取權,而不是本 adapter repository 本身有問題
|
| 47 |
+
|
| 48 |
### 驗證摘要
|
| 49 |
|
| 50 |
- Taiwan-md pair benchmark:Spearman `0.8319`、Pearson `0.8953`
|
|
|
|
| 94 |
|
| 95 |
This repository contains a LoRA adapter release, not a merged full checkpoint. Load it on top of the base model.
|
| 96 |
|
| 97 |
+
## Access Requirements
|
| 98 |
+
|
| 99 |
+
`BPVELA-G300M` is built on top of `google/embeddinggemma-300m`, so users must be able to access the upstream Gemma base model in addition to this adapter repository.
|
| 100 |
+
|
| 101 |
+
- Request and accept gated access for `google/embeddinggemma-300m` on Hugging Face first
|
| 102 |
+
- If you use a fine-grained token, enable read access to public gated repositories
|
| 103 |
+
- If loading fails with `401 Unauthorized` or `403 Forbidden` against `google/embeddinggemma-300m/resolve/...`, the issue is usually missing upstream Gemma access rather than a problem with this adapter repository
|
| 104 |
+
|
| 105 |
## Validation Snapshot
|
| 106 |
|
| 107 |
- Taiwan-md pair benchmark: Spearman `0.8319`, Pearson `0.8953`
|