Update README.md
Browse files
README.md
CHANGED
@@ -2384,7 +2384,7 @@ model-index:
|
|
2384 |
- type: precision_at_10
|
2385 |
value: 8.493
|
2386 |
- type: precision_at_20
|
2387 |
-
value: 5
|
2388 |
- type: precision_at_100
|
2389 |
value: 1.312
|
2390 |
- type: precision_at_1000
|
@@ -4426,7 +4426,7 @@ model-index:
|
|
4426 |
- type: recall_at_1000
|
4427 |
value: 75.224
|
4428 |
- type: precision_at_1
|
4429 |
-
value: 72
|
4430 |
- type: precision_at_3
|
4431 |
value: 50.917
|
4432 |
- type: precision_at_5
|
@@ -4440,7 +4440,7 @@ model-index:
|
|
4440 |
- type: precision_at_1000
|
4441 |
value: 2.0549999999999997
|
4442 |
- type: mrr_at_1
|
4443 |
-
value: 72
|
4444 |
- type: mrr_at_3
|
4445 |
value: 77.5417
|
4446 |
- type: mrr_at_5
|
@@ -8225,7 +8225,7 @@ model-index:
|
|
8225 |
type: mteb/trec-covid
|
8226 |
metrics:
|
8227 |
- type: ndcg_at_1
|
8228 |
-
value: 86
|
8229 |
- type: ndcg_at_3
|
8230 |
value: 86.542
|
8231 |
- type: ndcg_at_5
|
@@ -8267,11 +8267,11 @@ model-index:
|
|
8267 |
- type: recall_at_1000
|
8268 |
value: 54.290000000000006
|
8269 |
- type: precision_at_1
|
8270 |
-
value: 90
|
8271 |
- type: precision_at_3
|
8272 |
-
value: 92
|
8273 |
- type: precision_at_5
|
8274 |
-
value: 90
|
8275 |
- type: precision_at_10
|
8276 |
value: 88.6
|
8277 |
- type: precision_at_20
|
@@ -8281,7 +8281,7 @@ model-index:
|
|
8281 |
- type: precision_at_1000
|
8282 |
value: 25.81
|
8283 |
- type: mrr_at_1
|
8284 |
-
value: 90
|
8285 |
- type: mrr_at_3
|
8286 |
value: 94.6667
|
8287 |
- type: mrr_at_5
|
@@ -9036,16 +9036,13 @@ model-index:
|
|
9036 |
type: PairClassification
|
9037 |
pipeline_tag: sentence-similarity
|
9038 |
tags:
|
9039 |
-
- sentence-transformers
|
9040 |
- feature-extraction
|
9041 |
- sentence-similarity
|
9042 |
- mteb
|
9043 |
- arctic
|
9044 |
- snowflake-arctic-embed
|
9045 |
-
- transformers.js
|
9046 |
- onnx
|
9047 |
- teradata
|
9048 |
-
|
9049 |
---
|
9050 |
# A Teradata Vantage compatible Embeddings Model
|
9051 |
|
@@ -9197,5 +9194,4 @@ print("Cosine similiarity for embeddings calculated with ONNX:" + str(cos_sim(em
|
|
9197 |
print("Cosine similiarity for embeddings calculated with SentenceTransformer:" + str(cos_sim(embeddings_1_sentence_transformer, embeddings_2_sentence_transformer)))
|
9198 |
```
|
9199 |
|
9200 |
-
You can find the detailed ONNX vs. SentenceTransformer result comparison steps in the file [test_local.py](./test_local.py)
|
9201 |
-
|
|
|
2384 |
- type: precision_at_10
|
2385 |
value: 8.493
|
2386 |
- type: precision_at_20
|
2387 |
+
value: 5
|
2388 |
- type: precision_at_100
|
2389 |
value: 1.312
|
2390 |
- type: precision_at_1000
|
|
|
4426 |
- type: recall_at_1000
|
4427 |
value: 75.224
|
4428 |
- type: precision_at_1
|
4429 |
+
value: 72
|
4430 |
- type: precision_at_3
|
4431 |
value: 50.917
|
4432 |
- type: precision_at_5
|
|
|
4440 |
- type: precision_at_1000
|
4441 |
value: 2.0549999999999997
|
4442 |
- type: mrr_at_1
|
4443 |
+
value: 72
|
4444 |
- type: mrr_at_3
|
4445 |
value: 77.5417
|
4446 |
- type: mrr_at_5
|
|
|
8225 |
type: mteb/trec-covid
|
8226 |
metrics:
|
8227 |
- type: ndcg_at_1
|
8228 |
+
value: 86
|
8229 |
- type: ndcg_at_3
|
8230 |
value: 86.542
|
8231 |
- type: ndcg_at_5
|
|
|
8267 |
- type: recall_at_1000
|
8268 |
value: 54.290000000000006
|
8269 |
- type: precision_at_1
|
8270 |
+
value: 90
|
8271 |
- type: precision_at_3
|
8272 |
+
value: 92
|
8273 |
- type: precision_at_5
|
8274 |
+
value: 90
|
8275 |
- type: precision_at_10
|
8276 |
value: 88.6
|
8277 |
- type: precision_at_20
|
|
|
8281 |
- type: precision_at_1000
|
8282 |
value: 25.81
|
8283 |
- type: mrr_at_1
|
8284 |
+
value: 90
|
8285 |
- type: mrr_at_3
|
8286 |
value: 94.6667
|
8287 |
- type: mrr_at_5
|
|
|
9036 |
type: PairClassification
|
9037 |
pipeline_tag: sentence-similarity
|
9038 |
tags:
|
|
|
9039 |
- feature-extraction
|
9040 |
- sentence-similarity
|
9041 |
- mteb
|
9042 |
- arctic
|
9043 |
- snowflake-arctic-embed
|
|
|
9044 |
- onnx
|
9045 |
- teradata
|
|
|
9046 |
---
|
9047 |
# A Teradata Vantage compatible Embeddings Model
|
9048 |
|
|
|
9194 |
print("Cosine similiarity for embeddings calculated with SentenceTransformer:" + str(cos_sim(embeddings_1_sentence_transformer, embeddings_2_sentence_transformer)))
|
9195 |
```
|
9196 |
|
9197 |
+
You can find the detailed ONNX vs. SentenceTransformer result comparison steps in the file [test_local.py](./test_local.py)
|
|