Update README.md
Browse files
README.md
CHANGED
@@ -1086,7 +1086,7 @@ model-index:
|
|
1086 |
- type: map_at_5
|
1087 |
value: 17.471999999999998
|
1088 |
- type: mrr_at_1
|
1089 |
-
value: 71
|
1090 |
- type: mrr_at_10
|
1091 |
value: 79.176
|
1092 |
- type: mrr_at_100
|
@@ -1110,9 +1110,9 @@ model-index:
|
|
1110 |
- type: ndcg_at_5
|
1111 |
value: 45.826
|
1112 |
- type: precision_at_1
|
1113 |
-
value: 71
|
1114 |
- type: precision_at_10
|
1115 |
-
value: 35
|
1116 |
- type: precision_at_100
|
1117 |
value: 11.360000000000001
|
1118 |
- type: precision_at_1000
|
@@ -2046,7 +2046,7 @@ model-index:
|
|
2046 |
- type: map_at_5
|
2047 |
value: 68.447
|
2048 |
- type: mrr_at_1
|
2049 |
-
value: 64
|
2050 |
- type: mrr_at_10
|
2051 |
value: 71.212
|
2052 |
- type: mrr_at_100
|
@@ -2058,7 +2058,7 @@ model-index:
|
|
2058 |
- type: mrr_at_5
|
2059 |
value: 70.094
|
2060 |
- type: ndcg_at_1
|
2061 |
-
value: 64
|
2062 |
- type: ndcg_at_10
|
2063 |
value: 74.607
|
2064 |
- type: ndcg_at_100
|
@@ -2070,7 +2070,7 @@ model-index:
|
|
2070 |
- type: ndcg_at_5
|
2071 |
value: 71.41300000000001
|
2072 |
- type: precision_at_1
|
2073 |
-
value: 64
|
2074 |
- type: precision_at_10
|
2075 |
value: 9.933
|
2076 |
- type: precision_at_100
|
@@ -2088,7 +2088,7 @@ model-index:
|
|
2088 |
- type: recall_at_100
|
2089 |
value: 94.833
|
2090 |
- type: recall_at_1000
|
2091 |
-
value: 100
|
2092 |
- type: recall_at_3
|
2093 |
value: 72.628
|
2094 |
- type: recall_at_5
|
@@ -2121,7 +2121,7 @@ model-index:
|
|
2121 |
- type: dot_precision
|
2122 |
value: 87.16475095785441
|
2123 |
- type: dot_recall
|
2124 |
-
value: 91
|
2125 |
- type: euclidean_accuracy
|
2126 |
value: 99.85445544554456
|
2127 |
- type: euclidean_ap
|
@@ -2131,7 +2131,7 @@ model-index:
|
|
2131 |
- type: euclidean_precision
|
2132 |
value: 92.17046580773042
|
2133 |
- type: euclidean_recall
|
2134 |
-
value: 93
|
2135 |
- type: manhattan_accuracy
|
2136 |
value: 99.85445544554456
|
2137 |
- type: manhattan_ap
|
@@ -2222,7 +2222,7 @@ model-index:
|
|
2222 |
- type: map_at_5
|
2223 |
value: 1.077
|
2224 |
- type: mrr_at_1
|
2225 |
-
value: 82
|
2226 |
- type: mrr_at_10
|
2227 |
value: 89.86699999999999
|
2228 |
- type: mrr_at_100
|
@@ -2234,7 +2234,7 @@ model-index:
|
|
2234 |
- type: mrr_at_5
|
2235 |
value: 89.667
|
2236 |
- type: ndcg_at_1
|
2237 |
-
value: 79
|
2238 |
- type: ndcg_at_10
|
2239 |
value: 74.818
|
2240 |
- type: ndcg_at_100
|
@@ -2246,9 +2246,9 @@ model-index:
|
|
2246 |
- type: ndcg_at_5
|
2247 |
value: 79.81899999999999
|
2248 |
- type: precision_at_1
|
2249 |
-
value: 82
|
2250 |
- type: precision_at_10
|
2251 |
-
value: 78
|
2252 |
- type: precision_at_100
|
2253 |
value: 54.48
|
2254 |
- type: precision_at_1000
|
@@ -2490,14 +2490,11 @@ model-index:
|
|
2490 |
task:
|
2491 |
type: PairClassification
|
2492 |
tags:
|
2493 |
-
- sentence-transformers
|
2494 |
- feature-extraction
|
2495 |
- sentence-similarity
|
2496 |
-
- transformers
|
2497 |
- mteb
|
2498 |
- onnx
|
2499 |
- teradata
|
2500 |
-
|
2501 |
---
|
2502 |
# A Teradata Vantage compatible Embeddings Model
|
2503 |
|
@@ -2649,5 +2646,4 @@ print("Cosine similiarity for embeddings calculated with ONNX:" + str(cos_sim(em
|
|
2649 |
print("Cosine similiarity for embeddings calculated with SentenceTransformer:" + str(cos_sim(embeddings_1_sentence_transformer, embeddings_2_sentence_transformer)))
|
2650 |
```
|
2651 |
|
2652 |
-
You can find the detailed ONNX vs. SentenceTransformer result comparison steps in the file [test_local.py](./test_local.py)
|
2653 |
-
|
|
|
1086 |
- type: map_at_5
|
1087 |
value: 17.471999999999998
|
1088 |
- type: mrr_at_1
|
1089 |
+
value: 71
|
1090 |
- type: mrr_at_10
|
1091 |
value: 79.176
|
1092 |
- type: mrr_at_100
|
|
|
1110 |
- type: ndcg_at_5
|
1111 |
value: 45.826
|
1112 |
- type: precision_at_1
|
1113 |
+
value: 71
|
1114 |
- type: precision_at_10
|
1115 |
+
value: 35
|
1116 |
- type: precision_at_100
|
1117 |
value: 11.360000000000001
|
1118 |
- type: precision_at_1000
|
|
|
2046 |
- type: map_at_5
|
2047 |
value: 68.447
|
2048 |
- type: mrr_at_1
|
2049 |
+
value: 64
|
2050 |
- type: mrr_at_10
|
2051 |
value: 71.212
|
2052 |
- type: mrr_at_100
|
|
|
2058 |
- type: mrr_at_5
|
2059 |
value: 70.094
|
2060 |
- type: ndcg_at_1
|
2061 |
+
value: 64
|
2062 |
- type: ndcg_at_10
|
2063 |
value: 74.607
|
2064 |
- type: ndcg_at_100
|
|
|
2070 |
- type: ndcg_at_5
|
2071 |
value: 71.41300000000001
|
2072 |
- type: precision_at_1
|
2073 |
+
value: 64
|
2074 |
- type: precision_at_10
|
2075 |
value: 9.933
|
2076 |
- type: precision_at_100
|
|
|
2088 |
- type: recall_at_100
|
2089 |
value: 94.833
|
2090 |
- type: recall_at_1000
|
2091 |
+
value: 100
|
2092 |
- type: recall_at_3
|
2093 |
value: 72.628
|
2094 |
- type: recall_at_5
|
|
|
2121 |
- type: dot_precision
|
2122 |
value: 87.16475095785441
|
2123 |
- type: dot_recall
|
2124 |
+
value: 91
|
2125 |
- type: euclidean_accuracy
|
2126 |
value: 99.85445544554456
|
2127 |
- type: euclidean_ap
|
|
|
2131 |
- type: euclidean_precision
|
2132 |
value: 92.17046580773042
|
2133 |
- type: euclidean_recall
|
2134 |
+
value: 93
|
2135 |
- type: manhattan_accuracy
|
2136 |
value: 99.85445544554456
|
2137 |
- type: manhattan_ap
|
|
|
2222 |
- type: map_at_5
|
2223 |
value: 1.077
|
2224 |
- type: mrr_at_1
|
2225 |
+
value: 82
|
2226 |
- type: mrr_at_10
|
2227 |
value: 89.86699999999999
|
2228 |
- type: mrr_at_100
|
|
|
2234 |
- type: mrr_at_5
|
2235 |
value: 89.667
|
2236 |
- type: ndcg_at_1
|
2237 |
+
value: 79
|
2238 |
- type: ndcg_at_10
|
2239 |
value: 74.818
|
2240 |
- type: ndcg_at_100
|
|
|
2246 |
- type: ndcg_at_5
|
2247 |
value: 79.81899999999999
|
2248 |
- type: precision_at_1
|
2249 |
+
value: 82
|
2250 |
- type: precision_at_10
|
2251 |
+
value: 78
|
2252 |
- type: precision_at_100
|
2253 |
value: 54.48
|
2254 |
- type: precision_at_1000
|
|
|
2490 |
task:
|
2491 |
type: PairClassification
|
2492 |
tags:
|
|
|
2493 |
- feature-extraction
|
2494 |
- sentence-similarity
|
|
|
2495 |
- mteb
|
2496 |
- onnx
|
2497 |
- teradata
|
|
|
2498 |
---
|
2499 |
# A Teradata Vantage compatible Embeddings Model
|
2500 |
|
|
|
2646 |
print("Cosine similiarity for embeddings calculated with SentenceTransformer:" + str(cos_sim(embeddings_1_sentence_transformer, embeddings_2_sentence_transformer)))
|
2647 |
```
|
2648 |
|
2649 |
+
You can find the detailed ONNX vs. SentenceTransformer result comparison steps in the file [test_local.py](./test_local.py)
|
|