sasha-smirnov commited on
Commit
5ea8433
·
verified ·
1 Parent(s): 2f8ac1d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -18
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: 71.0
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: 71.0
1114
  - type: precision_at_10
1115
- value: 35.0
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: 64.0
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: 64.0
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: 64.0
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: 100.0
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: 91.0
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: 93.0
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: 82.0
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: 79.0
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: 82.0
2250
  - type: precision_at_10
2251
- value: 78.0
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)