Hugging Face's logo

oneryalcin
/
static-embedding-chess

Sentence Similarity
sentence-transformers
Safetensors
English
static-embedding
chess
retrieval
exploratory
Model card Files
xet
 Community
oneryalcin commited on
Commit
e2f07c0
·
verified ·
1 Parent(s): d2d59e9

Training in progress, step 150

Browse files
Files changed (3) hide show
  1. README.md +31 -21
  2. eval/Information-Retrieval_evaluation_chess-ir_results.csv +1 -0
  3. model.safetensors +1 -1
README.md CHANGED
@@ -64,31 +64,31 @@ model-index:
64
  type: chess-ir
65
  metrics:
66
  - type: cosine_accuracy@1
67
- value: 0.015
68
  name: Cosine Accuracy@1
69
  - type: cosine_accuracy@10
70
- value: 0.135
71
  name: Cosine Accuracy@10
72
  - type: cosine_precision@1
73
- value: 0.015
74
  name: Cosine Precision@1
75
  - type: cosine_precision@10
76
- value: 0.016
77
  name: Cosine Precision@10
78
  - type: cosine_recall@1
79
- value: 0.005
80
  name: Cosine Recall@1
81
  - type: cosine_recall@10
82
- value: 0.05333333333333333
83
  name: Cosine Recall@10
84
  - type: cosine_ndcg@10
85
- value: 0.03352606053277749
86
  name: Cosine Ndcg@10
87
  - type: cosine_mrr@10
88
- value: 0.04136111111111111
89
  name: Cosine Mrr@10
90
  - type: cosine_map@100
91
- value: 0.025214543549657912
92
  name: Cosine Map@100
93
  ---
94
 
@@ -155,7 +155,7 @@ print(query_embeddings.shape, document_embeddings.shape)
155
  # Get the similarity scores for the embeddings
156
  similarities = model.similarity(query_embeddings, document_embeddings)
157
  print(similarities)
158
- # tensor([[ 0.8804, -0.1477, 0.3899]])
159
  ```
160
  <!--
161
  ### Direct Usage (Transformers)
@@ -192,15 +192,15 @@ You can finetune this model on your own dataset.
192
 
193
  | Metric | Value |
194
  |:--------------------|:-----------|
195
- | cosine_accuracy@1 | 0.015 |
196
- | cosine_accuracy@10 | 0.135 |
197
- | cosine_precision@1 | 0.015 |
198
- | cosine_precision@10 | 0.016 |
199
- | cosine_recall@1 | 0.005 |
200
- | cosine_recall@10 | 0.0533 |
201
- | **cosine_ndcg@10** | **0.0335** |
202
- | cosine_mrr@10 | 0.0414 |
203
- | cosine_map@100 | 0.0252 |
204
 
205
  <!--
206
  ## Bias, Risks and Limitations
@@ -395,12 +395,22 @@ You can finetune this model on your own dataset.
395
  | 0.0316 | 90 | 1.3138 | - |
396
  | 0.0334 | 95 | 1.3596 | - |
397
  | 0.0351 | 100 | 1.3428 | 0.0335 |
 
 
 
 
 
 
 
 
 
 
398
 
399
 
400
  ### Training Time
401
- - **Training**: 9.2 seconds
402
  - **Evaluation**: 0.1 seconds
403
- - **Total**: 9.3 seconds
404
 
405
  ### Framework Versions
406
  - Python: 3.12.10
 
64
  type: chess-ir
65
  metrics:
66
  - type: cosine_accuracy@1
67
+ value: 0.02
68
  name: Cosine Accuracy@1
69
  - type: cosine_accuracy@10
70
+ value: 0.12
71
  name: Cosine Accuracy@10
72
  - type: cosine_precision@1
73
+ value: 0.02
74
  name: Cosine Precision@1
75
  - type: cosine_precision@10
76
+ value: 0.0155
77
  name: Cosine Precision@10
78
  - type: cosine_recall@1
79
+ value: 0.006666666666666666
80
  name: Cosine Recall@1
81
  - type: cosine_recall@10
82
+ value: 0.051666666666666666
83
  name: Cosine Recall@10
84
  - type: cosine_ndcg@10
85
+ value: 0.034539315152376744
86
  name: Cosine Ndcg@10
87
  - type: cosine_mrr@10
88
+ value: 0.04391468253968253
89
  name: Cosine Mrr@10
90
  - type: cosine_map@100
91
+ value: 0.02851338765635309
92
  name: Cosine Map@100
93
  ---
94
 
 
155
  # Get the similarity scores for the embeddings
156
  similarities = model.similarity(query_embeddings, document_embeddings)
157
  print(similarities)
158
+ # tensor([[ 0.9140, -0.1872, 0.3933]])
159
  ```
160
  <!--
161
  ### Direct Usage (Transformers)
 
192
 
193
  | Metric | Value |
194
  |:--------------------|:-----------|
195
+ | cosine_accuracy@1 | 0.02 |
196
+ | cosine_accuracy@10 | 0.12 |
197
+ | cosine_precision@1 | 0.02 |
198
+ | cosine_precision@10 | 0.0155 |
199
+ | cosine_recall@1 | 0.0067 |
200
+ | cosine_recall@10 | 0.0517 |
201
+ | **cosine_ndcg@10** | **0.0345** |
202
+ | cosine_mrr@10 | 0.0439 |
203
+ | cosine_map@100 | 0.0285 |
204
 
205
  <!--
206
  ## Bias, Risks and Limitations
 
395
  | 0.0316 | 90 | 1.3138 | - |
396
  | 0.0334 | 95 | 1.3596 | - |
397
  | 0.0351 | 100 | 1.3428 | 0.0335 |
398
+ | 0.0369 | 105 | 1.3302 | - |
399
+ | 0.0386 | 110 | 1.3281 | - |
400
+ | 0.0404 | 115 | 1.3520 | - |
401
+ | 0.0421 | 120 | 1.3127 | - |
402
+ | 0.0439 | 125 | 1.3362 | - |
403
+ | 0.0456 | 130 | 1.3174 | - |
404
+ | 0.0474 | 135 | 1.3103 | - |
405
+ | 0.0492 | 140 | 1.3428 | - |
406
+ | 0.0509 | 145 | 1.2886 | - |
407
+ | 0.0527 | 150 | 1.2895 | 0.0345 |
408
 
409
 
410
  ### Training Time
411
+ - **Training**: 13.2 seconds
412
  - **Evaluation**: 0.1 seconds
413
+ - **Total**: 13.3 seconds
414
 
415
  ### Framework Versions
416
  - Python: 3.12.10
eval/Information-Retrieval_evaluation_chess-ir_results.csv CHANGED
@@ -1,3 +1,4 @@
1
  epoch,steps,cosine-Accuracy@1,cosine-Accuracy@10,cosine-Precision@1,cosine-Recall@1,cosine-Precision@10,cosine-Recall@10,cosine-MRR@10,cosine-NDCG@10,cosine-MAP@100
2
  0.0175561797752809,50,0.015,0.115,0.015,0.005,0.013000000000000001,0.04333333333333333,0.03541269841269841,0.02770564804107805,0.021195015342589062
3
  0.0351123595505618,100,0.015,0.135,0.015,0.005,0.016,0.05333333333333333,0.04136111111111111,0.03352606053277749,0.025214543549657912
 
 
1
  epoch,steps,cosine-Accuracy@1,cosine-Accuracy@10,cosine-Precision@1,cosine-Recall@1,cosine-Precision@10,cosine-Recall@10,cosine-MRR@10,cosine-NDCG@10,cosine-MAP@100
2
  0.0175561797752809,50,0.015,0.115,0.015,0.005,0.013000000000000001,0.04333333333333333,0.03541269841269841,0.02770564804107805,0.021195015342589062
3
  0.0351123595505618,100,0.015,0.135,0.015,0.005,0.016,0.05333333333333333,0.04136111111111111,0.03352606053277749,0.025214543549657912
4
+ 0.05266853932584269,150,0.02,0.12,0.02,0.006666666666666666,0.0155,0.051666666666666666,0.04391468253968253,0.034539315152376744,0.02851338765635309
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5d4ef3bb5b44abe3bfb9417a3efa9ca89a41ae764f01f72b81aee073532c9c5f
3
  size 4343904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5645fc31a78d305cc82c0d937c9ceecfd92d40fa95c3e3d81de94d475b2b109
3
  size 4343904