Add new SentenceTransformer model

README.md

@@ -7,7 +7,7 @@ tags:
 - generated_from_trainer
 - dataset_size:90000
 - loss:MultipleNegativesRankingLoss
-base_model: sentence-transformers/all-MiniLM-L6-v2
+base_model: sentence-transformers/all-MiniLM-L12-v2
 widget:
 - source_sentence: who is the publisher of the norton anthology american literature
   sentences:
@@ -154,7 +154,7 @@ metrics:
 - cosine_mrr@10
 - cosine_map@100
 model-index:
-- name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
+- name: SentenceTransformer based on sentence-transformers/all-MiniLM-L12-v2
   results:
   - task:
       type: information-retrieval
@@ -164,49 +164,49 @@ model-index:
       type: NanoMSMARCO
     metrics:
     - type: cosine_accuracy@1
-      value: 0.24
+      value: 0.34
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.52
+      value: 0.54
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
       value: 0.64
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.76
+      value: 0.78
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.24
+      value: 0.34
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.1733333333333333
+      value: 0.18
       name: Cosine Precision@3
     - type: cosine_precision@5
       value: 0.128
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.07600000000000001
+      value: 0.07800000000000001
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.24
+      value: 0.34
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.52
+      value: 0.54
       name: Cosine Recall@3
     - type: cosine_recall@5
       value: 0.64
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.76
+      value: 0.78
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.49143458252672184
+      value: 0.5447080049645561
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.4057380952380952
+      value: 0.47073809523809523
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.4167225925544634
+      value: 0.4806962957327628
       name: Cosine Map@100
   - task:
       type: information-retrieval
@@ -216,49 +216,49 @@ model-index:
       type: NanoNQ
     metrics:
     - type: cosine_accuracy@1
-      value: 0.46
+      value: 0.44
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
       value: 0.62
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.66
+      value: 0.7
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.7
+      value: 0.78
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.46
+      value: 0.44
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.20666666666666664
+      value: 0.21333333333333332
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.136
+      value: 0.14800000000000002
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.07600000000000001
+      value: 0.08199999999999999
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.45
+      value: 0.43
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.58
+      value: 0.61
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.63
+      value: 0.67
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.69
+      value: 0.74
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.5748041984771892
+      value: 0.5924173512360595
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.5456666666666667
+      value: 0.5506349206349206
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.5421576333440519
+      value: 0.5491036387356644
       name: Cosine Map@100
   - task:
       type: nano-beir
@@ -268,61 +268,61 @@ model-index:
       type: NanoBEIR_mean
     metrics:
     - type: cosine_accuracy@1
-      value: 0.35
+      value: 0.39
       name: Cosine Accuracy@1
     - type: cosine_accuracy@3
-      value: 0.5700000000000001
+      value: 0.5800000000000001
       name: Cosine Accuracy@3
     - type: cosine_accuracy@5
-      value: 0.65
+      value: 0.6699999999999999
       name: Cosine Accuracy@5
     - type: cosine_accuracy@10
-      value: 0.73
+      value: 0.78
       name: Cosine Accuracy@10
     - type: cosine_precision@1
-      value: 0.35
+      value: 0.39
       name: Cosine Precision@1
     - type: cosine_precision@3
-      value: 0.18999999999999997
+      value: 0.19666666666666666
       name: Cosine Precision@3
     - type: cosine_precision@5
-      value: 0.132
+      value: 0.138
       name: Cosine Precision@5
     - type: cosine_precision@10
-      value: 0.07600000000000001
+      value: 0.08
       name: Cosine Precision@10
     - type: cosine_recall@1
-      value: 0.345
+      value: 0.385
       name: Cosine Recall@1
     - type: cosine_recall@3
-      value: 0.55
+      value: 0.575
       name: Cosine Recall@3
     - type: cosine_recall@5
-      value: 0.635
+      value: 0.655
       name: Cosine Recall@5
     - type: cosine_recall@10
-      value: 0.725
+      value: 0.76
       name: Cosine Recall@10
     - type: cosine_ndcg@10
-      value: 0.5331193905019556
+      value: 0.5685626781003078
       name: Cosine Ndcg@10
     - type: cosine_mrr@10
-      value: 0.47570238095238093
+      value: 0.5106865079365079
       name: Cosine Mrr@10
     - type: cosine_map@100
-      value: 0.47944011294925765
+      value: 0.5148999672342136
       name: Cosine Map@100
 ---
 
-# SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
+# SentenceTransformer based on sentence-transformers/all-MiniLM-L12-v2
 
-This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
+This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L12-v2). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
 
 ## Model Details
 
 ### Model Description
 - **Model Type:** Sentence Transformer
-- **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision c9745ed1d9f207416be6d2e6f8de32d1f16199bf -->
+- **Base model:** [sentence-transformers/all-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L12-v2) <!-- at revision 936af83a2ecce5fe87a09109ff5cbcefe073173a -->
 - **Maximum Sequence Length:** 128 tokens
 - **Output Dimensionality:** 384 dimensions
 - **Similarity Function:** Cosine Similarity
@@ -375,9 +375,9 @@ print(embeddings.shape)
 # Get the similarity scores for the embeddings
 similarities = model.similarity(embeddings, embeddings)
 print(similarities)
-# tensor([[1.0000, 0.7393, 0.1251],
-#         [0.7393, 1.0000, 0.1255],
-#         [0.1251, 0.1255, 1.0000]])
+# tensor([[ 1.0000,  0.7187, -0.0053],
+#         [ 0.7187,  1.0000,  0.0412],
+#         [-0.0053,  0.0412,  1.0000]])
 ```
 
 <!--
@@ -415,21 +415,21 @@ You can finetune this model on your own dataset.
 
 | Metric              | NanoMSMARCO | NanoNQ     |
 |:--------------------|:------------|:-----------|
-| cosine_accuracy@1   | 0.24        | 0.46       |
-| cosine_accuracy@3   | 0.52        | 0.62       |
-| cosine_accuracy@5   | 0.64        | 0.66       |
-| cosine_accuracy@10  | 0.76        | 0.7        |
-| cosine_precision@1  | 0.24        | 0.46       |
-| cosine_precision@3  | 0.1733      | 0.2067     |
-| cosine_precision@5  | 0.128       | 0.136      |
-| cosine_precision@10 | 0.076       | 0.076      |
-| cosine_recall@1     | 0.24        | 0.45       |
-| cosine_recall@3     | 0.52        | 0.58       |
-| cosine_recall@5     | 0.64        | 0.63       |
-| cosine_recall@10    | 0.76        | 0.69       |
-| **cosine_ndcg@10**  | **0.4914**  | **0.5748** |
-| cosine_mrr@10       | 0.4057      | 0.5457     |
-| cosine_map@100      | 0.4167      | 0.5422     |
+| cosine_accuracy@1   | 0.34        | 0.44       |
+| cosine_accuracy@3   | 0.54        | 0.62       |
+| cosine_accuracy@5   | 0.64        | 0.7        |
+| cosine_accuracy@10  | 0.78        | 0.78       |
+| cosine_precision@1  | 0.34        | 0.44       |
+| cosine_precision@3  | 0.18        | 0.2133     |
+| cosine_precision@5  | 0.128       | 0.148      |
+| cosine_precision@10 | 0.078       | 0.082      |
+| cosine_recall@1     | 0.34        | 0.43       |
+| cosine_recall@3     | 0.54        | 0.61       |
+| cosine_recall@5     | 0.64        | 0.67       |
+| cosine_recall@10    | 0.78        | 0.74       |
+| **cosine_ndcg@10**  | **0.5447**  | **0.5924** |
+| cosine_mrr@10       | 0.4707      | 0.5506     |
+| cosine_map@100      | 0.4807      | 0.5491     |
 
 #### Nano BEIR
 
@@ -447,21 +447,21 @@ You can finetune this model on your own dataset.
 
 | Metric              | Value      |
 |:--------------------|:-----------|
-| cosine_accuracy@1   | 0.35       |
-| cosine_accuracy@3   | 0.57       |
-| cosine_accuracy@5   | 0.65       |
-| cosine_accuracy@10  | 0.73       |
-| cosine_precision@1  | 0.35       |
-| cosine_precision@3  | 0.19       |
-| cosine_precision@5  | 0.132      |
-| cosine_precision@10 | 0.076      |
-| cosine_recall@1     | 0.345      |
-| cosine_recall@3     | 0.55       |
-| cosine_recall@5     | 0.635      |
-| cosine_recall@10    | 0.725      |
-| **cosine_ndcg@10**  | **0.5331** |
-| cosine_mrr@10       | 0.4757     |
-| cosine_map@100      | 0.4794     |
+| cosine_accuracy@1   | 0.39       |
+| cosine_accuracy@3   | 0.58       |
+| cosine_accuracy@5   | 0.67       |
+| cosine_accuracy@10  | 0.78       |
+| cosine_precision@1  | 0.39       |
+| cosine_precision@3  | 0.1967     |
+| cosine_precision@5  | 0.138      |
+| cosine_precision@10 | 0.08       |
+| cosine_recall@1     | 0.385      |
+| cosine_recall@3     | 0.575      |
+| cosine_recall@5     | 0.655      |
+| cosine_recall@10    | 0.76       |
+| **cosine_ndcg@10**  | **0.5686** |
+| cosine_mrr@10       | 0.5107     |
+| cosine_map@100      | 0.5149     |
 
 <!--
 ## Bias, Risks and Limitations
@@ -535,9 +535,9 @@ You can finetune this model on your own dataset.
 - `eval_strategy`: steps
 - `per_device_train_batch_size`: 128
 - `per_device_eval_batch_size`: 128
-- `learning_rate`: 0.0001
-- `weight_decay`: 0.001
-- `max_steps`: 1687
+- `learning_rate`: 8e-05
+- `weight_decay`: 0.005
+- `max_steps`: 1125
 - `warmup_ratio`: 0.1
 - `fp16`: True
 - `dataloader_drop_last`: True
@@ -564,14 +564,14 @@ You can finetune this model on your own dataset.
 - `gradient_accumulation_steps`: 1
 - `eval_accumulation_steps`: None
 - `torch_empty_cache_steps`: None
-- `learning_rate`: 0.0001
-- `weight_decay`: 0.001
+- `learning_rate`: 8e-05
+- `weight_decay`: 0.005
 - `adam_beta1`: 0.9
 - `adam_beta2`: 0.999
 - `adam_epsilon`: 1e-08
 - `max_grad_norm`: 1.0
 - `num_train_epochs`: 3.0
-- `max_steps`: 1687
+- `max_steps`: 1125
 - `lr_scheduler_type`: linear
 - `lr_scheduler_kwargs`: {}
 - `warmup_ratio`: 0.1
@@ -678,13 +678,11 @@ You can finetune this model on your own dataset.
 ### Training Logs
 | Epoch  | Step | Training Loss | Validation Loss | NanoMSMARCO_cosine_ndcg@10 | NanoNQ_cosine_ndcg@10 | NanoBEIR_mean_cosine_ndcg@10 |
 |:------:|:----:|:-------------:|:---------------:|:--------------------------:|:---------------------:|:----------------------------:|
-| 0      | 0    | -             | 0.0803          | 0.5540                     | 0.5931                | 0.5735                       |
-| 0.3556 | 250  | 0.0897        | 0.0764          | 0.5354                     | 0.5945                | 0.5650                       |
-| 0.7112 | 500  | 0.0917        | 0.0749          | 0.5252                     | 0.5638                | 0.5445                       |
-| 1.0669 | 750  | 0.0823        | 0.0653          | 0.5267                     | 0.5904                | 0.5586                       |
-| 1.4225 | 1000 | 0.0459        | 0.0630          | 0.5236                     | 0.5689                | 0.5462                       |
-| 1.7781 | 1250 | 0.0429        | 0.0613          | 0.5312                     | 0.5676                | 0.5494                       |
-| 2.1337 | 1500 | 0.0393        | 0.0600          | 0.4914                     | 0.5748                | 0.5331                       |
+| 0      | 0    | -             | 0.0731          | 0.5887                     | 0.5786                | 0.5836                       |
+| 0.3556 | 250  | 0.0821        | 0.0701          | 0.5325                     | 0.5977                | 0.5651                       |
+| 0.7112 | 500  | 0.0805        | 0.0640          | 0.5523                     | 0.5631                | 0.5577                       |
+| 1.0669 | 750  | 0.0712        | 0.0572          | 0.5369                     | 0.5819                | 0.5594                       |
+| 1.4225 | 1000 | 0.0371        | 0.0551          | 0.5447                     | 0.5924                | 0.5686                       |
 
 
 ### Framework Versions