Sentence Similarity
sentence-transformers
Safetensors
Japanese
modernbert
medical
japanese
ruri
embedding
text-embeddings-inference
Instructions to use genshiai-daichi/med-ruri-v3-310m-v2-from-med with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use genshiai-daichi/med-ruri-v3-310m-v2-from-med with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("genshiai-daichi/med-ruri-v3-310m-v2-from-med") sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Notebooks
- Google Colab
- Kaggle
latest = step 9238 (nDCG@10=0.5348)
Browse files- model.safetensors +1 -1
- trainer_state.json +53 -6
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1258462760
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3dbc0607278096f2ff7c9f006a3d96f225ba870166156b0152f77eb210bf925
|
| 3 |
size 1258462760
|
trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 0.
|
| 4 |
-
"best_model_checkpoint": "checkpoints/med-ruri-v3-310m-v2-from-med/checkpoint-
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -1610,6 +1610,53 @@
|
|
| 1610 |
"eval_samples_per_second": 0.0,
|
| 1611 |
"eval_steps_per_second": 0.0,
|
| 1612 |
"step": 9000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1613 |
}
|
| 1614 |
],
|
| 1615 |
"logging_steps": 50,
|
|
@@ -1624,7 +1671,7 @@
|
|
| 1624 |
"should_evaluate": false,
|
| 1625 |
"should_log": false,
|
| 1626 |
"should_save": true,
|
| 1627 |
-
"should_training_stop":
|
| 1628 |
},
|
| 1629 |
"attributes": {}
|
| 1630 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 9238,
|
| 3 |
+
"best_metric": 0.5347855089896354,
|
| 4 |
+
"best_model_checkpoint": "checkpoints/med-ruri-v3-310m-v2-from-med/checkpoint-9238",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 9238,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 1610 |
"eval_samples_per_second": 0.0,
|
| 1611 |
"eval_steps_per_second": 0.0,
|
| 1612 |
"step": 9000
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.9796492747347911,
|
| 1616 |
+
"grad_norm": 14.185355186462402,
|
| 1617 |
+
"learning_rate": 2.1536007292616228e-07,
|
| 1618 |
+
"loss": 0.22177120208740234,
|
| 1619 |
+
"step": 9050
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.9850617016670274,
|
| 1623 |
+
"grad_norm": 12.419501304626465,
|
| 1624 |
+
"learning_rate": 1.5838650865998178e-07,
|
| 1625 |
+
"loss": 0.20167236328125,
|
| 1626 |
+
"step": 9100
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.9904741285992639,
|
| 1630 |
+
"grad_norm": 10.942581176757812,
|
| 1631 |
+
"learning_rate": 1.0141294439380129e-07,
|
| 1632 |
+
"loss": 0.20480478286743165,
|
| 1633 |
+
"step": 9150
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.9958865555315003,
|
| 1637 |
+
"grad_norm": 11.836773872375488,
|
| 1638 |
+
"learning_rate": 4.443938012762079e-08,
|
| 1639 |
+
"loss": 0.18890615463256835,
|
| 1640 |
+
"step": 9200
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 1.0,
|
| 1644 |
+
"eval_med-ir_cosine_accuracy@1": 0.309,
|
| 1645 |
+
"eval_med-ir_cosine_accuracy@10": 0.7785,
|
| 1646 |
+
"eval_med-ir_cosine_accuracy@5": 0.6665,
|
| 1647 |
+
"eval_med-ir_cosine_map@10": 0.4575869047619047,
|
| 1648 |
+
"eval_med-ir_cosine_mrr@10": 0.457586904761906,
|
| 1649 |
+
"eval_med-ir_cosine_ndcg@10": 0.5347855089896354,
|
| 1650 |
+
"eval_med-ir_cosine_precision@1": 0.309,
|
| 1651 |
+
"eval_med-ir_cosine_precision@10": 0.07784999999999999,
|
| 1652 |
+
"eval_med-ir_cosine_precision@5": 0.13329999999999997,
|
| 1653 |
+
"eval_med-ir_cosine_recall@1": 0.309,
|
| 1654 |
+
"eval_med-ir_cosine_recall@10": 0.7785,
|
| 1655 |
+
"eval_med-ir_cosine_recall@5": 0.6665,
|
| 1656 |
+
"eval_runtime": 198.2635,
|
| 1657 |
+
"eval_samples_per_second": 0.0,
|
| 1658 |
+
"eval_steps_per_second": 0.0,
|
| 1659 |
+
"step": 9238
|
| 1660 |
}
|
| 1661 |
],
|
| 1662 |
"logging_steps": 50,
|
|
|
|
| 1671 |
"should_evaluate": false,
|
| 1672 |
"should_log": false,
|
| 1673 |
"should_save": true,
|
| 1674 |
+
"should_training_stop": true
|
| 1675 |
},
|
| 1676 |
"attributes": {}
|
| 1677 |
}
|