mmrech/florence2-pitvqa-finetuned

README.md

@@ -19,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/Florence-2-base](https://huggingface.co/microsoft/Florence-2-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.5691
+- Loss: 1.5346
 
 ## Model description
 
@@ -54,11 +54,11 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 1.758         | 25.0  | 100  | 1.7403          |
-| 1.1935        | 50.0  | 200  | 1.4054          |
-| 1.1311        | 75.0  | 300  | 1.5244          |
-| 1.0502        | 100.0 | 400  | 1.6417          |
-| 1.0318        | 125.0 | 500  | 1.5691          |
+| 1.7452        | 25.0  | 100  | 1.7258          |
+| 1.1957        | 50.0  | 200  | 1.4125          |
+| 1.1343        | 75.0  | 300  | 1.4571          |
+| 1.0557        | 100.0 | 400  | 1.5572          |
+| 1.0336        | 125.0 | 500  | 1.5346          |
 
 
 ### Framework versions

adapter_config.json

@@ -29,12 +29,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
-    "k_proj",
     "fc1",
-    "o_proj",
     "q_proj",
-    "fc2"
+    "o_proj",
+    "v_proj",
+    "fc2",
+    "k_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a1fc76c93208e5a01b11ffea0308afee7a854318ed5e43d9b5257407f1584010
+oid sha256:cceaae29d857c72ae6d9289c765c30de9bfa49412b39aa7cf3ebc6d221b70430
 size 9491464

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59f11c502f2482f3ff296a91e2026419affca175b1795d6aad7d16216fb4ab23
+oid sha256:62bb70ff497df21bc2f1ec207021af1265d488bff9d9e1a680e9d721127965ce
 size 5841