ft-vit-base-patch16-224-in21k-on-food101-lora

README.md

@@ -20,8 +20,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the food101 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2032
-- Accuracy: 0.938
+- Loss: 2.4845
+- Accuracy: 1.0
 
 ## Model description
 
@@ -55,8 +55,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 2.2286        | 0.96  | 17   | 0.2821          | 0.914    |
-| 0.2266        | 1.92  | 34   | 0.2032          | 0.938    |
+| No log        | 1.0   | 2    | 2.4845          | 1.0      |
+| No log        | 2.0   | 4    | 1.0485          | 1.0      |
 
 
 ### Framework versions

adapter_config.json

@@ -12,7 +12,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
+  "lora_alpha": 32,
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
@@ -25,6 +25,8 @@
   "revision": null,
   "target_modules": [
     "query",
+    "key",
+    "dense",
     "value"
   ],
   "task_type": null

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec29b1a584af9d87b709977625cfaeff56d132b7ac08db4bad87810146e333a9
-size 2677140
+oid sha256:7fd8401a192a4ddfa9fc56309ed5da3d251fef2cbc1de90274e43c0508a0dcb8
+size 10948068

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95907d32c4832fa8afb419253f987f866aa5c007ed34e06b200a92b458275153
+oid sha256:1ec7f67e55cc437105cadbc2322de5b0e6ca66d4f235fcc0c6577d215da8745a
 size 4411