sophonai/lora-test-distilbert-base-uncased

README.md

@@ -2,6 +2,8 @@
 base_model: distilbert-base-uncased
 library_name: peft
 license: apache-2.0
+metrics:
+- accuracy
 tags:
 - generated_from_trainer
 model-index:
@@ -15,6 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 # distilbert-base-uncased-lora-text-classification
 
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.0845
+- Accuracy: {'accuracy': 0.884}
 
 ## Model description
 
@@ -39,7 +44,63 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 1
+- num_epochs: 50
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy            |
+|:-------------:|:-----:|:----:|:---------------:|:-------------------:|
+| No log        | 1.0   | 125  | 0.3046          | {'accuracy': 0.89}  |
+| No log        | 2.0   | 250  | 0.3899          | {'accuracy': 0.853} |
+| No log        | 3.0   | 375  | 0.5440          | {'accuracy': 0.844} |
+| 0.2813        | 4.0   | 500  | 0.4597          | {'accuracy': 0.867} |
+| 0.2813        | 5.0   | 625  | 0.5280          | {'accuracy': 0.88}  |
+| 0.2813        | 6.0   | 750  | 0.5096          | {'accuracy': 0.877} |
+| 0.2813        | 7.0   | 875  | 0.6435          | {'accuracy': 0.881} |
+| 0.0793        | 8.0   | 1000 | 0.6281          | {'accuracy': 0.883} |
+| 0.0793        | 9.0   | 1125 | 0.7278          | {'accuracy': 0.868} |
+| 0.0793        | 10.0  | 1250 | 0.6827          | {'accuracy': 0.876} |
+| 0.0793        | 11.0  | 1375 | 0.8269          | {'accuracy': 0.86}  |
+| 0.0502        | 12.0  | 1500 | 0.7706          | {'accuracy': 0.866} |
+| 0.0502        | 13.0  | 1625 | 0.8552          | {'accuracy': 0.867} |
+| 0.0502        | 14.0  | 1750 | 1.0037          | {'accuracy': 0.872} |
+| 0.0502        | 15.0  | 1875 | 0.8915          | {'accuracy': 0.874} |
+| 0.0237        | 16.0  | 2000 | 0.7394          | {'accuracy': 0.882} |
+| 0.0237        | 17.0  | 2125 | 0.8335          | {'accuracy': 0.878} |
+| 0.0237        | 18.0  | 2250 | 1.0328          | {'accuracy': 0.875} |
+| 0.0237        | 19.0  | 2375 | 0.8622          | {'accuracy': 0.879} |
+| 0.0232        | 20.0  | 2500 | 0.8536          | {'accuracy': 0.872} |
+| 0.0232        | 21.0  | 2625 | 0.9345          | {'accuracy': 0.874} |
+| 0.0232        | 22.0  | 2750 | 0.9473          | {'accuracy': 0.876} |
+| 0.0232        | 23.0  | 2875 | 1.0162          | {'accuracy': 0.874} |
+| 0.0121        | 24.0  | 3000 | 0.9833          | {'accuracy': 0.868} |
+| 0.0121        | 25.0  | 3125 | 0.9889          | {'accuracy': 0.878} |
+| 0.0121        | 26.0  | 3250 | 0.9712          | {'accuracy': 0.874} |
+| 0.0121        | 27.0  | 3375 | 0.9586          | {'accuracy': 0.877} |
+| 0.0098        | 28.0  | 3500 | 0.9557          | {'accuracy': 0.875} |
+| 0.0098        | 29.0  | 3625 | 1.0105          | {'accuracy': 0.873} |
+| 0.0098        | 30.0  | 3750 | 0.9052          | {'accuracy': 0.869} |
+| 0.0098        | 31.0  | 3875 | 0.9571          | {'accuracy': 0.878} |
+| 0.0093        | 32.0  | 4000 | 1.0417          | {'accuracy': 0.871} |
+| 0.0093        | 33.0  | 4125 | 1.1209          | {'accuracy': 0.871} |
+| 0.0093        | 34.0  | 4250 | 1.1851          | {'accuracy': 0.869} |
+| 0.0093        | 35.0  | 4375 | 1.4009          | {'accuracy': 0.871} |
+| 0.0048        | 36.0  | 4500 | 1.1369          | {'accuracy': 0.879} |
+| 0.0048        | 37.0  | 4625 | 1.1375          | {'accuracy': 0.879} |
+| 0.0048        | 38.0  | 4750 | 1.1177          | {'accuracy': 0.877} |
+| 0.0048        | 39.0  | 4875 | 1.1163          | {'accuracy': 0.881} |
+| 0.0004        | 40.0  | 5000 | 1.1190          | {'accuracy': 0.874} |
+| 0.0004        | 41.0  | 5125 | 1.1133          | {'accuracy': 0.879} |
+| 0.0004        | 42.0  | 5250 | 1.1240          | {'accuracy': 0.878} |
+| 0.0004        | 43.0  | 5375 | 1.0499          | {'accuracy': 0.885} |
+| 0.0023        | 44.0  | 5500 | 1.0496          | {'accuracy': 0.884} |
+| 0.0023        | 45.0  | 5625 | 1.0538          | {'accuracy': 0.884} |
+| 0.0023        | 46.0  | 5750 | 1.0807          | {'accuracy': 0.879} |
+| 0.0023        | 47.0  | 5875 | 1.0642          | {'accuracy': 0.88}  |
+| 0.0002        | 48.0  | 6000 | 1.0655          | {'accuracy': 0.879} |
+| 0.0002        | 49.0  | 6125 | 1.0755          | {'accuracy': 0.881} |
+| 0.0002        | 50.0  | 6250 | 1.0845          | {'accuracy': 0.884} |
+
 
 ### Framework versions
 

adapter_config.json

@@ -19,7 +19,7 @@
     "score"
   ],
   "peft_type": "LORA",
-  "r": 4,
+  "r": 1024,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6eb7e6bd9ef6adf3e7c2e302f4c5cb9247abd5f129b44aac3e1196372c39d222
-size 2518104
+oid sha256:1be07c69b61d4b8e9c4dcf727a1fdc663f293ded1f0ec91730474dea43d84e21
+size 40119480

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b248f83145c01feab35b2ac49cda366c7ac2c06a3e427a8d672263f8d7bd2cc
+oid sha256:48d608d87a4a289efccf3cc694efa37c422e7bdf9dc08320ccff5b567902df4c
 size 5176