richardchiu
/

work1_AIA_LLM_B_132001

Text Generation

Generated from Trainer

text-generation-inference

Inference Endpoints

8-bit precision

Model card Files Files and versions Metrics Training metrics Community

richardchiu commited on Mar 23

Commit

c3244ee

•

1 Parent(s): 5db90e8

End of training

Files changed (2) hide show

README.md +10 -10
model.safetensors +1 -1

README.md CHANGED Viewed

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 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: 1.2488
-- Matthews Correlation: 0.5406
 ## Model description
@@ -37,10 +37,10 @@ More information needed
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 16
 - eval_batch_size: 16
-- seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 5
@@ -49,11 +49,11 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
 |:-------------:|:-----:|:----:|:---------------:|:--------------------:|
-| 0.1282        | 1.0   | 535  | 0.7020          | 0.5086               |
-| 0.0923        | 2.0   | 1070 | 1.0002          | 0.5173               |
-| 0.0693        | 3.0   | 1605 | 1.0410          | 0.5195               |
-| 0.0524        | 4.0   | 2140 | 1.2027          | 0.5309               |
-| 0.042         | 5.0   | 2675 | 1.2488          | 0.5406               |
 ### Framework versions

 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: 0.9305
+- Matthews Correlation: 0.5116
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 4.6144778537845165e-05
+- train_batch_size: 32
 - eval_batch_size: 16
+- seed: 3
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 5
 | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
 |:-------------:|:-----:|:----:|:---------------:|:--------------------:|
+| No log        | 1.0   | 268  | 0.4654          | 0.4612               |
+| 0.415         | 2.0   | 536  | 0.5235          | 0.4858               |
+| 0.415         | 3.0   | 804  | 0.6812          | 0.4887               |
+| 0.1523        | 4.0   | 1072 | 0.8019          | 0.5029               |
+| 0.1523        | 5.0   | 1340 | 0.9305          | 0.5116               |
 ### Framework versions

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ccdc4bb498f954a3872115444406f355d43610eaabacd686d34bce8c8aa4a1a
 size 267832560

 version https://git-lfs.github.com/spec/v1
+oid sha256:f6924014bf59ea78b5fb2877725379b58387c390575cf05d8b2214c1e897a754
 size 267832560