End of training

README.md

@@ -14,6 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
 # codet5-small-v6
 
 This model is a fine-tuned version of [Salesforce/codet5-small](https://huggingface.co/Salesforce/codet5-small) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1977
+- Bleu Score: 0.0006
+- Gen Len: 14.9669
 
 ## Model description
 
@@ -33,14 +37,23 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 500
-- eval_batch_size: 500
+- train_batch_size: 20
+- eval_batch_size: 20
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 3
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Bleu Score | Gen Len |
+|:-------------:|:-----:|:-----:|:---------------:|:----------:|:-------:|
+| 0.1709        | 1.0   | 6738  | 0.2015          | 0.0006     | 14.9169 |
+| 0.1249        | 2.0   | 13476 | 0.1950          | 0.0006     | 14.9628 |
+| 0.1076        | 3.0   | 20214 | 0.1977          | 0.0006     | 14.9669 |
+
+
 ### Framework versions
 
 - Transformers 4.38.0.dev0

generation_config.json

@@ -1,5 +1,4 @@
 {
-  "_from_model_config": true,
   "bos_token_id": 1,
   "decoder_start_token_id": 0,
   "eos_token_id": 2,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00c338decac37bdb20abb4c5368e2ffb4517b848b8794b81f1058030ee05eacb
+oid sha256:5a49d4901270091344a16798e0f6144668700c8b0a41f357d02c7027070e12b5
 size 241990696

runs/Jan28_13-51-34_bbab5494bb78/events.out.tfevents.1706449895.bbab5494bb78.13686.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68c50599ae3fde65dca015b471f874b0e606bbfba9ea6722a278486dbefff97f
-size 12459
+oid sha256:15d921c79bd0c76b94209ec33855ef651d5d0f4d2b7b6f4f5006e3cfb652e33f
+size 13362