batch_size=64

README.md

@@ -15,23 +15,9 @@ probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/flan-t5-small](https://huggingface.co/google/flan-t5-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.4162
-- Epoch: 0
-
-<pre>
-  WRONG!!! It is fine-tuned on validation data
-  
-  {'eval_loss': 0.3034636080265045,
- 'eval_bleu': 21.208712916114965,
- 'eval_rouge1': 57.0,
- 'eval_rouge2': 34.71,
- 'eval_rougeL': 50.2,
- 'eval_rougeLsum': 50.22,
- 'eval_exact': 0.03128947624137596,
- 'eval_runtime': 347.407,
- 'eval_samples_per_second': 29.622,
- 'eval_steps_per_second': 0.927}
-</pre>
+- Train Loss: 0.3095
+- Validation Loss: 0.3903
+- Epoch: 3
 
 ## Model description
 
@@ -55,9 +41,12 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Train Loss | Epoch |
-|:----------:|:-----:|
-| 0.4162     | 0     |
+| Train Loss | Validation Loss | Epoch |
+|:----------:|:---------------:|:-----:|
+| 0.6099     | 0.4054          | 0     |
+| 0.3919     | 0.3899          | 1     |
+| 0.3451     | 0.3880          | 2     |
+| 0.3095     | 0.3903          | 3     |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:510e4d4d93ebb2e2a6806c29348a9875a15cacbe0c0c7efa7554e609f4409706
+oid sha256:8969cc622ef945ebc9df0f83c11e1250b64574e87d1096d5017b3a788bcb23c5
 size 439831352