Model save

README.md

@@ -1,11 +1,9 @@
 ---
 library_name: transformers
 license: apache-2.0
-base_model: google/umt5-small
+base_model: urarik/t5-asr-CV16
 tags:
 - generated_from_trainer
-metrics:
-- wer
 model-index:
 - name: t5-asr-CV16
   results: []
@@ -16,10 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # t5-asr-CV16
 
-This model is a fine-tuned version of [google/umt5-small](https://huggingface.co/google/umt5-small) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 15.7975
-- Wer: 3.5813
+This model is a fine-tuned version of [urarik/t5-asr-CV16](https://huggingface.co/urarik/t5-asr-CV16) on an unknown dataset.
 
 ## Model description
 
@@ -38,12 +33,12 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0002
+- learning_rate: 0.001
 - train_batch_size: 64
 - eval_batch_size: 64
 - seed: 42
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 512
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 1024
 - optimizer: Use paged_lion_8bit and the args are:
 No additional optimizer arguments
 - lr_scheduler_type: cosine
@@ -51,22 +46,6 @@ No additional optimizer arguments
 - num_epochs: 10
 - mixed_precision_training: Native AMP
 
-### Training results
-
-| Training Loss | Epoch | Step | Validation Loss | Wer    |
-|:-------------:|:-----:|:----:|:---------------:|:------:|
-| 29.9836       | 1.0   | 61   | 14.7810         | 5.8702 |
-| 27.3223       | 2.0   | 122  | 15.0561         | 3.9158 |
-| 25.0282       | 3.0   | 183  | 15.6140         | 3.4144 |
-| 24.21         | 4.0   | 244  | 15.6346         | 3.5611 |
-| 22.764        | 5.0   | 305  | 15.7074         | 3.5764 |
-| 21.5313       | 6.0   | 366  | 15.7883         | 3.5908 |
-| 20.9904       | 7.0   | 427  | 15.7766         | 3.6450 |
-| 21.4041       | 8.0   | 488  | 15.7942         | 3.7102 |
-| 20.8573       | 9.0   | 549  | 15.7979         | 3.6254 |
-| 20.4032       | 10.0  | 610  | 15.7975         | 3.5813 |
-
-
 ### Framework versions
 
 - Transformers 4.48.3

runs/Feb11_06-00-43_e2bd624041e9/events.out.tfevents.1739253644.e2bd624041e9.18.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:807a04e126524f3a68af6f61a65108f719deb368266695627ec6f105caadb3a6
+size 4184

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20a46ac256746594ed7e1e3ef733b83fbc5a6f0922aa7480eda961743de080ef
-size 16837459
+oid sha256:9525bc713574a220597bd7239039b25127ff49100039ecaccb165d49e8b69036
+size 16837557

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed93c71fd96065245d83d3991e15c95c6d6209e28cc69110e3703d1e5e5d608c
+oid sha256:46c9c9f9b97ed870547437e64c3e130fcd4428fd509b6a99cc29424a3b0bb685
 size 5560