Training in progress, step 28

README.md

@@ -25,17 +25,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the ntnu-smil/sandi2025-ds dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 1.0717
-- eval_wer: 79.1953
-- eval_cer: 220.4455
-- eval_decode_runtime: 259.7244
-- eval_wer_runtime: 0.2083
-- eval_cer_runtime: 0.4860
-- eval_runtime: 508.4684
-- eval_samples_per_second: 6.39
-- eval_steps_per_second: 0.201
-- epoch: 6.1071
-- step: 21
+- eval_loss: 1.0370
+- eval_wer: 78.0238
+- eval_cer: 215.7449
+- eval_decode_runtime: 252.2954
+- eval_wer_runtime: 0.1988
+- eval_cer_runtime: 0.4668
+- eval_runtime: 500.0374
+- eval_samples_per_second: 6.498
+- eval_steps_per_second: 0.204
+- epoch: 9.0357
+- step: 28
 
 ## Model description
 

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74eedd2c1a6e38ad270f6a20916295f986257c8d75adcb7b73949e6708f609cd
+oid sha256:1fecb3f265fbe855c99926ba441dfdae31227cfbbafc16f0cbf5f9352d139427
 size 230831688

runs/May04_10-33-22_b84328dfdf7a/events.out.tfevents.1746355042.b84328dfdf7a.38805.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5bb33151364851bb412d7f700b7ebadfa3da91a10d016a8445d4fd103c0bba1
-size 9231826
+oid sha256:a39ff60c1fbfe6dd0d136d761a750fd78f8af10820e922e24da80b7e3c8d74df
+size 12310548