End of training

README.md

@@ -18,23 +18,23 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 0.1359
-- eval_wer: 13.9763
-- eval_avg_precision_Exact: 0.8872
-- eval_avg_recall_Exact: 0.8877
-- eval_avg_f1_Exact: 0.8870
-- eval_avg_precision_Letter_Shift: 0.9084
-- eval_avg_recall_Letter_Shift: 0.9090
-- eval_avg_f1_Letter_Shift: 0.9083
-- eval_avg_precision_Word_Level: 0.9114
-- eval_avg_recall_Word_Level: 0.9124
-- eval_avg_f1_Word_Level: 0.9115
-- eval_avg_precision_Word_Shift: 0.9614
-- eval_avg_recall_Word_Shift: 0.9640
-- eval_avg_f1_Word_Shift: 0.9622
-- eval_precision_median_exact: 0.9375
-- eval_recall_median_exact: 0.9412
-- eval_f1_median_exact: 0.9524
+- eval_loss: 0.7434
+- eval_wer: 12.1548
+- eval_avg_precision_Exact: 0.9045
+- eval_avg_recall_Exact: 0.9047
+- eval_avg_f1_Exact: 0.9042
+- eval_avg_precision_Letter_Shift: 0.9263
+- eval_avg_recall_Letter_Shift: 0.9265
+- eval_avg_f1_Letter_Shift: 0.9260
+- eval_avg_precision_Word_Level: 0.9288
+- eval_avg_recall_Word_Level: 0.9291
+- eval_avg_f1_Word_Level: 0.9285
+- eval_avg_precision_Word_Shift: 0.9744
+- eval_avg_recall_Word_Shift: 0.9757
+- eval_avg_f1_Word_Shift: 0.9746
+- eval_precision_median_exact: 1.0
+- eval_recall_median_exact: 1.0
+- eval_f1_median_exact: 1.0
 - eval_precision_max_exact: 1.0
 - eval_recall_max_exact: 1.0
 - eval_f1_max_exact: 1.0
@@ -47,14 +47,13 @@ It achieves the following results on the evaluation set:
 - eval_precision_min_Word_Level: 0.0
 - eval_recall_min_Word_Level: 0.0
 - eval_f1_min_Word_Level: 0.0
-- eval_precision_min_Word_Shift: 0.1429
-- eval_recall_min_Word_Shift: 0.1111
-- eval_f1_min_Word_Shift: 0.125
-- eval_runtime: 1569.5335
-- eval_samples_per_second: 1.715
-- eval_steps_per_second: 0.054
-- epoch: 0.75
-- step: 8000
+- eval_precision_min_Word_Shift: 0.0
+- eval_recall_min_Word_Shift: 0.0
+- eval_f1_min_Word_Shift: 0.0
+- eval_runtime: 1534.2005
+- eval_samples_per_second: 1.755
+- eval_steps_per_second: 0.055
+- step: 0
 
 ## Model description
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "openai/whisper-medium",
+  "_name_or_path": "./whisper-medium-he-teamim-allNusah-13-03-24-warmup-100-RandomFalse",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,

runs/Mar13_12-11-55_sipl-7542-ct/events.out.tfevents.1710333506.sipl-7542-ct.218115.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b3b863ca2b3fc5f0079f0d0da9231201a73deccaf8a6edd65cd8f1734b348c1
+size 2104

runs/Mar13_12-53-10_sipl-7542-ct/events.out.tfevents.1710336255.sipl-7542-ct.235753.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99fe2583c6b8d5f833616a2f3a201b3d6c3d07ff512224fb14c7870492a27964
+size 2104

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:34ff411ffb26f014712eed70f0cf9d41d5bbdde7e2c8abee75240c21b0d52f07
+oid sha256:9eaee8a9a97af593321c796d7f8e4b8b4942c2637ae56f631af638fbdaf31fca
 size 5176