Model save

README.md

@@ -1,7 +1,5 @@
 ---
 library_name: transformers
-language:
-- en
 license: apache-2.0
 base_model: openai/whisper-large-v3
 tags:
@@ -9,20 +7,20 @@ tags:
 metrics:
 - wer
 model-index:
-- name: ./7264
+- name: whisper-large-v3-cit-do01-wd0-lr3e-06-FULL5
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# ./7264
+# whisper-large-v3-cit-do01-wd0-lr3e-06-FULL5
 
-This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the 7264 FULL-2024-10-24 dataset.
+This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3984
-- Wer Ortho: 22.8797
-- Wer: 15.9936
+- Loss: 0.3911
+- Wer Ortho: 22.6474
+- Wer: 15.5576
 
 ## Model description
 
@@ -58,14 +56,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Wer Ortho | Wer     |
 |:-------------:|:------:|:----:|:---------------:|:---------:|:-------:|
-| 0.6878        | 0.4893 | 200  | 0.4758          | 27.2036   | 19.6510 |
-| 0.522         | 0.9786 | 400  | 0.4338          | 25.3302   | 18.0924 |
-| 0.424         | 1.4679 | 600  | 0.4133          | 23.7006   | 16.6155 |
-| 0.41          | 1.9572 | 800  | 0.4043          | 23.0951   | 16.2346 |
-| 0.3445        | 2.4465 | 1000 | 0.4023          | 22.8756   | 16.0791 |
-| 0.342         | 2.9358 | 1200 | 0.3967          | 23.0260   | 16.0519 |
-| 0.304         | 3.4251 | 1400 | 0.3989          | 22.9731   | 16.0403 |
-| 0.3           | 3.9144 | 1600 | 0.3984          | 22.8797   | 15.9936 |
+| 0.686         | 0.4851 | 200  | 0.4602          | 26.0150   | 18.7885 |
+| 0.5255        | 0.9703 | 400  | 0.4216          | 24.3312   | 17.1358 |
+| 0.4328        | 1.4554 | 600  | 0.4028          | 23.2291   | 15.9895 |
+| 0.4064        | 1.9406 | 800  | 0.3945          | 23.2291   | 16.1897 |
+| 0.3579        | 2.4257 | 1000 | 0.3945          | 22.8195   | 15.7618 |
+| 0.3409        | 2.9109 | 1200 | 0.3894          | 22.6884   | 15.5812 |
+| 0.3131        | 3.3960 | 1400 | 0.3909          | 22.6556   | 15.6008 |
+| 0.3021        | 3.8811 | 1600 | 0.3911          | 22.6474   | 15.5576 |
 
 
 ### Framework versions

config.json

@@ -2,7 +2,7 @@
   "_name_or_path": "openai/whisper-large-v3",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
-  "apply_spec_augment": true,
+  "apply_spec_augment": false,
   "architectures": [
     "WhisperForConditionalGeneration"
   ],
@@ -16,7 +16,7 @@
   "decoder_layerdrop": 0.0,
   "decoder_layers": 32,
   "decoder_start_token_id": 50258,
-  "dropout": 0.1,
+  "dropout": 0,
   "encoder_attention_heads": 20,
   "encoder_ffn_dim": 5120,
   "encoder_layerdrop": 0.0,

runs/Oct26_09-48-42_whisper-finetune-2/events.out.tfevents.1729936151.whisper-finetune-2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a33ac1b63fb7cd33c2343818e87e963a17fd6a7562f28301722e0e29b0deda26
-size 10068
+oid sha256:9b667b3da0aff60d99dc7e349c464150cdfd76067da8af6496f442bb20991042
+size 10793