End of training

README.md

@@ -1,22 +1,22 @@
 ---
-base_model: openai/whisper-medium
+base_model: openai/whisper-large-v3
 library_name: peft
 license: apache-2.0
 tags:
 - generated_from_trainer
 model-index:
-- name: whisper_medium_peft
+- name: whisper-large-v3-peft
   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. -->
 
-# whisper_medium_peft
+# whisper-large-v3-peft
 
-This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the CFN Proposal Audio dataset.
+This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the CFN Proposal Audio dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3352
+- Loss: 0.2785
 
 ## Model description
 
@@ -49,9 +49,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.3982        | 1.0   | 191  | 0.3951          |
-| 0.2585        | 2.0   | 382  | 0.3351          |
-| 0.1703        | 3.0   | 573  | 0.3352          |
+| 0.3111        | 1.0   | 191  | 0.3036          |
+| 0.1985        | 2.0   | 382  | 0.2830          |
+| 0.1435        | 3.0   | 573  | 0.2785          |
 
 
 ### Framework versions

adapter_config.json

@@ -4,7 +4,7 @@
     "base_model_class": "WhisperForConditionalGeneration",
     "parent_library": "transformers.models.whisper.modeling_whisper"
   },
-  "base_model_name_or_path": "openai/whisper-medium",
+  "base_model_name_or_path": "openai/whisper-large-v3",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b88d17c1a19931b82bf1e4bcb26a24865041a6c41ba36f6fc2fa24a43242589e
-size 37789960
+oid sha256:0f28f9df5ceaf236fef735ce8b1734bff1a4c6f4cf561b6ebfa6210a31af5299
+size 62969640

preprocessor_config.json

@@ -1,7 +1,7 @@
 {
   "chunk_length": 30,
   "feature_extractor_type": "WhisperFeatureExtractor",
-  "feature_size": 80,
+  "feature_size": 128,
   "hop_length": 160,
   "n_fft": 400,
   "n_samples": 480000,

runs/Aug08_16-33-57_hal/events.out.tfevents.1723149363.hal.538048.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51d89004f168a96a5b969a963d24c5e909a0319e33f88bfca79436093d50e54b
+size 11495

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71eb4c55a569b4892d14346e90bf42eeefdf66d1583c526ea7045590d1f20832
+oid sha256:01abdfca3525e542df8c0933c04942a21e0ddcc1a907c9a68dd0e00371de3025
 size 5304