End of training

README.md

@@ -6,8 +6,8 @@ license: apache-2.0
 base_model: openai/whisper-small
 tags:
 - generated_from_trainer
-datasets:
-- ihanif/common_voice_ps_20_0
+metrics:
+- wer
 model-index:
 - name: Whisper small Ps - ZFA
   results: []
@@ -18,9 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Whisper small Ps - ZFA
 
-This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 20.0 dataset.
+This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Sir Shibli dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7437
+- Loss: 0.7353
+- Wer: 26.1787
 
 ## Model description
 
@@ -39,27 +40,32 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-05
-- train_batch_size: 4
-- eval_batch_size: 4
+- learning_rate: 2e-05
+- train_batch_size: 1
+- eval_batch_size: 1
 - seed: 42
-- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 8
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 3.0
+- lr_scheduler_warmup_steps: 200
+- training_steps: 1500
+- mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss |
-|:-------------:|:------:|:----:|:---------------:|
-| 0.2314        | 0.7418 | 500  | 0.8094          |
-| 0.1647        | 1.4837 | 1000 | 0.7804          |
-| 0.0946        | 2.2255 | 1500 | 0.7939          |
-| 0.0309        | 2.9674 | 2000 | 0.7437          |
+| Training Loss | Epoch   | Step | Validation Loss | Wer     |
+|:-------------:|:-------:|:----:|:---------------:|:-------:|
+| No log        | 2.8341  | 300  | 0.5501          | 32.4650 |
+| 5.4773        | 5.6635  | 600  | 0.5894          | 28.7836 |
+| 5.4773        | 8.4929  | 900  | 0.6551          | 27.0398 |
+| 0.1933        | 11.3223 | 1200 | 0.7140          | 26.7600 |
+| 0.0125        | 14.1517 | 1500 | 0.7353          | 26.1787 |
 
 
 ### Framework versions
 
-- Transformers 4.56.2
-- Pytorch 2.7.0+cu126
-- Datasets 4.1.1
-- Tokenizers 0.22.0
+- Transformers 5.0.0
+- Pytorch 2.9.0+cu126
+- Datasets 4.0.0
+- Tokenizers 0.22.2

config.json

@@ -6,7 +6,6 @@
     "WhisperForConditionalGeneration"
   ],
   "attention_dropout": 0.0,
-  "begin_suppress_tokens": null,
   "bos_token_id": 50257,
   "classifier_proj_size": 256,
   "d_model": 768,
@@ -22,20 +21,8 @@
   "encoder_layerdrop": 0.0,
   "encoder_layers": 12,
   "eos_token_id": 50257,
-  "forced_decoder_ids": [
-    [
-      1,
-      50259
-    ],
-    [
-      2,
-      50359
-    ],
-    [
-      3,
-      50363
-    ]
-  ],
+  "forced_decoder_ids": null,
+  "gradient_checkpointing": false,
   "init_std": 0.02,
   "is_encoder_decoder": true,
   "mask_feature_length": 10,
@@ -44,7 +31,6 @@
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
   "mask_time_prob": 0.05,
-  "max_length": null,
   "max_source_positions": 1500,
   "max_target_positions": 448,
   "median_filter_width": 7,
@@ -53,8 +39,10 @@
   "num_mel_bins": 80,
   "pad_token_id": 50257,
   "scale_embedding": false,
-  "transformers_version": "4.56.2",
-  "use_cache": true,
+  "suppress_tokens": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.0.0",
+  "use_cache": false,
   "use_weighted_layer_sum": false,
   "vocab_size": 51865
 }

generation_config.json

@@ -41,16 +41,38 @@
       5
     ]
   ],
+  "assistant_confidence_threshold": 0.4,
+  "assistant_lookbehind": 10,
   "begin_suppress_tokens": [
     220,
     50257
   ],
   "bos_token_id": 50257,
   "decoder_start_token_id": 50258,
+  "diversity_penalty": 0.0,
+  "do_sample": false,
+  "early_stopping": false,
+  "encoder_no_repeat_ngram_size": 0,
+  "encoder_repetition_penalty": 1.0,
   "eos_token_id": [
     50257
   ],
-  "forced_decoder_ids": null,
+  "epsilon_cutoff": 0.0,
+  "eta_cutoff": 0.0,
+  "forced_decoder_ids": [
+    [
+      1,
+      50340
+    ],
+    [
+      2,
+      50359
+    ],
+    [
+      3,
+      50363
+    ]
+  ],
   "is_multilingual": true,
   "lang_to_id": {
     "<|af|>": 50327,
@@ -153,12 +175,23 @@
     "<|yo|>": 50325,
     "<|zh|>": 50260
   },
-  "language": "pashto",
+  "length_penalty": 1.0,
   "max_initial_timestamp_index": 50,
   "max_length": 448,
+  "min_length": 0,
+  "no_repeat_ngram_size": 0,
   "no_timestamps_token_id": 50363,
+  "num_assistant_tokens": 20,
+  "num_assistant_tokens_schedule": "constant",
+  "num_beam_groups": 1,
+  "num_beams": 1,
+  "num_return_sequences": 1,
+  "output_scores": false,
   "pad_token_id": 50257,
   "prev_sot_token_id": 50361,
+  "remove_invalid_values": false,
+  "repetition_penalty": 1.0,
+  "return_dict_in_generate": false,
   "return_timestamps": false,
   "suppress_tokens": [
     1,
@@ -244,14 +277,21 @@
     49870,
     50254,
     50258,
+    50358,
+    50359,
     50360,
     50361,
     50362
   ],
-  "task": "transcribe",
+  "target_lookbehind": 10,
   "task_to_id": {
     "transcribe": 50359,
     "translate": 50358
   },
-  "transformers_version": "4.56.2"
+  "temperature": 1.0,
+  "top_k": 50,
+  "top_p": 1.0,
+  "transformers_version": "5.0.0",
+  "typical_p": 1.0,
+  "use_cache": true
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0b8e8eb4ef66a77252bdb11f79e1283058141825c38ed47ebb1730899d20a9d
+oid sha256:4ae5fc169fe3f2580bf45eae61f9e26933d21adf4a29d3cdddf435015f054983
 size 966995080

processor_config.json

@@ -0,0 +1,17 @@
+{
+  "feature_extractor": {
+    "chunk_length": 30,
+    "dither": 0.0,
+    "feature_extractor_type": "WhisperFeatureExtractor",
+    "feature_size": 80,
+    "hop_length": 160,
+    "n_fft": 400,
+    "n_samples": 480000,
+    "nb_max_frames": 3000,
+    "padding_side": "right",
+    "padding_value": 0.0,
+    "return_attention_mask": false,
+    "sampling_rate": 16000
+  },
+  "processor_class": "WhisperProcessor"
+}

runs/Feb07_07-42-34_d8ef86bfc6b2/events.out.tfevents.1770450154.d8ef86bfc6b2.1775.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b94535f93a4634db1a4da9a9a6fffdd162c4da05f1553e190a78a1b344cd707b
+size 5057

runs/Feb07_07-56-25_d8ef86bfc6b2/events.out.tfevents.1770450985.d8ef86bfc6b2.1775.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36e9aad18c3e01e6b3f65e9cdbdcaf4b7a65cd9aa5e7ae4ae6802207285ae77a
+size 5057

runs/Feb07_08-18-44_d8ef86bfc6b2/events.out.tfevents.1770452324.d8ef86bfc6b2.1775.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a60f6c382708510c57dcfc6f865001a981e345d3cdfcb13131d27d5a291794e
+size 5906

runs/Feb07_09-06-10_d8ef86bfc6b2/events.out.tfevents.1770455170.d8ef86bfc6b2.27087.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:673e4fa6a08337462b4294b0d95d6c1f27b88ef7d0b1b27a8fef7ceff010e6ba
+size 7636

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0a5392fbd6d7d272e54fe9b56130db6a6cfaaec42b8e231024748589cc2e27b
-size 5905
+oid sha256:fca71909df634f830182537b67d18d35ea04811ecd33f61c71febff2b1f5886d
+size 5329