Fix Cell 11: bypass WhisperProcessor entirely to avoid torchvision crash

WhisperProcessor -> processing_utils -> image_utils -> torchvision fails
when torch CUDA 13.0 and torchvision CUDA 12.8 are mismatched in memory
(reinstalling torchvision in Cell 2 fixes disk but not the live process).

Build processor from WhisperFeatureExtractor + WhisperTokenizer directly
-- neither touches image_utils or torchvision. A minimal _Processor class
exposes feature_extractor, tokenizer, get_decoder_prompt_ids, and
save_pretrained, which is all the rest of the notebook needs.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

notebooks/kaggle_master_trainer.ipynb

@@ -111,7 +111,7 @@
    "id": "cell-prepare",
    "metadata": {},
    "outputs": [],
-   "source": "# -- Cell 11: Whisper processor + prepare_dataset -----------------------------\n# Use the direct module path to avoid the top-level transformers lazy loader\n# which triggers torchvision -> torch._dynamo and breaks on Kaggle Python 3.12.\nimport numpy as np\n\ntry:\n    from transformers.models.whisper.processing_whisper import WhisperProcessor\nexcept ImportError:\n    from transformers import WhisperProcessor\n\nprint(f'Loading processor: {WHISPER_MODEL_ID} ...')\nprocessor = WhisperProcessor.from_pretrained(WHISPER_MODEL_ID, token=HF_TOKEN)\nprint('Processor ready')\n\n\ndef prepare_dataset(batch, text_col='transcription', lang=TRAIN_LANG):\n    \"\"\"\n    Unified feature-extraction for all data sources.\n    Resamples audio to 16 kHz, cleans text, returns Whisper input_features + labels.\n    \"\"\"\n    audio        = batch['audio']\n    audio_array  = np.array(audio['array'], dtype=np.float32)\n    orig_sr      = audio['sampling_rate']\n\n    if orig_sr != TARGET_SR:\n        try:\n            import torchaudio.functional as F_audio\n            import torch\n            audio_array = F_audio.resample(\n                torch.from_numpy(audio_array).unsqueeze(0),\n                orig_sr, TARGET_SR,\n            ).squeeze(0).numpy()\n        except Exception:\n            import librosa\n            audio_array = librosa.resample(audio_array, orig_sr=orig_sr, target_sr=TARGET_SR)\n\n    batch['input_features'] = processor.feature_extractor(\n        audio_array, sampling_rate=TARGET_SR\n    ).input_features[0]\n\n    raw_text       = batch.get(text_col, '') or ''\n    cleaned        = clean_text(str(raw_text), lang=lang)\n    batch['labels'] = processor.tokenizer(cleaned).input_ids\n    return batch\n\n\nprint('prepare_dataset ready')"
+   "source": "# -- Cell 11: Whisper processor + prepare_dataset -----------------------------\n# WhisperProcessor imports processing_utils -> image_utils -> torchvision,\n# which crashes when torch/torchvision have mismatched CUDA versions.\n# Fix: build the processor manually from its two sub-components.\n# WhisperFeatureExtractor and WhisperTokenizer have no torchvision dependency.\nimport numpy as np\n\nfrom transformers.models.whisper.feature_extraction_whisper import WhisperFeatureExtractor\nfrom transformers.models.whisper.tokenization_whisper import WhisperTokenizer\n\nprint(f'Loading Whisper feature extractor + tokenizer: {WHISPER_MODEL_ID} ...')\n_feat_ext  = WhisperFeatureExtractor.from_pretrained(WHISPER_MODEL_ID, token=HF_TOKEN)\n_tokenizer = WhisperTokenizer.from_pretrained(WHISPER_MODEL_ID, token=HF_TOKEN)\n\n\nclass _Processor:\n    \"\"\"Minimal WhisperProcessor substitute that avoids the torchvision import chain.\"\"\"\n    def __init__(self, feature_extractor, tokenizer):\n        self.feature_extractor = feature_extractor\n        self.tokenizer         = tokenizer\n\n    def get_decoder_prompt_ids(self, language, task='transcribe'):\n        return self.tokenizer.get_decoder_prompt_ids(language=language, task=task)\n\n    def save_pretrained(self, path):\n        self.feature_extractor.save_pretrained(path)\n        self.tokenizer.save_pretrained(path)\n\n\nprocessor = _Processor(_feat_ext, _tokenizer)\nprint('Processor ready')\n\n\ndef prepare_dataset(batch, text_col='transcription', lang=TRAIN_LANG):\n    \"\"\"\n    Resample to 16 kHz, extract log-mel features, tokenise text.\n    Works on any dict with 'audio' (HF Audio column) and a text column.\n    \"\"\"\n    audio       = batch['audio']\n    audio_array = np.array(audio['array'], dtype=np.float32)\n    orig_sr     = audio['sampling_rate']\n\n    if orig_sr != TARGET_SR:\n        try:\n            import torchaudio.functional as F_audio, torch\n            audio_array = F_audio.resample(\n                torch.from_numpy(audio_array).unsqueeze(0),\n                orig_sr, TARGET_SR,\n            ).squeeze(0).numpy()\n        except Exception:\n            import librosa\n            audio_array = librosa.resample(audio_array, orig_sr=orig_sr, target_sr=TARGET_SR)\n\n    batch['input_features'] = processor.feature_extractor(\n        audio_array, sampling_rate=TARGET_SR\n    ).input_features[0]\n\n    raw_text        = batch.get(text_col, '') or ''\n    cleaned         = clean_text(str(raw_text), lang=lang)\n    batch['labels'] = processor.tokenizer(cleaned).input_ids\n    return batch\n\n\nprint('prepare_dataset ready')"
   },
   {
    "cell_type": "code",