Update app.py

app.py

@@ -1,62 +1,12 @@
-import os 
-import sys 
-os.system("pip install transformers==4.27.0")
-os.system("pip install numpy==1.23")
-from transformers import pipeline, WhisperModel, WhisperTokenizer, WhisperFeatureExtractor, AutoFeatureExtractor, AutoProcessor, WhisperConfig, WhisperProcessor, WhisperForConditionalGeneration
-os.system("pip install jiwer")
-from jiwer import wer
-os.system("pip install datasets[audio]")
-from evaluate import evaluator, load
-from transformers import AutoModelForSequenceClassification, pipeline, BertTokenizer, AutoTokenizer, GPT2Model
-from datasets import load_dataset, Audio, disable_caching, set_caching_enabled
-import gradio as gr
-import torch
-from datasets import load_dataset
-from transformers import WhisperForConditionalGeneration, WhisperProcessor
+# TEST MODEL
 
+from transformers import pipeline
 
-processor = WhisperProcessor.from_pretrained("mskov/whisper-small-esc50")
-model = WhisperForConditionalGeneration.from_pretrained("mskov/whisper-small-esc50").to("cuda")
+repository_id="mskov/roberta-base-toxicity"
+classifier = pipeline('text-classification',repository_id)
 
-def map_to_pred(batch):
-    audio = batch["audio"]
-    input_features = processor(audio["array"], sampling_rate=16000, return_tensors="pt").input_features
-    batch["reference"] = processor.tokenizer._normalize(batch['category'])
+text = "Kederis proclaims innocence Olympic champion Kostas Kederis today left hospital ahead of his date with IOC inquisitors claiming his innocence and vowing: quot;After the crucifixion comes the resurrection. quot; .."
+result = classifier(text)
 
-    with torch.no_grad():
-        predicted_ids = model.generate(input_features.to("cuda"))[0]
-    transcription = processor.decode(predicted_ids)
-    batch["prediction"] = processor.tokenizer._normalize(transcription)
-    print(batch["prediction"])
-    return batch
-
-result = librispeech_test_clean.map(map_to_pred)
-
-wer = load("wer")
-print(100 * wer.compute(references=result["reference"], predictions=result["prediction"]))
-
-
-    
-def transcribe(audio):
-    text = pipe(audio)["text"]
-    return text, test
-
-iface = gr.Interface(
-    fn=transcribe, 
-    inputs=gr.Audio(source="microphone", type="filepath"), 
-    outputs="text",
-    title="Whisper Small ESC50 Test",
-)
-
-iface.launch()
-
-
-'''
-print("check check")
-print(inputs)
-input_features = inputs.input_features
-decoder_input_ids = torch.tensor([[1, 1]]) * model.config.decoder_start_token_id
-last_hidden_state = model(input_features, decoder_input_ids=decoder_input_ids).last_hidden_state
-list(last_hidden_state.shape)
-print(list(last_hidden_state.shape))
-'''
+predicted_label = result[0]["label"]
+print(f"Predicted label: {predicted_label}")