Create app.py

app.py

@@ -0,0 +1,58 @@
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+from transformers import pipeline, AutoTokenizer, AutoModelForSequenceClassification
+import torch
+
+app = FastAPI()
+
+# Model configuration
+MODEL_NAME = "nlptown/bert-base-multilingual-uncased-sentiment"
+DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
+
+# Initialize model and tokenizer
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+model = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME)
+classifier = pipeline("sentiment-analysis", model=model, tokenizer=tokenizer, device=DEVICE)
+
+class TextInput(BaseModel):
+    text: str
+
+@app.post("/analyze-sentiment")
+async def analyze_sentiment(input_data: TextInput):
+    try:
+        result = classifier(input_data.text)
+        return {
+            "sentiment": result[0]['label'],
+            "score": float(result[0]['score'])
+        }
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+# Przykład dla większego modelu (np. GPT-2)
+MODEL_NAME_LARGE = "gpt2-large"
+tokenizer_large = AutoTokenizer.from_pretrained(MODEL_NAME_LARGE)
+model_large = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME_LARGE)
+
+class GenerationInput(BaseModel):
+    prompt: str
+    max_length: int = 100
+
+@app.post("/generate-text")
+async def generate_text(input_data: GenerationInput):
+    try:
+        inputs = tokenizer_large(input_data.prompt, return_tensors="pt")
+        outputs = model_large.generate(
+            inputs["input_ids"],
+            max_length=input_data.max_length,
+            num_return_sequences=1,
+            no_repeat_ngram_size=2
+        )
+        generated_text = tokenizer_large.decode(outputs[0], skip_special_tokens=True)
+        return {"generated_text": generated_text}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+# Dodanie podstawowego health checka
+@app.get("/health")
+async def health_check():
+    return {"status": "healthy"}