Merge pull request #36 from danilyef/pr3_fashapi

PR3: Write a FastAPI server for your model, with tests and CI integration.

homework_9/pr1/README.md

@@ -1,7 +1,7 @@
 ## Run
 ```bash
 pip install -r requirements.txt
-streamlit run project/main.py
+streamlit run main.py
 ```
 
 ## Tests

homework_9/pr2/README.md

@@ -1,7 +1,7 @@
 ## Run
 ```bash
 pip install -r requirements.txt
-python project/main.py
+python main.py
 ```
 
 ## Tests

homework_9/pr3/README.md

@@ -0,0 +1,8 @@
+## Run
+
+```bash
+cd homework_9
+uvicorn pr3.app:app --reload
+```
+
+## Tests

homework_9/pr3/app.py

@@ -0,0 +1,42 @@
+from fastapi import FastAPI
+from pydantic import BaseModel
+from .utils import Model
+
+model = Model(model_name='distilbert-base-uncased-finetuned-sst-2-english')
+
+app = FastAPI()
+
+
+class SentimentRequest(BaseModel):
+    text: str
+
+class SentimentResponse(BaseModel):
+    text: str
+    sentiment: str
+    probability: float
+
+class ProbabilityResponse(BaseModel):
+    text: str 
+    probability: float
+
+@app.get("/")
+def read_root():
+    return {"message": "Welcome to the sentiment analysis API"}
+
+@app.post("/predict")
+def predict_sentiment(request: SentimentRequest) -> SentimentResponse:
+    label = model.predict(request.text)
+    probability = model.predict_proba(request.text)
+    return SentimentResponse(
+        text=request.text,
+        sentiment=label,
+        probability=float(probability)
+    )
+
+@app.post("/probability")
+def get_probability(request: SentimentRequest) -> ProbabilityResponse:
+    probability = model.predict_proba(request.text)
+    return ProbabilityResponse(
+        text=request.text,
+        probability=float(probability)
+    )

homework_9/pr3/utils.py

@@ -0,0 +1,26 @@
+from transformers import DistilBertTokenizer, DistilBertForSequenceClassification
+import torch
+
+class Model:
+    def __init__(self, model_name="distilbert-base-uncased-finetuned-sst-2-english"):
+        self.tokenizer = DistilBertTokenizer.from_pretrained(model_name)
+        self.model = DistilBertForSequenceClassification.from_pretrained(model_name)
+        self.model.eval()
+
+    def predict(self, text):
+        inputs = self.tokenizer(
+            text, return_tensors="pt", truncation=True, padding=True
+        )
+        with torch.no_grad():
+            outputs = self.model(**inputs)
+        predicted_class_id = torch.argmax(outputs.logits, dim=1).item()
+        return self.model.config.id2label[predicted_class_id]
+
+    def predict_proba(self, text):
+        inputs = self.tokenizer(
+            text, return_tensors="pt", truncation=True, padding=True
+        )
+        with torch.no_grad():
+            outputs = self.model(**inputs)
+        probabilities = torch.softmax(outputs.logits, dim=1)
+        return probabilities.squeeze().max().item()

homework_9/requirements.txt

@@ -4,4 +4,8 @@ pandas==2.2.1
 torch==2.2.1
 streamlit==1.39.0
 pytest==8.3.0
-gradio==4.44.1
+gradio==4.44.1
+fastapi[standard]==0.114.0
+pydantic==2.8.2
+uvicorn==0.30.5
+requests==2.32.2

homework_9/tests/fastapi_tests/test_fastapi.py

@@ -0,0 +1,67 @@
+from fastapi.testclient import TestClient
+from pr3.app import app
+
+client = TestClient(app)
+
+def test_read_root():
+    response = client.get("/")
+    assert response.status_code == 200
+    assert response.json() == {"message": "Welcome to the sentiment analysis API"}
+
+def test_predict_sentiment():
+    # Test with positive text
+    response = client.post(
+        "/predict",
+        json={"text": "I love this movie!"}
+    )
+    assert response.status_code == 200
+    data = response.json()
+    assert "text" in data
+    assert "sentiment" in data
+    assert "probability" in data
+    assert data["text"] == "I love this movie!"
+    assert isinstance(data["sentiment"], str)
+    assert isinstance(data["probability"], float)
+    assert 0 <= data["probability"] <= 1
+
+    # Test with negative text
+    response = client.post(
+        "/predict",
+        json={"text": "I hate this movie!"}
+    )
+    assert response.status_code == 200
+    data = response.json()
+    assert "text" in data
+    assert "sentiment" in data
+    assert "probability" in data
+    assert data["text"] == "I hate this movie!"
+    assert isinstance(data["sentiment"], str)
+    assert isinstance(data["probability"], float)
+    assert 0 <= data["probability"] <= 1
+
+def test_get_probability():
+    response = client.post(
+        "/probability",
+        json={"text": "This is a test message"}
+    )
+    assert response.status_code == 200
+    data = response.json()
+    assert "text" in data
+    assert "probability" in data
+    assert data["text"] == "This is a test message"
+    assert isinstance(data["probability"], float)
+    assert 0 <= data["probability"] <= 1
+
+def test_invalid_request():
+    # Test missing text field
+    response = client.post(
+        "/predict",
+        json={}
+    )
+    assert response.status_code == 422
+
+    response = client.post(
+        "/probability",
+        json={}
+    )
+    assert response.status_code == 422

homework_9/tests/gradio_func/test_gradio.py

@@ -1,5 +1,5 @@
 import pytest
-from pr2.func_gradio.utils import Model
+from pr2.utils import Model
 
 @pytest.fixture
 def model():
@@ -36,4 +36,7 @@ def test_model_consistency(model):
     # Test multiple times to ensure consistency
     for _ in range(3):
         assert model.predict(text) == first_prediction
-        assert abs(model.predict_proba(text) - first_probability) < 1e-6
+        assert abs(model.predict_proba(text) - first_probability) < 1e-6
+
+
+

homework_9/tests/streamlit_func/test_streamlit.py

@@ -1,5 +1,5 @@
 import pytest
-from pr1.func_st.utils import Model
+from pr1.utils import Model
 
 @pytest.fixture
 def model():