Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -18,9 +18,29 @@ model = AutoModelForCausalLM.from_pretrained(
|
|
| 18 |
device_map="auto"
|
| 19 |
)
|
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
class InputText(BaseModel):
|
| 22 |
text: str
|
| 23 |
|
|
|
|
| 24 |
@app.post("/generate")
|
| 25 |
async def generate_text(data: InputText):
|
| 26 |
inputs = tokenizer(data.text, return_tensors="pt").to(model.device)
|
|
|
|
| 18 |
device_map="auto"
|
| 19 |
)
|
| 20 |
|
| 21 |
+
# ---------- ROUTES -------------
|
| 22 |
+
|
| 23 |
+
@app.get("/")
|
| 24 |
+
async def root():
|
| 25 |
+
return {
|
| 26 |
+
"message": "🚀 FastAPI MPT Model Running on Hugging Face Spaces",
|
| 27 |
+
"endpoints": ["/", "/status", "/generate"]
|
| 28 |
+
}
|
| 29 |
+
|
| 30 |
+
@app.get("/status")
|
| 31 |
+
async def status():
|
| 32 |
+
return {
|
| 33 |
+
"status": "ok",
|
| 34 |
+
"model": MODEL_REPO,
|
| 35 |
+
"device": str(model.device),
|
| 36 |
+
"torch_dtype": str(model.dtype)
|
| 37 |
+
}
|
| 38 |
+
|
| 39 |
+
|
| 40 |
class InputText(BaseModel):
|
| 41 |
text: str
|
| 42 |
|
| 43 |
+
|
| 44 |
@app.post("/generate")
|
| 45 |
async def generate_text(data: InputText):
|
| 46 |
inputs = tokenizer(data.text, return_tensors="pt").to(model.device)
|