Update app.py

app.py

@@ -1,51 +1,51 @@
-from fastapi import FastAPI, HTTPException, Depends, Request
-from transformers import AutoTokenizer, AutoModelForCausalLM
-import torch
-import os
-import huggingface_hub
-
-app = FastAPI()
-
-EXPECTED_TOKEN = os.environ.get("EXPECTED_TOKEN")
-REPO_ID = "Day23/coder-personal-use"
-MODEL_FOLDER = "model"
-
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-
-model_dir = huggingface_hub.snapshot_download(repo_id=REPO_ID, allow_patterns=["model/*"])
-
-tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained(
-    model_dir,
-    trust_remote_code=True,
-    device_map=device,
-)
-
-@app.post("/generate")
-async def generate_text(request: Request):
-    """Gera um texto com base na entrada fornecida."""
-    
-    data = await request.json()
-    user_message = data.get("message")
-    
-    if not user_message:
-        raise HTTPException(status_code=400, detail="O campo 'message' é obrigatório.")
-    
-    messages = [{'role': 'user', 'content': user_message}]
-    
-    inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(device)
-    
-    with torch.no_grad():
-        outputs = model.generate(
-            inputs,
-            max_new_tokens=512,
-            do_sample=True,
-            top_k=50,
-            top_p=0.95,
-            num_return_sequences=1,
-            eos_token_id=tokenizer.eos_token_id
-        )
-    
-    generated_text = tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=True)
-    
+from fastapi import FastAPI, HTTPException, Depends, Request
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+import os
+import huggingface_hub
+
+app = FastAPI()
+
+EXPECTED_TOKEN = os.environ.get("EXPECTED_TOKEN")
+REPO_ID = "Day23/coder-personal-use"
+MODEL_FOLDER = "model"
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+model_dir = huggingface_hub.snapshot_download(repo_id=REPO_ID, allow_patterns=["model/*"])
+
+tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    model_dir,
+    trust_remote_code=True,
+    device_map=device,
+)
+
+@app.get("/generate")
+async def generate_text(request: Request):
+    """Gera um texto com base na entrada fornecida."""
+    
+    data = await request.json()
+    user_message = data.get("message")
+    
+    if not user_message:
+        raise HTTPException(status_code=400, detail="O campo 'message' é obrigatório.")
+    
+    messages = [{'role': 'user', 'content': user_message}]
+    
+    inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(device)
+    
+    with torch.no_grad():
+        outputs = model.generate(
+            inputs,
+            max_new_tokens=512,
+            do_sample=True,
+            top_k=50,
+            top_p=0.95,
+            num_return_sequences=1,
+            eos_token_id=tokenizer.eos_token_id
+        )
+    
+    generated_text = tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=True)
+    
     return {"response": generated_text}