Update app.py
Browse files
app.py
CHANGED
|
@@ -17,7 +17,7 @@ os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:50'
|
|
| 17 |
model_id = "TheBloke/Yi-34B-200K-Llamafied-GPTQ"
|
| 18 |
|
| 19 |
gptq_config = GPTQConfig(bits=4, exllama_config={"version": 2})
|
| 20 |
-
tokenizer = YiTokenizer.from_pretrained("./")
|
| 21 |
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype="auto", trust_remote_code=True, quantization_config=gptq_config)
|
| 22 |
|
| 23 |
def predict(message, max_new_tokens=4056, temperature=3.5, top_p=0.9, top_k=800):
|
|
|
|
| 17 |
model_id = "TheBloke/Yi-34B-200K-Llamafied-GPTQ"
|
| 18 |
|
| 19 |
gptq_config = GPTQConfig(bits=4, exllama_config={"version": 2})
|
| 20 |
+
tokenizer = YiTokenizer.from_pretrained("./")
|
| 21 |
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", torch_dtype="auto", trust_remote_code=True, quantization_config=gptq_config)
|
| 22 |
|
| 23 |
def predict(message, max_new_tokens=4056, temperature=3.5, top_p=0.9, top_k=800):
|