A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 06:45 UTC
Fix model docstrings (#1764)
Woosuk Kwon committed
Nov 23, 2023 at 23:04 UTC
7c600440f7560348e571f021f2b2d1469de5264d
14 files changed
+14
-70
vllm/model_executor/models/aquila.py
+1
-5
@@ -20,11 +20,7 @@
20
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
21
# See the License for the specific language governing permissions and
22
# limitations under the License.
23
-"""Inference-only LLaMA model compatible with HuggingFace weights.
24
-
25
-The input of the model is flattened to a 1D tensor of tokens. The model uses
26
-InputMetadata to extract the original 2D shape of the input.
27
-"""
23
+"""Inference-only LLaMA model compatible with HuggingFace weights."""
24
from typing import Any, Dict, List, Optional, Tuple
25
26
import torch
vllm/model_executor/models/baichuan.py
+1
-5
@@ -17,11 +17,7 @@
17
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
18
# See the License for the specific language governing permissions and
19
# limitations under the License.
20
-"""Inference-only BaiChuan model compatible with HuggingFace weights.
21
-
22
-The input of the model is flattened to a 1D tensor of tokens. The model uses
23
-InputMetadata to extract the original 2D shape of the input.
24
-"""
20
+"""Inference-only BaiChuan model compatible with HuggingFace weights."""
21
import math
22
from typing import List, Optional, Tuple
23
vllm/model_executor/models/bloom.py
+1
-5
@@ -15,11 +15,7 @@
15
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16
# See the License for the specific language governing permissions and
17
# limitations under the License.
18
-"""Inference-only BLOOM model compatible with HuggingFace weights.
19
-
20
-The input of the model is flattened to a 1D tensor of tokens. The model uses
21
-InputMetadata to extract the original 2D shape of the input.
22
-"""
18
+"""Inference-only BLOOM model compatible with HuggingFace weights."""
19
import math
20
from typing import List, Optional, Tuple
21
vllm/model_executor/models/chatglm.py
+1
-5
@@ -1,11 +1,7 @@
1
# coding=utf-8
2
# Adapted from
3
# https://github.com/THUDM/ChatGLM2-6B
4
-"""Inference-only ChatGLM model compatible with THUDM weights.
5
-
6
-The input of the model is flattened to a 1D tensor of tokens. The model uses
7
-InputMetadata to extract the original 2D shape of the input.
8
-"""
4
+"""Inference-only ChatGLM model compatible with THUDM weights."""
5
from typing import List, Optional, Tuple
6
7
import torch
vllm/model_executor/models/gpt2.py
+1
-5
@@ -16,11 +16,7 @@
16
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
17
# See the License for the specific language governing permissions and
18
# limitations under the License.
19
-"""Inference-only GPT-2 model compatible with HuggingFace weights.
20
-
21
-The input of the model is flattened to a 1D tensor of tokens. The model uses
22
-InputMetadata to extract the original 2D shape of the input.
23
-"""
19
+"""Inference-only GPT-2 model compatible with HuggingFace weights."""
20
from typing import List, Optional, Tuple
21
22
import torch
vllm/model_executor/models/gpt_bigcode.py
+1
-5
@@ -17,11 +17,7 @@
17
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
18
# See the License for the specific language governing permissions and
19
# limitations under the License.
20
-"""Inference-only GPTBigCode model compatible with HuggingFace weights.
21
-
22
-The input of the model is flattened to a 1D tensor of tokens. The model uses
23
-InputMetadata to extract the original 2D shape of the input.
24
-"""
20
+"""Inference-only GPTBigCode model compatible with HuggingFace weights."""
21
from typing import List, Optional, Tuple
22
23
import torch
vllm/model_executor/models/gpt_j.py
+1
-5
@@ -15,11 +15,7 @@
15
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16
# See the License for the specific language governing permissions and
17
# limitations under the License.
18
-"""Inference-only GPT-J model compatible with HuggingFace weights.
19
-
20
-The input of the model is flattened to a 1D tensor of tokens. The model uses
21
-InputMetadata to extract the original 2D shape of the input.
22
-"""
18
+"""Inference-only GPT-J model compatible with HuggingFace weights."""
19
from typing import List, Optional, Tuple
20
21
import torch
vllm/model_executor/models/gpt_neox.py
+1
-5
@@ -15,11 +15,7 @@
15
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16
# See the License for the specific language governing permissions and
17
# limitations under the License.
18
-"""Inference-only GPT-NeoX model compatible with HuggingFace weights.
19
-
20
-The input of the model is flattened to a 1D tensor of tokens. The model uses
21
-InputMetadata to extract the original 2D shape of the input.
22
-"""
18
+"""Inference-only GPT-NeoX model compatible with HuggingFace weights."""
19
from typing import List, Optional, Tuple
20
21
import torch
vllm/model_executor/models/llama.py
+1
-5
@@ -20,11 +20,7 @@
20
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
21
# See the License for the specific language governing permissions and
22
# limitations under the License.
23
-"""Inference-only LLaMA model compatible with HuggingFace weights.
24
-
25
-The input of the model is flattened to a 1D tensor of tokens. The model uses
26
-InputMetadata to extract the original 2D shape of the input.
27
-"""
23
+"""Inference-only LLaMA model compatible with HuggingFace weights."""
24
from typing import Any, Dict, List, Optional, Tuple
25
26
import torch
vllm/model_executor/models/mistral.py
+1
-5
@@ -20,11 +20,7 @@
20
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
21
# See the License for the specific language governing permissions and
22
# limitations under the License.
23
-"""Inference-only Mistral model compatible with HuggingFace weights.
24
-
25
-The input of the model is flattened to a 1D tensor of tokens. The model uses
26
-InputMetadata to extract the original 2D shape of the input.
27
-"""
23
+"""Inference-only Mistral model compatible with HuggingFace weights."""
24
from typing import List, Optional, Tuple
25
26
import torch
vllm/model_executor/models/opt.py
+1
-5
@@ -16,11 +16,7 @@
16
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
17
# See the License for the specific language governing permissions and
18
# limitations under the License.
19
-"""Inference-only OPT model compatible with HuggingFace weights.
20
-
21
-The input of the model is flattened to a 1D tensor of tokens. The model uses
22
-InputMetadata to extract the original 2D shape of the input.
23
-"""
19
+"""Inference-only OPT model compatible with HuggingFace weights."""
20
from typing import List, Optional, Tuple
21
22
import torch
vllm/model_executor/models/phi_1_5.py
+1
-5
@@ -34,11 +34,7 @@
34
# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
35
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
36
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
37
-"""Inference-only Phi-1.5 model compatible with HuggingFace weights.
38
-
39
-The input of the model is flattened to a 1D tensor of tokens. The model uses
40
-InputMetadata to extract the original 2D shape of the input.
41
-"""
37
+"""Inference-only Phi-1.5 model compatible with HuggingFace weights."""
38
from typing import List, Optional, Tuple
39
40
import torch
vllm/model_executor/models/qwen.py
+1
-5
@@ -3,11 +3,7 @@
3
# https://huggingface.co/Qwen/Qwen-7B/blob/main/modeling_qwen.py
4
# Copyright (c) Alibaba Cloud.
5
# LICENSE: https://huggingface.co/Qwen/Qwen-7B/blob/main/LICENSE
6
-"""Inference-only QWen model compatible with HuggingFace weights.
7
-
8
-The input of the model is flattened to a 1D tensor of tokens. The model uses
9
-InputMetadata to extract the original 2D shape of the input.
10
-"""
6
+"""Inference-only QWen model compatible with HuggingFace weights."""
7
from typing import Any, Dict, List, Optional, Tuple
8
9
import torch
vllm/model_executor/models/yi.py
+1
-5
@@ -20,11 +20,7 @@
20
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
21
# See the License for the specific language governing permissions and
22
# limitations under the License.
23
-"""Inference-only Yi model (https://01.ai) compatible with HuggingFace weights.
24
-
25
-The input of the model is flattened to a 1D tensor of tokens. The model uses
26
-InputMetadata to extract the original 2D shape of the input.
27
-"""
23
+"""Inference-only Yi model (https://01.ai) compatible with HuggingFace weights."""
24
from typing import Any, Dict, List, Optional, Tuple
25
26
import torch