Update README.md
Browse files
README.md
CHANGED
@@ -28,7 +28,7 @@ from deepsparse import TextGeneration
|
|
28 |
system_message = ""
|
29 |
prompt = "Write a quick sort algorithm in Python"
|
30 |
formatted_prompt = f"<|im_start|>system\n{system_message}<|im_end|>\n<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant"
|
31 |
-
model = TextGeneration(model="hf:mgoin/
|
32 |
print(model(formatted_prompt, max_new_tokens=500).generations[0].text)
|
33 |
```
|
34 |
|
@@ -51,7 +51,7 @@ See the `recipe.yaml` in this repo and follow the instructions below.
|
|
51 |
git clone https://github.com/neuralmagic/sparseml
|
52 |
pip install -e "sparseml[transformers]"
|
53 |
python sparseml/src/sparseml/transformers/sparsification/obcq/obcq.py teknium/OpenHermes-2.5-Mistral-7B open_platypus --recipe recipe.yaml --save True
|
54 |
-
python sparseml/src/sparseml/transformers/sparsification/obcq/export.py --task text-generation --model_path obcq_deployment
|
55 |
cp deployment/model.onnx deployment/model-orig.onnx
|
56 |
```
|
57 |
|
|
|
28 |
system_message = ""
|
29 |
prompt = "Write a quick sort algorithm in Python"
|
30 |
formatted_prompt = f"<|im_start|>system\n{system_message}<|im_end|>\n<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant"
|
31 |
+
model = TextGeneration(model="hf:mgoin/OpenHermes-2.5-Mistral-7B-pruned50-quant-ds")
|
32 |
print(model(formatted_prompt, max_new_tokens=500).generations[0].text)
|
33 |
```
|
34 |
|
|
|
51 |
git clone https://github.com/neuralmagic/sparseml
|
52 |
pip install -e "sparseml[transformers]"
|
53 |
python sparseml/src/sparseml/transformers/sparsification/obcq/obcq.py teknium/OpenHermes-2.5-Mistral-7B open_platypus --recipe recipe.yaml --save True
|
54 |
+
python sparseml/src/sparseml/transformers/sparsification/obcq/export.py --task text-generation --model_path obcq_deployment --sequence_length 4096
|
55 |
cp deployment/model.onnx deployment/model-orig.onnx
|
56 |
```
|
57 |
|