Update README.md
Browse files
README.md
CHANGED
@@ -40,13 +40,15 @@ from transformers import pipeline
|
|
40 |
# Chose the model inference precision
|
41 |
dtype = torch.float16 # options are torch.float16, torch.bfloat16, torch.float32
|
42 |
|
43 |
-
model = pipeline(
|
|
|
|
|
|
|
|
|
44 |
|
45 |
prompt = "YOUR PROMPT HERE"
|
46 |
|
47 |
output = model(prompt, max_length=2048, do_sample=True)
|
48 |
-
|
49 |
-
|
50 |
```
|
51 |
|
52 |
|
@@ -60,14 +62,17 @@ from transformers import pipeline
|
|
60 |
dtype = torch.float16 # options are torch.bfloat16, torch.float32
|
61 |
model = pipeline(model="VMware/flan-ul2-alpaca-lora",device_map = 'auto',torch_dtype=dtype )
|
62 |
|
63 |
-
prompt_template = "
|
64 |
-
|
65 |
-
prompt = "YOUR PROMPT HERE"
|
66 |
|
67 |
-
|
|
|
68 |
|
|
|
69 |
|
|
|
70 |
|
|
|
71 |
```
|
72 |
|
73 |
# Training Details
|
@@ -81,10 +86,6 @@ The model was trained on 3xV100 GPUs using PEFT-LORA and Deepspeed
|
|
81 |
* epochs = 3
|
82 |
|
83 |
|
84 |
-
```
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
# Limitations and Bias
|
89 |
|
90 |
The model is based on a large and diverse dataset, but it may still have limitations and biases in certain areas. Some limitations include:
|
|
|
40 |
# Chose the model inference precision
|
41 |
dtype = torch.float16 # options are torch.float16, torch.bfloat16, torch.float32
|
42 |
|
43 |
+
model = pipeline(
|
44 |
+
model = "VMware/flan-ul2-alpaca-lora",
|
45 |
+
device_map = 'auto',
|
46 |
+
torch_dtype=dtype
|
47 |
+
)
|
48 |
|
49 |
prompt = "YOUR PROMPT HERE"
|
50 |
|
51 |
output = model(prompt, max_length=2048, do_sample=True)
|
|
|
|
|
52 |
```
|
53 |
|
54 |
|
|
|
62 |
dtype = torch.float16 # options are torch.bfloat16, torch.float32
|
63 |
model = pipeline(model="VMware/flan-ul2-alpaca-lora",device_map = 'auto',torch_dtype=dtype )
|
64 |
|
65 |
+
prompt_template = """
|
66 |
+
Below is an instruction that describes a task. Write a response that appropriately completes the request.
|
|
|
67 |
|
68 |
+
### Instruction:
|
69 |
+
{instruction}
|
70 |
|
71 |
+
### Response:"""
|
72 |
|
73 |
+
prompt = "YOUR INSTRUCTION HERE"
|
74 |
|
75 |
+
output = model(prompt_template.format(instruction=prompt), max_length=2048, do_sample=True)
|
76 |
```
|
77 |
|
78 |
# Training Details
|
|
|
86 |
* epochs = 3
|
87 |
|
88 |
|
|
|
|
|
|
|
|
|
89 |
# Limitations and Bias
|
90 |
|
91 |
The model is based on a large and diverse dataset, but it may still have limitations and biases in certain areas. Some limitations include:
|