Update README.md
Browse files
README.md
CHANGED
@@ -66,7 +66,8 @@ We started the second-stage training on top of [EurusPRM-Stage1](https://hugging
|
|
66 |
|
67 |
## Usage
|
68 |
|
69 |
-
Due to the training format of **EurusPRM**, a clearly separated answer should be provided to elicit a better performance. We strongly recommend
|
|
|
70 |
We show an example leveraging **EurusPRM-Stage2** below:
|
71 |
|
72 |
```python
|
@@ -75,10 +76,10 @@ from transformers import AutoTokenizer,AutoModelForCausalLM
|
|
75 |
coef=0.001
|
76 |
d = {'query':'Convert the point $(0,3)$ in rectangular coordinates to polar coordinates. Enter your answer in the form $(r,\\theta),$ where $r > 0$ and $0 \\le \\theta < 2 \\pi.$',
|
77 |
'answer':[
|
78 |
-
"Step 1: To convert the point (0,3) from rectangular coordinates to polar coordinates, we need to find the radius (r) and the angle theta (\u03b8).",
|
79 |
-
"Step
|
80 |
-
"Step
|
81 |
-
"Step
|
82 |
]
|
83 |
}
|
84 |
model = AutoModelForCausalLM.from_pretrained('PRIME-RL/EurusPRM-Stage2')
|
|
|
66 |
|
67 |
## Usage
|
68 |
|
69 |
+
***Due to the training format of **EurusPRM**, a clearly separated answer should be provided to elicit a better performance. We strongly recommend using answers with each step starting with ``Step K``.***
|
70 |
+
|
71 |
We show an example leveraging **EurusPRM-Stage2** below:
|
72 |
|
73 |
```python
|
|
|
76 |
coef=0.001
|
77 |
d = {'query':'Convert the point $(0,3)$ in rectangular coordinates to polar coordinates. Enter your answer in the form $(r,\\theta),$ where $r > 0$ and $0 \\le \\theta < 2 \\pi.$',
|
78 |
'answer':[
|
79 |
+
"Step 1: To convert the point (0,3) from rectangular coordinates to polar coordinates, we need to find the radius (r) and the angle theta (\u03b8).",
|
80 |
+
"Step 2: Find the radius (r). The radius is the distance from the origin (0,0) to the point (0,3). Since the x-coordinate is 0, the distance is simply the absolute value of the y-coordinate. So, r = |3| = 3.",
|
81 |
+
"Step 3: Find the angle theta (\u03b8). The angle theta is measured counterclockwise from the positive x-axis. Since the point (0,3) lies on the positive y-axis, the angle theta is 90 degrees or \u03c0\/2 radians.",
|
82 |
+
"Step 4: Write the polar coordinates. The polar coordinates are (r, \u03b8), where r > 0 and 0 \u2264 \u03b8 < 2\u03c0. In this case, r = 3 and \u03b8 = \u03c0\/2.\n\nTherefore, the polar coordinates of the point (0,3) are (3, \u03c0\/2).\n\n\n\\boxed{(3,\\frac{\\pi}{2})}"
|
83 |
]
|
84 |
}
|
85 |
model = AutoModelForCausalLM.from_pretrained('PRIME-RL/EurusPRM-Stage2')
|