Instructions to use cs-552-2026-camykaz/math_model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use cs-552-2026-camykaz/math_model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="cs-552-2026-camykaz/math_model") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("cs-552-2026-camykaz/math_model") model = AutoModelForCausalLM.from_pretrained("cs-552-2026-camykaz/math_model", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use cs-552-2026-camykaz/math_model with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "cs-552-2026-camykaz/math_model" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cs-552-2026-camykaz/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/cs-552-2026-camykaz/math_model
- SGLang
How to use cs-552-2026-camykaz/math_model with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "cs-552-2026-camykaz/math_model" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cs-552-2026-camykaz/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "cs-552-2026-camykaz/math_model" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cs-552-2026-camykaz/math_model", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use cs-552-2026-camykaz/math_model with Docker Model Runner:
docker model run hf.co/cs-552-2026-camykaz/math_model
Update Automated MNLP evaluation report (2026-06-08)
Browse files- EVAL_REPORT.md +64 -61
EVAL_REPORT.md
CHANGED
|
@@ -2,22 +2,22 @@
|
|
| 2 |
|
| 3 |
- **Model repo:** [`cs-552-2026-camykaz/math_model`](https://huggingface.co/cs-552-2026-camykaz/math_model)
|
| 4 |
- **Owner(s):** group **camykaz**
|
| 5 |
-
- **Generated at:** 2026-06-
|
| 6 |
- **Pipeline:** [mnlp-project-ci](https://github.com/eric11eca/mnlp-project-ci)
|
| 7 |
|
| 8 |
_This PR is opened automatically by the course CI. It is **non-blocking** — you do not need to merge it. The next nightly run will refresh this file._
|
| 9 |
|
| 10 |
## Evaluated checkpoint
|
| 11 |
|
| 12 |
-
- **Commit:** [`
|
| 13 |
-
- **Message:** Upload v7 checkpoint
|
| 14 |
-
- **Committed:** 2026-06-
|
| 15 |
|
| 16 |
## Summary
|
| 17 |
|
| 18 |
| Benchmark | Accuracy | Status |
|
| 19 |
|---|---:|---|
|
| 20 |
-
| Math | 0.
|
| 21 |
| Knowledge | — | not run |
|
| 22 |
| Multilingual | — | not run |
|
| 23 |
| Safety | — | not run |
|
|
@@ -30,87 +30,90 @@ _Prompts are intentionally omitted to avoid revealing benchmark contents. For mu
|
|
| 30 |
|
| 31 |
**Correct** (1 shown)
|
| 32 |
|
| 33 |
-
- **reference**: `
|
| 34 |
-
- **overall** (
|
| 35 |
-
- **extracted** (✓): `
|
| 36 |
- **completion**:
|
| 37 |
|
| 38 |
```text
|
| 39 |
-
Okay,
|
|
|
|
|
|
|
| 40 |
|
| 41 |
... [truncated 823 chars] ...
|
| 42 |
|
| 43 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
-
|
| 46 |
-
- The sum of the roots \(\alpha^2 + \beta^2 + \gamma^2 = \frac{n}{6}\)
|
| 47 |
-
- The sum of the products of the roots taken two at a time \(\alpha^2\beta^2 + \beta^2\gamma^2 + \gamma^2\alpha^2 = n - 11\)
|
| 48 |
-
- The product of the roots \(\alpha^2\beta^2\gamma^2 = 400\)
|
| 49 |
|
| 50 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
|
| 52 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
|
| 54 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
```
|
| 56 |
|
| 57 |
**Incorrect** (1 shown)
|
| 58 |
|
| 59 |
-
- **reference**: `
|
| 60 |
- **overall** (0/8 completions correct)
|
| 61 |
-
- **extracted** (✗): `
|
| 62 |
- **completion**:
|
| 63 |
|
| 64 |
```text
|
| 65 |
-
Okay,
|
| 66 |
|
| 67 |
-
First,
|
| 68 |
|
| 69 |
... [truncated 822 chars] ...
|
| 70 |
|
| 71 |
-
|
| 72 |
-
</think>
|
| 73 |
|
| 74 |
-
|
| 75 |
|
| 76 |
-
|
| 77 |
-
\[
|
| 78 |
-
1 + 1 + g(1) \mid 1 \cdot 1 + g(1) + 1 \cdot g(1) + 1
|
| 79 |
-
\]
|
| 80 |
-
This simplifies to:
|
| 81 |
-
\[
|
| 82 |
-
2 + g(1) \mid 2 + 2g(1)
|
| 83 |
-
\]
|
| 84 |
-
This implies:
|
| 85 |
-
\[
|
| 86 |
-
2 + g(1) \mid 2g(1)
|
| 87 |
-
\]
|
| 88 |
-
which means \( 2 + g(1) \) divides \( 2g(1) \). Testing values, we find \( g(1) = 0 \) or \( g(1) = 2 \).
|
| 89 |
|
| 90 |
-
|
| 91 |
-
\[
|
| 92 |
-
m + n + g(m) \mid mn + g(n) + n g(m) + n
|
| 93 |
-
\]
|
| 94 |
-
we get:
|
| 95 |
-
\[
|
| 96 |
-
m + n + (m^2 - m) \mid mn + (n^2 - n) + n(m^2 - m) + n
|
| 97 |
-
\]
|
| 98 |
-
Simplifying the right-hand side:
|
| 99 |
-
\[
|
| 100 |
-
mn + n^2 - n + nm^2 - nm + n = mn + n^2 + nm^2
|
| 101 |
-
\]
|
| 102 |
-
Thus, the condition becomes:
|
| 103 |
-
\[
|
| 104 |
-
m + n + m^2 - m \mid mn + n^2 + nm^2
|
| 105 |
-
\]
|
| 106 |
-
\[
|
| 107 |
-
n + m^2 \mid n(m^2 + n)
|
| 108 |
-
\]
|
| 109 |
-
This is always true because \( n + m^2 \) divides \( n(m^2 + n) \).
|
| 110 |
|
| 111 |
-
|
| 112 |
|
| 113 |
-
|
| 114 |
-
\boxed{g(n) = n^2 - n}
|
| 115 |
-
\]
|
| 116 |
```
|
|
|
|
| 2 |
|
| 3 |
- **Model repo:** [`cs-552-2026-camykaz/math_model`](https://huggingface.co/cs-552-2026-camykaz/math_model)
|
| 4 |
- **Owner(s):** group **camykaz**
|
| 5 |
+
- **Generated at:** 2026-06-08T04:40:54+00:00 (UTC)
|
| 6 |
- **Pipeline:** [mnlp-project-ci](https://github.com/eric11eca/mnlp-project-ci)
|
| 7 |
|
| 8 |
_This PR is opened automatically by the course CI. It is **non-blocking** — you do not need to merge it. The next nightly run will refresh this file._
|
| 9 |
|
| 10 |
## Evaluated checkpoint
|
| 11 |
|
| 12 |
+
- **Commit:** [`cc7a572`](https://huggingface.co/cs-552-2026-camykaz/math_model/commit/cc7a572d698cb4899527567350c6317c5a0751b7)
|
| 13 |
+
- **Message:** Upload v7 checkpoint 2400 math model
|
| 14 |
+
- **Committed:** 2026-06-06T21:39:38+00:00
|
| 15 |
|
| 16 |
## Summary
|
| 17 |
|
| 18 |
| Benchmark | Accuracy | Status |
|
| 19 |
|---|---:|---|
|
| 20 |
+
| Math | 0.5400 | ok |
|
| 21 |
| Knowledge | — | not run |
|
| 22 |
| Multilingual | — | not run |
|
| 23 |
| Safety | — | not run |
|
|
|
|
| 30 |
|
| 31 |
**Correct** (1 shown)
|
| 32 |
|
| 33 |
+
- **reference**: `40_9`
|
| 34 |
+
- **overall** (8/8 completions correct)
|
| 35 |
+
- **extracted** (✓): `40_9`
|
| 36 |
- **completion**:
|
| 37 |
|
| 38 |
```text
|
| 39 |
+
Okay, so I need to subtract two numbers in base 9: 58_9 minus 18_9. Hmm, let me remember how base conversions work. First, maybe I should convert both numbers to base 10 to make the subtraction easier, then convert the result back to base 9. That seems like a plan. Let me try that.
|
| 40 |
+
|
| 41 |
+
Starting with the first number, 58_9. In base 9, each digit represents a power of 9. The rightmost digit is 9^0, whi
|
| 42 |
|
| 43 |
... [truncated 823 chars] ...
|
| 44 |
|
| 45 |
+
us result. So that seems correct. Therefore, the answer is 40_9.
|
| 46 |
+
</think>
|
| 47 |
+
|
| 48 |
+
To solve the problem of evaluating \(58_9 - 18_9\) and expressing the result in base 9, we can follow these steps:
|
| 49 |
+
|
| 50 |
+
1. **Convert the numbers from base 9 to base 10:**
|
| 51 |
+
|
| 52 |
+
- Convert \(58_9\) to base 10:
|
| 53 |
+
\[
|
| 54 |
+
58_9 = 5 \cdot 9^1 + 8 \cdot 9^0 = 5 \cdot 9 + 8 \cdot 1 = 45 + 8 = 53_{10}
|
| 55 |
+
\]
|
| 56 |
+
|
| 57 |
+
- Convert \(18_9\) to base 10:
|
| 58 |
+
\[
|
| 59 |
+
18_9 = 1 \cdot 9^1 + 8 \cdot 9^0 = 1 \cdot 9 + 8 \cdot 1 = 9 + 8 = 17_{10}
|
| 60 |
+
\]
|
| 61 |
+
|
| 62 |
+
2. **Perform the subtraction in base 10:**
|
| 63 |
+
|
| 64 |
+
\[
|
| 65 |
+
53_{10} - 17_{10} = 36_{10}
|
| 66 |
+
\]
|
| 67 |
|
| 68 |
+
3. **Convert the result from base 10 back to base 9:**
|
|
|
|
|
|
|
|
|
|
| 69 |
|
| 70 |
+
To convert \(36_{10}\) to base 9, we need to find the highest power of 9 that is less than or equal to 36. The powers of 9 are:
|
| 71 |
+
\[
|
| 72 |
+
9^0 = 1, \quad 9^1 = 9, \quad 9^2 = 81
|
| 73 |
+
\]
|
| 74 |
+
Since \(9^2 = 81\) is greater than 36, we use \(9^1 = 9\).
|
| 75 |
|
| 76 |
+
- Determine how many times 9 fits into 36:
|
| 77 |
+
\[
|
| 78 |
+
36 \div 9 = 4
|
| 79 |
+
\]
|
| 80 |
+
- The quotient 4 is the coefficient for \(9^1\).
|
| 81 |
+
- The remainder is:
|
| 82 |
+
\[
|
| 83 |
+
36 - (4 \cdot 9) = 36 - 36 = 0
|
| 84 |
+
\]
|
| 85 |
|
| 86 |
+
Therefore, \(36_{10} = 40_9\).
|
| 87 |
+
|
| 88 |
+
Thus, the result of \(58_9 - 18_9\) is:
|
| 89 |
+
\[
|
| 90 |
+
\boxed{40_9}
|
| 91 |
+
\]
|
| 92 |
```
|
| 93 |
|
| 94 |
**Incorrect** (1 shown)
|
| 95 |
|
| 96 |
+
- **reference**: `19`
|
| 97 |
- **overall** (0/8 completions correct)
|
| 98 |
+
- **extracted** (✗): `<no answer>`
|
| 99 |
- **completion**:
|
| 100 |
|
| 101 |
```text
|
| 102 |
+
Okay, let's try to tackle this problem step by step. So, we have an 11-sided polygon that's non-convex, and there are three conditions given. The goal is to find the sum m + n + p + q for the perimeter expression given. Hmm, let's parse the problem again.
|
| 103 |
|
| 104 |
+
First, the polygon is A1A2A3...A11. For each i from 2 to 10, the triangle A_iA1A_{i+1} has area 1. Also, the cosine of the angle at each of the
|
| 105 |
|
| 106 |
... [truncated 822 chars] ...
|
| 107 |
|
| 108 |
+
aybe x_2 = 26/5, x_3 = 1, x_4 = 26/5, etc., but we saw that this leads to a perimeter of 24.63, which is still not 20.
|
|
|
|
| 109 |
|
| 110 |
+
Given that all these approaches are not working, perhaps there's a different insight. Let's recall that the problem states the polygon is non-convex. This might mean that some of the interior angles are reflex angles (greater than 180 degrees), which could affect the lengths of the sides. However, I'm not sure how this directly impacts the sides adjacent to A1.
|
| 111 |
|
| 112 |
+
Alternatively, maybe we can consider that the polygon is composed of triangles A_iA1A_{i+1} with area 1, and the sum of all these triangles' areas is 10, but the polygon's area is less than that. However, this is not directly helpful.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 113 |
|
| 114 |
+
Alternatively, perhaps we can use the fact that the sum of the areas of the triangles is 10, but the area of the polygon is related to these triangles. However, since the polygon is non-convex, some areas might overlap, making this approach complicated.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 115 |
|
| 116 |
+
Alternatively, perhaps the key is to use the Law of Cosines for the entire polygon. But with 11 sides, this would be very complex.
|
| 117 |
|
| 118 |
+
Alternatively, perhaps the problem requires recognizing that the sides A1A2 and A1A11 form a
|
|
|
|
|
|
|
| 119 |
```
|