Spaces:
Sleeping
Sleeping
Update src/about.py
#1
by
hou12q
- opened
- src/about.py +24 -63
src/about.py
CHANGED
|
@@ -1,72 +1,33 @@
|
|
| 1 |
-
|
| 2 |
-
from enum import Enum
|
| 3 |
|
| 4 |
-
|
| 5 |
-
class Task:
|
| 6 |
-
benchmark: str
|
| 7 |
-
metric: str
|
| 8 |
-
col_name: str
|
| 9 |
|
|
|
|
| 10 |
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
# task_key in the json file, metric_key in the json file, name to display in the leaderboard
|
| 15 |
-
task0 = Task("anli_r1", "acc", "ANLI")
|
| 16 |
-
task1 = Task("logiqa", "acc_norm", "LogiQA")
|
| 17 |
|
| 18 |
-
|
| 19 |
-
|
|
|
|
|
|
|
|
|
|
| 20 |
|
|
|
|
|
|
|
|
|
|
| 21 |
|
|
|
|
| 22 |
|
| 23 |
-
|
| 24 |
-
TITLE = """<h1 align="center" id="space-title">Demo leaderboard</h1>"""
|
| 25 |
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
Intro text
|
| 29 |
-
"""
|
| 30 |
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
"""
|
| 39 |
-
|
| 40 |
-
EVALUATION_QUEUE_TEXT = """
|
| 41 |
-
## Some good practices before submitting a model
|
| 42 |
-
|
| 43 |
-
### 1) Make sure you can load your model and tokenizer using AutoClasses:
|
| 44 |
-
```python
|
| 45 |
-
from transformers import AutoConfig, AutoModel, AutoTokenizer
|
| 46 |
-
config = AutoConfig.from_pretrained("your model name", revision=revision)
|
| 47 |
-
model = AutoModel.from_pretrained("your model name", revision=revision)
|
| 48 |
-
tokenizer = AutoTokenizer.from_pretrained("your model name", revision=revision)
|
| 49 |
-
```
|
| 50 |
-
If this step fails, follow the error messages to debug your model before submitting it. It's likely your model has been improperly uploaded.
|
| 51 |
-
|
| 52 |
-
Note: make sure your model is public!
|
| 53 |
-
Note: if your model needs `use_remote_code=True`, we do not support this option yet but we are working on adding it, stay posted!
|
| 54 |
-
|
| 55 |
-
### 2) Convert your model weights to [safetensors](https://huggingface.co/docs/safetensors/index)
|
| 56 |
-
It's a new format for storing weights which is safer and faster to load and use. It will also allow us to add the number of parameters of your model to the `Extended Viewer`!
|
| 57 |
-
|
| 58 |
-
### 3) Make sure your model has an open license!
|
| 59 |
-
This is a leaderboard for Open LLMs, and we'd love for as many people as possible to know they can use your model 🤗
|
| 60 |
-
|
| 61 |
-
### 4) Fill up your model card
|
| 62 |
-
When we add extra information about models to the leaderboard, it will be automatically taken from the model card
|
| 63 |
-
|
| 64 |
-
## In case of model failure
|
| 65 |
-
If your model is displayed in the `FAILED` category, its execution stopped.
|
| 66 |
-
Make sure you have followed the above steps first.
|
| 67 |
-
If everything is done, check you can launch the EleutherAIHarness on your model locally, using the above command without modifications (you can add `--limit` to limit the number of examples per task).
|
| 68 |
-
"""
|
| 69 |
-
|
| 70 |
-
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
|
| 71 |
-
CITATION_BUTTON_TEXT = r"""
|
| 72 |
-
"""
|
|
|
|
| 1 |
+
# Report
|
|
|
|
| 2 |
|
| 3 |
+
## 1. 模型及类别选择
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
|
| 5 |
+
本次实验选用了三类大模型:Llama 3, Mistral 7B, ChatGPT。
|
| 6 |
|
| 7 |
+
- **Llama 3**:开源社区广泛使用,适合中英文任务。
|
| 8 |
+
- **Mistral 7B**:轻量级,适合边缘设备。
|
| 9 |
+
- **ChatGPT**:闭源,适合通用对话任务,表现最优。
|
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
+
| 模型名称 | 参数量 | 开源情况 | 主要用途 |
|
| 12 |
+
|------------|--------|---------|----------------|
|
| 13 |
+
| Llama 3 | 70B | 是 | 多语言任务 |
|
| 14 |
+
| Mistral 7B | 7B | 是 | 低功耗推理任务 |
|
| 15 |
+
| ChatGPT | 未公开 | 否 | 通用对话、推理任务 |
|
| 16 |
|
| 17 |
+
**选择理由**:
|
| 18 |
+
- Llama 3和Mistral为开源,方便定制与修改;
|
| 19 |
+
- ChatGPT性能优越,作为基准。
|
| 20 |
|
| 21 |
+
---
|
| 22 |
|
| 23 |
+
## 2. 系统实现细节
|
|
|
|
| 24 |
|
| 25 |
+
### Gradio交互界面截图
|
| 26 |
+

|
|
|
|
|
|
|
| 27 |
|
| 28 |
+
### 输入与输出流程图
|
| 29 |
+
```mermaid
|
| 30 |
+
graph TD
|
| 31 |
+
A[用户输入] --> B[Gradio界面]
|
| 32 |
+
B --> C[模型推理]
|
| 33 |
+
C --> D[返回结果]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|