Commit graph

4890 commits

Author SHA1 Message Date
didi
1de0653d70 Update token_counter.py 2024-09-10 19:07:50 +08:00
didi
bdf865eb0d Merge branch 'main' of https://github.com/didiforgithub/MetaGPT-MathAI 2024-09-10 19:06:17 +08:00
didi
7f45ef6231 Merge branch 'main' of https://github.com/didiforgithub/MetaGPT-MathAI 2024-09-10 19:04:05 +08:00
Zhaoyang Yu
445a2e6048 Update QA 2024-09-10 18:59:13 +08:00
didi
ab112462a5 添加了cost计算 2024-09-10 18:51:18 +08:00
Zhaoyang Yu
257b994409 Update drop.py 2024-09-10 18:45:54 +08:00
Zhaoyang Yu
68e87da378 Update Hotpotqa 2024-09-10 18:27:20 +08:00
Zhaoyang Yu
4ce18d7f48 Update baselines 2024-09-10 16:51:26 +08:00
duyipan
923109e882 add aide.py update README 2024-09-10 16:27:29 +08:00
Yizhou Chi
a373e684ae update di instruction 2024-09-10 16:05:22 +08:00
Yizhou Chi
b776c7309b add random tree search 2024-09-10 15:30:23 +08:00
Rayhao
e3663f2322 implement autuglu exp 2024-09-09 23:21:56 -07:00
Yizhou Chi
d34a482faf give dev label 2024-09-10 14:11:47 +08:00
Yizhou Chi
60e8e3eab8 fix hfdataset; make dirs when save notebook 2024-09-10 13:51:54 +08:00
didi
0b0a49d772 更新 hotpotqa Baseline 2024-09-10 12:48:28 +08:00
didi
c7c34cda7d Update human Eval 2024-09-10 10:58:39 +08:00
didi
62ffa730e0 update humaneval baseline & hotpotqa baseline 2024-09-10 10:57:21 +08:00
duyipan
af41f1f1cf Update README.md add aide setup and run 2024-09-09 11:26:23 +00:00
didi
7ffe68b499 重构了Evaluator 2024-09-09 18:19:22 +08:00
didi
4e0a896bdc 提交baseline例子;修改context-fill 格式识别方式 2024-09-09 17:17:15 +08:00
Yizhou Chi
294d0fe709 fix nbclient 2024-09-09 16:59:39 +08:00
Yizhou Chi
9ba0d217fc update mcts logic 2024-09-09 14:52:25 +08:00
林义章
e3fccce73d Merge branch 'experimenter' into 'expo'
Experimenter Update

See merge request agents/exp_optimizer!5
2024-09-09 06:13:07 +00:00
Yizhou Chi
401ca97846 update dataset.py 2024-09-09 13:54:09 +08:00
Yizhou Chi
93ff1f8f2b update datasets.yaml 2024-09-09 13:50:36 +08:00
Yizhou Chi
72dd44ae32 add ds agent's datasets 2024-09-09 13:47:59 +08:00
Yizhou Chi
9728b3a891 add greedy to run_experiment; add save_notebook to experimenter.py 2024-09-09 13:44:38 +08:00
didi
ca560a844f 合入Eval与Optimize
当前问题
1. eval 存在部分参数差异(path,csv测试)
2. optimize 尝试新流程(优化后的optimize曲线);optimize 模版书写
3. optimize 在各个数据集上跑通
4. 创建baseline folder
5. 创建experiment data收集方法
6. 从ags中移出
2024-09-08 23:04:01 +08:00
Yizhou Chi
df6fe9854d fix dataset bug 2024-09-06 19:27:36 +08:00
Yizhou Chi
c0262bcd8f 1. add support to hf dataset
2. add support to datasets that have both train and test
3. create data folder
4. fix new instruction bug
2024-09-06 19:05:10 +08:00
Yizhou Chi
376d1b7661 allow new instruction even if there's no insights 2024-09-06 17:02:49 +08:00
Yizhou Chi
6d40ec463f update result key 2024-09-06 14:18:04 +08:00
Yizhou Chi
a6f71f4498 1. avoid circular reference
2. add greedy
2024-09-06 13:09:18 +08:00
Yizhou Chi
e07ed0df8b fix mcts bug 2024-09-06 10:42:24 +08:00
Yizhou Chi
0e27e3d8be 使用code block做reflection 2024-09-06 10:31:01 +08:00
Yizhou Chi
defca81ebb fix time calculation 2024-09-06 10:24:49 +08:00
林义章
3583e2af63 Merge branch 'experimenter' into 'expo'
增加dataset_info到提示词

See merge request agents/exp_optimizer!4
2024-09-05 06:35:16 +00:00
Yizhou Chi
96ffcd285f 更新Prompt相关readme 2024-09-05 14:30:36 +08:00
Yizhou Chi
d27a48adb2 update prompt to include dataset info path 2024-09-05 13:01:00 +08:00
Yizhou Chi
45d176b48b remove deprecated scripts 2024-09-05 11:00:07 +08:00
Yizhou Chi
c16286a006 Refactor MCTS class to handle role running errors and improve error logging 2024-09-05 10:12:37 +08:00
林义章
d9bce92156 Merge branch 'experimenter' into 'expo'
fix import

See merge request agents/exp_optimizer!3
2024-09-04 11:06:16 +00:00
Yizhou Chi
58d7b14007 fix import 2024-09-04 18:46:16 +08:00
Yizhou Chi
2c43944ec0 fix import 2024-09-04 18:08:59 +08:00
林义章
697deb97db Merge branch 'experimenter' into 'expo'
Experimenter

See merge request agents/exp_optimizer!2
2024-09-04 10:00:56 +00:00
Yizhou Chi
ab8a1d6824 format code 2024-09-04 17:52:02 +08:00
Yizhou Chi
fcd1ba66a6 增加try catch 2024-09-04 16:38:33 +08:00
Yizhou Chi
aea524b4ea 1. update readme
2. fix duration
2024-09-04 14:49:18 +08:00
Yizhou Chi
72bd1665b1 ensure experimenter not evaluating csv from other experiments 2024-09-04 10:41:10 +08:00
Yizhou Chi
f23d2a72c9 add autogluon 2024-09-04 10:39:28 +08:00