Skip to content

Commit 91fd960

Browse files
committed
Add instructions for testing other models
1 parent e8a130e commit 91fd960

File tree

2 files changed

+13
-2
lines changed

2 files changed

+13
-2
lines changed

README.md

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -332,6 +332,14 @@ For other general inference settings, please modify the command or
332332
directly modify the
333333
[script](https://github.com/hkust-nlp/dart-math/blob/main/pipeline/gen.py).
334334

335+
- To test **base** models, please add the corresponding **ID** to
336+
`BASE_MODEL_IDS` from
337+
[dart_math.utils](https://github.com/hkust-nlp/dart-math/blob/main/dart_math/utils.py).
338+
- To test **instruct** models, please add the corresponding **prompt
339+
template** to `PROMPT_TEMPLATE_ID2DICT` from
340+
[dart_math.utils](https://github.com/hkust-nlp/dart-math/blob/main/dart_math/utils.py)
341+
and specify with `--prompt_template`.
342+
335343
You can also add the `--gen_only` option to only generate responses
336344
without evaluation and use the
337345
[`EvaluatorMathBatch`](https://hkust-nlp.github.io/dart-math/eval.html#evaluatormathbatch)

nbs/index.ipynb

Lines changed: 5 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -352,6 +352,9 @@
352352
"\n",
353353
"For other general inference settings, please modify the command or directly modify the [script](https://github.com/hkust-nlp/dart-math/blob/main/pipeline/gen.py).\n",
354354
"\n",
355+
"- To test **base** models, please add the corresponding **ID** to `BASE_MODEL_IDS` from [dart_math.utils](https://github.com/hkust-nlp/dart-math/blob/main/dart_math/utils.py).\n",
356+
"- To test **instruct** models, please add the corresponding **prompt template** to `PROMPT_TEMPLATE_ID2DICT` from [dart_math.utils](https://github.com/hkust-nlp/dart-math/blob/main/dart_math/utils.py) and specify with `--prompt_template`.\n",
357+
"\n",
355358
"You can also add the `--gen_only` option to only generate responses without evaluation and use the `EvaluatorMathBatch` to grade the generations by yourself. Please check the [grading script](pipeline/grade.py) for example.\n"
356359
]
357360
},
@@ -527,7 +530,7 @@
527530
"cell_type": "markdown",
528531
"metadata": {},
529532
"source": [
530-
"## 🌟 Star History"
533+
"## 🌟 Star History\n"
531534
]
532535
},
533536
{
@@ -540,7 +543,7 @@
540543
" <source media=\"(prefers-color-scheme: light)\" srcset=\"https://api.star-history.com/svg?repos=hkust-nlp/dart-math&type=Date\" />\n",
541544
" <img alt=\"Star History Chart\" src=\"https://api.star-history.com/svg?repos=hkust-nlp/dart-math&type=Date\" />\n",
542545
" </picture>\n",
543-
"</a>"
546+
"</a>\n"
544547
]
545548
},
546549
{

0 commit comments

Comments
 (0)