Replace ModelSpecs with CompletionFn #594
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jwang47
force-pushed
the
alvin/evals_refactor_modelspecs
branch
4 times, most recently
from
bc4ceda
to
481c456
Compare
jwang47
force-pushed
the
alvin/evals_refactor_modelspecs
branch
from
481c456
to
59b85aa
Compare
jwang47
changed the title
| @@ -26,10 +24,8 @@ def _purple(str): | |||
|
|
|||
| def get_parser() -> argparse.ArgumentParser: | |||
| parser = argparse.ArgumentParser(description="Run evals through the API") | |||
| parser.add_argument("model", type=str, help="Name of a completion model.") | |||
| parser.add_argument("completion_fn", type=str, help="One or more CompletionFn URLs, separated by commas (,). The format of a CompletionFn URL can be two forms: 1) an OpenAI API model followed by query parameters (e.g. `gpt-3.5-turbo?api_key=..`) or 2) a path to a Python class followed by query parameters (e.g. `evals.api:OpenAICompletionFn?model=text-davinci-003`).") | |||
There was a problem hiding this comment.
I tried multiple completion_fns but doesnt seem to work, e.g.
oaieval text-davinci-003,gpt-3.5-turbo pattern_identification --max_samples 1
errors with
AssertionError: Got type <class 'list'>, with val <class 'dict'> for prompt, expected str or list[int] or list[str]
There was a problem hiding this comment.
Looking at the error, realize it's independent of the multiple completion_fns but bc OpenAICompletionFn does not like the chat prompt. I think we should change this assert to be the same as the one for OpenAIChatCompletionFn because to_openai_create_prompt should be able to handle converting the Chat format prompt into strings
There was a problem hiding this comment.
Changed to be same assert as OpenAIChatCompletionFn
evals/elsuite/basic/includes.py
Outdated
|
|
||
| def eval_sample(self, sample: Any, *_): | ||
| prompt = sample["input"] | ||
| result = self._completion_fn( | ||
| result = self.completion_fn( | ||
| prompt=prompt, | ||
| max_tokens=self.max_tokens, |
There was a problem hiding this comment.
Similarly, max_tokens needs to be moved as not relevant to all completionfns
|
|
||
| def eval_sample(self, test_sample, rng): | ||
| del rng | ||
| prompt, correct_answers = test_sample["input"], test_sample["ideal"] | ||
| result = self._completion_fn( | ||
| result = self.completion_fn( | ||
| prompt=prompt, | ||
| temperature=0.0, # Q: why are these hardcoded? |
There was a problem hiding this comment.
Also these need to be moved
|
LGTM, some minor comments. I am also curious about where do we expect most people to write their implementations of CompletionFns |
jwang47
force-pushed
the
alvin/evals_refactor_modelspecs
branch
from
8d7c964
to
9ec3693
Compare
jwang47
force-pushed
the
alvin/evals_refactor_modelspecs
branch
from
9ec3693
to
99ca1b3
Compare
|
Some issues we found: I believe |
jwang47
force-pushed
the
alvin/evals_refactor_modelspecs
branch
from
842dd17
to
ba8708d
Compare
# Thank you for contributing an eval!♥️ 🚨 Please make sure your PR follows these guidelines, __failure to follow the guidelines below will result in the PR being closed automatically__. Note that even if the criteria are met, that does not guarantee the PR will be merged nor GPT-4 access granted. 🚨 __PLEASE READ THIS__: In order for a PR to be merged, it must fail on GPT-4. We are aware that right now, users do not have access, so you will not be able to tell if the eval fails or not. Please run your eval with GPT-3.5-Turbo, but keep in mind as we run the eval, if GPT-4 gets higher than 90% on the eval, we will likely reject since GPT-4 is already capable of completing the task. We plan to roll out a way for users submitting evals to see the eval performance on GPT-4 soon. Stay tuned! Until then, you will not be able to see the eval performance on GPT-4. We encourage partial PR's with ~5-10 example that we can then run the evals on and share the results with you so you know how your eval does with GPT-4 before writing all 100 examples. ## Eval details 📑 ### Eval name [Insert Eval name here] ### Eval description [Insert a short description of what your eval does here] ### What makes this a useful eval? [Insert why this eval is worth including and any additional context] ## Criteria for a good eval ✅ Below are some of the criteria we look for in a good eval. In general, we are seeking cases where the model does not do a good job despite being capable of generating a good response (note that there are some things large language models cannot do, so those would not make good evals). Your eval should be: - [ ] Thematically consistent: The eval should be thematically consistent. We'd like to see a number of prompts all demonstrating some particular failure mode. For example, we can create an eval on cases where the model fails to reason about the physical world. - [ ] Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not. - [ ] Includes good signal around what is the right behavior. This means either a correct answer for `Basic` evals or the `Fact` Model-graded eval, or an exhaustive rubric for evaluating answers for the `Criteria` Model-graded eval. - [ ] Include at least 100 high quality examples (it is okay to only contribute 5-10 meaningful examples and have us test them with GPT-4 before adding all 100) If there is anything else that makes your eval worth including, please document it below. ### Unique eval value > Insert what makes your eval high quality that was not mentioned above. (Not required) ## Eval structure 🏗️ Your eval should - [ ] Check that your data is in `evals/registry/data/{name}` - [ ] Check that your yaml is registered at `evals/registry/evals/{name}.yaml` - [ ] Ensure you have the right to use the data you submit via this eval (For now, we will only be approving evals that use one of the existing eval classes. You may still write custom eval classes for your own cases, and we may consider merging them in the future.) ## Final checklist 👀 ### Submission agreement By contributing to Evals, you are agreeing to make your evaluation logic and data under the same MIT license as this repository. You must have adequate rights to upload any data used in an Eval. OpenAI reserves the right to use this data in future service improvements to our product. Contributions to OpenAI Evals will be subject to our usual Usage Policies (https://platform.openai.com/docs/usage-policies). - [ ] I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies. ### Email address validation If your submission is accepted, we will be granting GPT-4 access to a limited number of contributors. Access will be given to the email address associated with the merged pull request. - [ ] I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request. ### Limited availability acknowledgement We know that you might be excited to contribute to OpenAI's mission, help improve our models, and gain access to GPT-4. However, due to the requirements mentioned above and high volume of submissions, we will not be able to accept all submissions and thus not grant everyone who opens a PR GPT-4 access. We know this is disappointing, but we hope to set the right expectation before you open this PR. - [ ] I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access granted. ### Submit eval - [ ] I have filled out all required fields in the evals PR form - [ ] (Ignore if not submitting code) I have run `pip install pre-commit; pre-commit install` and have verified that `black`, `isort`, and `autoflake` are running when I commit and push Failure to fill out all required fields will result in the PR being closed. ### Eval JSON data Since we are using Git LFS, we are asking eval submitters to add in as many Eval Samples (at least 5) from their contribution here: <details> <summary>View evals in JSON</summary> ### Eval ```jsonl INSERT_EVAL_HERE ``` </details> --------- Co-authored-by: Alvin Wang <[email protected]>
andrew-openai
approved these changes
Apr 6, 2023
jwang47
added a commit
that referenced
this pull request
Apr 6, 2023
- [evals] Refactor evals package to expose `completion_fn`. - Add `record_raw_samples` - Andrew/evals refactor (#579) - update manifest and pyproject to support fetching data on pip install (#592) - we need to still use the interop for string/list[dicts] for modelgraded evals - refactor simple evals to not use result.prompt (#593) - Clean up duplicate recordings - Replace ModelSpecs with CompletionFn (#594) - Add --registry_path CLI arg # Thank you for contributing an eval!♥️ 🚨 Please make sure your PR follows these guidelines, __failure to follow the guidelines below will result in the PR being closed automatically__. Note that even if the criteria are met, that does not guarantee the PR will be merged nor GPT-4 access granted. 🚨 __PLEASE READ THIS__: In order for a PR to be merged, it must fail on GPT-4. We are aware that right now, users do not have access, so you will not be able to tell if the eval fails or not. Please run your eval with GPT-3.5-Turbo, but keep in mind as we run the eval, if GPT-4 gets higher than 90% on the eval, we will likely reject since GPT-4 is already capable of completing the task. We plan to roll out a way for users submitting evals to see the eval performance on GPT-4 soon. Stay tuned! Until then, you will not be able to see the eval performance on GPT-4. We encourage partial PR's with ~5-10 example that we can then run the evals on and share the results with you so you know how your eval does with GPT-4 before writing all 100 examples. ## Eval details 📑 ### Eval name [Insert Eval name here] ### Eval description [Insert a short description of what your eval does here] ### What makes this a useful eval? [Insert why this eval is worth including and any additional context] ## Criteria for a good eval ✅ Below are some of the criteria we look for in a good eval. In general, we are seeking cases where the model does not do a good job despite being capable of generating a good response (note that there are some things large language models cannot do, so those would not make good evals). Your eval should be: - [ ] Thematically consistent: The eval should be thematically consistent. We'd like to see a number of prompts all demonstrating some particular failure mode. For example, we can create an eval on cases where the model fails to reason about the physical world. - [ ] Contains failures where a human can do the task, but either GPT-4 or GPT-3.5-Turbo could not. - [ ] Includes good signal around what is the right behavior. This means either a correct answer for `Basic` evals or the `Fact` Model-graded eval, or an exhaustive rubric for evaluating answers for the `Criteria` Model-graded eval. - [ ] Include at least 100 high quality examples (it is okay to only contribute 5-10 meaningful examples and have us test them with GPT-4 before adding all 100) If there is anything else that makes your eval worth including, please document it below. ### Unique eval value > Insert what makes your eval high quality that was not mentioned above. (Not required) ## Eval structure 🏗️ Your eval should - [ ] Check that your data is in `evals/registry/data/{name}` - [ ] Check that your yaml is registered at `evals/registry/evals/{name}.yaml` - [ ] Ensure you have the right to use the data you submit via this eval (For now, we will only be approving evals that use one of the existing eval classes. You may still write custom eval classes for your own cases, and we may consider merging them in the future.) ## Final checklist 👀 ### Submission agreement By contributing to Evals, you are agreeing to make your evaluation logic and data under the same MIT license as this repository. You must have adequate rights to upload any data used in an Eval. OpenAI reserves the right to use this data in future service improvements to our product. Contributions to OpenAI Evals will be subject to our usual Usage Policies (https://platform.openai.com/docs/usage-policies). - [ ] I agree that my submission will be made available under an MIT license and complies with OpenAI's usage policies. ### Email address validation If your submission is accepted, we will be granting GPT-4 access to a limited number of contributors. Access will be given to the email address associated with the merged pull request. - [ ] I acknowledge that GPT-4 access will only be granted, if applicable, to the email address used for my merged pull request. ### Limited availability acknowledgement We know that you might be excited to contribute to OpenAI's mission, help improve our models, and gain access to GPT-4. However, due to the requirements mentioned above and high volume of submissions, we will not be able to accept all submissions and thus not grant everyone who opens a PR GPT-4 access. We know this is disappointing, but we hope to set the right expectation before you open this PR. - [ ] I understand that opening a PR, even if it meets the requirements above, does not guarantee the PR will be merged nor GPT-4 access granted. ### Submit eval - [ ] I have filled out all required fields in the evals PR form - [ ] (Ignore if not submitting code) I have run `pip install pre-commit; pre-commit install` and have verified that `black`, `isort`, and `autoflake` are running when I commit and push Failure to fill out all required fields will result in the PR being closed. ### Eval JSON data Since we are using Git LFS, we are asking eval submitters to add in as many Eval Samples (at least 5) from their contribution here: <details> <summary>View evals in JSON</summary> ### Eval ```jsonl INSERT_EVAL_HERE ``` </details>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Replace ModelSpec with CompletionFn and allow users to specify CompletionFn instances from the CLI.
Testing done: