Skip to content

armgit / dspy

Public Imported from GitHub
0

DSPy: The framework for programming—not prompting—language models

Python

Fix typos (#1707)

Viktor Szépe649be4a6f49f
diff --git a/.github/workflows/precommits_check.yml b/.github/workflows/precommits_check.yml
index 7ac80deda..20c9c8368 100644
--- a/.github/workflows/precommits_check.yml
+++ b/.github/workflows/precommits_check.yml
@@ -24,7 +24,7 @@ jobs:
run: |
echo "Changed files"
echo ${{ steps.files.outputs.all }}
- echo "Github Client version"
+ echo "GitHub Client version"
echo $(gh --version)
- name: Pre-Commit Checks
run: |
diff --git a/README.md b/README.md
index 9a9eb4384..d892e1806 100644
--- a/README.md
+++ b/README.md
@@ -415,7 +415,7 @@ Guidance, LMQL, RELM, and Outlines are all exciting new libraries for controllin
This is very useful in many settings, but it's generally focused on low-level, structured control of a single LM call. It doesn't help ensure the JSON (or structured output) you get is going to be correct or useful for your task.
-In contrast, **DSPy** automatically optimizes the prompts in your programs to align them with various task needs, which may also include producing valid structured ouputs. That said, we are considering allowing **Signatures** in **DSPy** to express regex-like constraints that are implemented by these libraries.
+In contrast, **DSPy** automatically optimizes the prompts in your programs to align them with various task needs, which may also include producing valid structured outputs. That said, we are considering allowing **Signatures** in **DSPy** to express regex-like constraints that are implemented by these libraries.
</details>
## Testing
diff --git a/docs/docs/cheatsheet.md b/docs/docs/cheatsheet.md
index 754d5295e..c6d3a8f0c 100644
--- a/docs/docs/cheatsheet.md
+++ b/docs/docs/cheatsheet.md
@@ -230,7 +230,7 @@ def gsm8k_metric(gold, pred, trace=None) -> int:
class FactJudge(dspy.Signature):
"""Judge if the answer is factually correct based on the context."""
- context = dspy.InputField(desc="Context for the prediciton")
+ context = dspy.InputField(desc="Context for the prediction")
question = dspy.InputField(desc="Question to be answered")
answer = dspy.InputField(desc="Answer for the question")
factually_correct = dspy.OutputField(desc="Is the answer factually correct based on the context?", prefix="Factual[Yes/No]:")
@@ -417,7 +417,7 @@ optimized_program = teleprompter.compile(
optimized_program.save(f"mipro_optimized")
# Evaluate optimized program
-print(f"Evluate optimized program...")
+print(f"Evaluate optimized program...")
evaluate(optimized_program, devset=devset[:])
```
@@ -446,7 +446,7 @@ optimized_program = teleprompter.compile(
optimized_program.save(f"mipro_optimized")
# Evaluate optimized program
-print(f"Evluate optimized program...")
+print(f"Evaluate optimized program...")
evaluate(optimized_program, devset=devset[:])
```
### Signature Optimizer with Types
diff --git a/docs/docs/deep-dive/data-handling/loading-custom-data.md b/docs/docs/deep-dive/data-handling/loading-custom-data.md
index 405a90f48..196d0119b 100644
--- a/docs/docs/deep-dive/data-handling/loading-custom-data.md
+++ b/docs/docs/deep-dive/data-handling/loading-custom-data.md
@@ -86,7 +86,7 @@ Using the Dataset base class now makes loading custom datasets incredibly easy a
!!! caution
- We did not populate `_test` attribute in the above code, which is fine and won't cause any unneccesary error as such. However it'll give you an error if you try to access the test split.
+ We did not populate `_test` attribute in the above code, which is fine and won't cause any unnecessary error as such. However it'll give you an error if you try to access the test split.
```python
dataset.test[:5]
@@ -110,6 +110,6 @@ Using the Dataset base class now makes loading custom datasets incredibly easy a
To prevent that you'll just need to make sure `_test` is not `None` and populated with the appropriate data.
-You can overide the methods in `Dataset` class to customize your class even more.
+You can override the methods in `Dataset` class to customize your class even more.
In summary, the Dataset base class provides a simplistic way to load and preprocess custom datasets with minimal code!
diff --git a/docs/docs/deep-dive/language_model_clients/lm_local_models/HFClientTGI.md b/docs/docs/deep-dive/language_model_clients/lm_local_models/HFClientTGI.md
index 0dd420056..3c136e0de 100644
--- a/docs/docs/deep-dive/language_model_clients/lm_local_models/HFClientTGI.md
+++ b/docs/docs/deep-dive/language_model_clients/lm_local_models/HFClientTGI.md
@@ -81,7 +81,7 @@ The constructor initializes the `HFModel` base class to support the handling of
- `model` (_str_): ID of Hugging Face model connected to the TGI server.
- `port` (_int_ or _list_): Port for communicating to the TGI server. This can be a single port number (`8080`) or a list of TGI ports (`[8080, 8081, 8082]`) to route the requests to.
- `url` (_str_): Base URL of hosted TGI server. This will often be `"http://localhost"`.
-- `http_request_kwargs` (_dict_): Dictionary of additional keyword agruments to pass to the HTTP request function to the TGI server. This is `None` by default.
+- `http_request_kwargs` (_dict_): Dictionary of additional keyword arguments to pass to the HTTP request function to the TGI server. This is `None` by default.
- `**kwargs`: Additional keyword arguments to configure the TGI client.
Example of the TGI constructor:
diff --git a/docs/docs/deep-dive/modules/predict.md b/docs/docs/deep-dive/modules/predict.md
index 46f0a09ee..b5169ea12 100644
--- a/docs/docs/deep-dive/modules/predict.md
+++ b/docs/docs/deep-dive/modules/predict.md
@@ -46,7 +46,7 @@ class Predict(Parameter):
This method serves as a wrapper for the `forward` method. It allows making predictions using the `Predict` class by providing keyword arguments.
-**Paramters:**
+**Parameters:**
- `**kwargs`: Keyword arguments required for prediction.
**Returns:**
diff --git a/docs/docs/deep-dive/optimizers/bfrs.md b/docs/docs/deep-dive/optimizers/bfrs.md
index 8728398a3..c15004af1 100644
--- a/docs/docs/deep-dive/optimizers/bfrs.md
+++ b/docs/docs/deep-dive/optimizers/bfrs.md
@@ -23,7 +23,7 @@ In terms of API `BootstrapFewShotWithRandomSearch` teleprompter is quite similar
## Working Example
-Let's take the example of optimizing a simple CoT pipeline for GSM8k dataset, we'll take the example in [BootstrapFewShot](/deep-dive/optimizers/bootstrap-fewshot) as our running example for optimizers. We're gonna assume our data and pipeline is same as the on in `BootstrapFewShot` article. So let's start by intializing the optimizer:
+Let's take the example of optimizing a simple CoT pipeline for GSM8k dataset, we'll take the example in [BootstrapFewShot](/deep-dive/optimizers/bootstrap-fewshot) as our running example for optimizers. We're gonna assume our data and pipeline is same as the on in `BootstrapFewShot` article. So let's start by initializing the optimizer:
```python
from dspy.teleprompt import BootstrapFewShotWithRandomSearch
diff --git a/docs/docs/deep-dive/optimizers/miprov2.md b/docs/docs/deep-dive/optimizers/miprov2.md
index 142eef084..4d60167cf 100644
--- a/docs/docs/deep-dive/optimizers/miprov2.md
+++ b/docs/docs/deep-dive/optimizers/miprov2.md
@@ -88,7 +88,7 @@ optimized_program = teleprompter.compile(
optimized_program.save(f"mipro_optimized")
# Evaluate optimized program
-print(f"Evluate optimized program...")
+print(f"Evaluate optimized program...")
evaluate(optimized_program, devset=devset[:])
```
@@ -119,7 +119,7 @@ zeroshot_optimized_program = teleprompter.compile(
zeroshot_optimized_program.save(f"mipro_zeroshot_optimized")
# Evaluate optimized program
-print(f"Evluate optimized program...")
+print(f"Evaluate optimized program...")
evaluate(zeroshot_optimized_program, devset=devset[:])
```
@@ -156,7 +156,7 @@ optimized_program = teleprompter.compile(
optimized_program.save(f"mipro_optimized")
# Evaluate optimized program
-print(f"Evluate optimized program...")
+print(f"Evaluate optimized program...")
evaluate(optimized_program, devset=devset[:])
```
@@ -170,7 +170,7 @@ evaluate(optimized_program, devset=devset[:])
| `prompt_model` | `dspy.LM` | LM specified in `dspy.settings` | Model used for prompt generation. |
| `task_model` | `dspy.LM` | LM specified in `dspy.settings` | Model used for task execution. |
| `auto` | `Optional[str]` | None | If set to `light`, `medium`, or `heavy`, this will automatically configure the following hyperparameters: `num_candidates`, `num_trials`, `minibatch`, and will also cap the size of `valset` up to 100, 300, and 1000 for `light`, `medium`, and `heavy` runs respectively. |
-| `num_candidates` | `int` | `10` | Number of candidate instructions & few-shot examples to generate and evaluate for each predictor. If `num_candidates=10`, this means for a 2 module LM program we'll be optimizing over 10 candidates x 2 modules x 2 variables (few-shot ex. and instructions for each module)= 40 total variables. Therfore, if we increase `num_candidates`, we will probably want to increase `num_trials` as well (see Compile parameters). |
+| `num_candidates` | `int` | `10` | Number of candidate instructions & few-shot examples to generate and evaluate for each predictor. If `num_candidates=10`, this means for a 2 module LM program we'll be optimizing over 10 candidates x 2 modules x 2 variables (few-shot ex. and instructions for each module)= 40 total variables. Therefore, if we increase `num_candidates`, we will probably want to increase `num_trials` as well (see Compile parameters). |
| `num_threads` | `int` | `6` | Threads to use for evaluation. |
| `max_errors` | `int` | `10` | Maximum errors during an evaluation run that can be made before throwing an Exception. |
| `teacher_settings` | `dict` | `{}` | Settings to use for the teacher model that bootstraps few-shot examples. An example dict would be `{lm=<dspy.LM object>}`. If your LM program with your default model is struggling to bootstrap any examples, it could be worth using a more powerful teacher model for bootstrapping. |
@@ -210,7 +210,7 @@ At a high level, `MIPROv2` works by creating both few-shot examples and new inst
These steps are broken down in more detail below:
1) **Bootstrap Few-Shot Examples**: The same bootstrapping technique used in `BootstrapFewshotWithRandomSearch` is used to create few-shot examples. This works by randomly sampling examples from your training set, which are then run through your LM program. If the output from the program is correct for this example, it is kept as a valid few-shot example candidate. Otherwise, we try another example until we've curated the specified amount of few-shot example candidates. This step creates `num_candidates` sets of `max_bootstrapped_demos` bootstrapped examples and `max_labeled_demos` basic examples sampled from the training set.
2) **Propose Instruction Candidates**. Next, we propose instruction candidates for each predictor in the program. This is done using another LM program as a proposer, which bootstraps & summarizes relevant information about the task to generate high quality instructions. Specifically, the instruction proposer includes (1) a generated summary of properties of the training dataset, (2) a generated summary of your LM program's code and the specific predictor that an instruction is being generated for, (3) the previously bootstrapped few-shot examples to show reference inputs / outputs for a given predictor and (4) a randomly sampled tip for generation (i.e. "be creative", "be concise", etc.) to help explore the feature space of potential instructions.
-3. **Find an Optimized Combination of Few-Shot Examples & Instructions**. Finally, now that we've created these few-shot examples and instructions, we use Bayesian Optimization to choose which set of these would work best for each predictor in our program. This works by running a series of `num_trials` trials, where a new set of prompts are evaluated over our validation set at each trial. This helps the Bayesian Optimizer learn which combination of prompts work best over time. If `minibatch` is set to `True` (which it is by default), then the new set of prompts are only evaluated on a minibatch of size `minibatch_size` at each trial which generally allows for more efficient exploration / exploitation. The best averaging set of prompts is then evalauted on the full validation set every `minibatch_full_eval_steps` get a less noisey performance benchmark. At the end of the optimization process, the LM program with the set of prompts that performed best on the full validation set is returned.
+3. **Find an Optimized Combination of Few-Shot Examples & Instructions**. Finally, now that we've created these few-shot examples and instructions, we use Bayesian Optimization to choose which set of these would work best for each predictor in our program. This works by running a series of `num_trials` trials, where a new set of prompts are evaluated over our validation set at each trial. This helps the Bayesian Optimizer learn which combination of prompts work best over time. If `minibatch` is set to `True` (which it is by default), then the new set of prompts are only evaluated on a minibatch of size `minibatch_size` at each trial which generally allows for more efficient exploration / exploitation. The best averaging set of prompts is then evaluated on the full validation set every `minibatch_full_eval_steps` get a less noisey performance benchmark. At the end of the optimization process, the LM program with the set of prompts that performed best on the full validation set is returned.
For those interested in more details, more information on `MIPROv2` along with a study on `MIPROv2` compared with other DSPy optimizers can be found in [this paper](https://arxiv.org/abs/2406.11695).
\ No newline at end of file
diff --git a/docs/docs/deep-dive/retrieval_models_clients/MilvusRM.md b/docs/docs/deep-dive/retrieval_models_clients/MilvusRM.md
index 8e290eb48..acd3cd4ea 100644
--- a/docs/docs/deep-dive/retrieval_models_clients/MilvusRM.md
+++ b/docs/docs/deep-dive/retrieval_models_clients/MilvusRM.md
@@ -51,7 +51,7 @@ Search the Milvus collection for the top `k` passages matching the given query o
from dspy.retrieve.milvus_rm import MilvusRM
import os
-os.envrion["OPENAI_API_KEY"] = "<YOUR_OPENAI_API_KEY>"
+os.environ["OPENAI_API_KEY"] = "<YOUR_OPENAI_API_KEY>"
retriever_model = MilvusRM(
collection_name="<YOUR_COLLECTION_NAME>",
diff --git a/docs/docs/deep-dive/retrieval_models_clients/SnowflakeRM.md b/docs/docs/deep-dive/retrieval_models_clients/SnowflakeRM.md
index e6689ea07..28211afe5 100644
--- a/docs/docs/deep-dive/retrieval_models_clients/SnowflakeRM.md
+++ b/docs/docs/deep-dive/retrieval_models_clients/SnowflakeRM.md
@@ -64,7 +64,7 @@ connection_parameters = {
snowpark = Session.builder.configs(connection_parameters).create()
snowflake_retriever = SnowflakeRM(snowflake_session=snowpark,
- cortex_search_service="<YOUR_CORTEX_SERACH_SERVICE_NAME>",
+ cortex_search_service="<YOUR_CORTEX_SEARCH_SERVICE_NAME>",
snowflake_database="<YOUR_SNOWFLAKE_DATABASE_NAME>",
snowflake_schema="<YOUR_SNOWFLAKE_SCHEMA_NAME>",
auto_filter=True,
diff --git a/docs/docs/dspy-usecases.md b/docs/docs/dspy-usecases.md
index a8cfc6208..e791d0ca9 100644
--- a/docs/docs/dspy-usecases.md
+++ b/docs/docs/dspy-usecases.md
@@ -58,27 +58,27 @@ WIP. This list mainly includes companies that have public posts or have OKed bei
| **Name** | **Description/Link** |
|---|---|
-| **Stanford CS 224U Homework** | [Github](https://github.com/cgpotts/cs224u/blob/main/hw_openqa.ipynb) |
-| **STORM Report Generation (10,000 GitHub stars)** | [Github](https://github.com/stanford-oval/storm) |
-| **DSPy Redteaming** | [Github](https://github.com/haizelabs/dspy-redteam) |
-| **DSPy Theory of Mind** | [Github](https://github.com/plastic-labs/dspy-opentom) |
-| **Indic cross-lingual Natural Language Inference** | [Github](https://github.com/saifulhaq95/DSPy-Indic/blob/main/indicxlni.ipynb) |
-| **Optimizing LM for Text2SQL using DSPy** | [Github](https://github.com/jjovalle99/DSPy-Text2SQL) |
+| **Stanford CS 224U Homework** | [GitHub](https://github.com/cgpotts/cs224u/blob/main/hw_openqa.ipynb) |
+| **STORM Report Generation (10,000 GitHub stars)** | [GitHub](https://github.com/stanford-oval/storm) |
+| **DSPy Redteaming** | [GitHub](https://github.com/haizelabs/dspy-redteam) |
+| **DSPy Theory of Mind** | [GitHub](https://github.com/plastic-labs/dspy-opentom) |
+| **Indic cross-lingual Natural Language Inference** | [GitHub](https://github.com/saifulhaq95/DSPy-Indic/blob/main/indicxlni.ipynb) |
+| **Optimizing LM for Text2SQL using DSPy** | [GitHub](https://github.com/jjovalle99/DSPy-Text2SQL) |
| **DSPy PII Masking Demo by Eric Ness** | [Colab](https://colab.research.google.com/drive/1KZR1sGTp_RLWUJPAiK1FKPKI-Qn9neUm?usp=sharing) |
-| **DSPy on BIG-Bench Hard Example** | [Github](https://drchrislevy.github.io/posts/dspy/dspy.html) |
-| **Building a chess playing agent using DSPy** | [Github](https://medium.com/thoughts-on-machine-learning/building-a-chess-playing-agent-using-dspy-9b87c868f71e) |
-| **Ittia Research Fact Checking** | [Github](https://github.com/ittia-research/check) |
-| **Strategic Debate via Tree-of-Thought** | [Github](https://github.com/zbambergerNLP/strategic-debate-tot) |
-| **Sanskrit to English Translation App**| [Github](https://github.com/ganarajpr/sanskrit-translator-dspy) |
-| **DSPy for extracting features from PDFs on arXiv**| [Github](https://github.com/S1M0N38/dspy-arxiv) |
-| **DSPygen: DSPy in Ruby on Rails**| [Github](https://github.com/seanchatmangpt/dspygen) |
-| **DSPy Inspector**| [Github](https://github.com/Neoxelox/dspy-inspector) |
-| **DSPy with FastAPI**| [Github](https://github.com/diicellman/dspy-rag-fastapi) |
-| **DSPy for Indian Languages**| [Github](https://github.com/saifulhaq95/DSPy-Indic) |
-| **Hurricane: Blog Posts with Generative Feedback Loops!**| [Github](https://github.com/weaviate-tutorials/Hurricane) |
-| **RAG example using DSPy, Gradio, FastAPI, and Ollama**| [Github](https://github.com/diicellman/dspy-gradio-rag) |
-| **Synthetic Data Generation**| [Github](https://colab.research.google.com/drive/1CweVOu0qhTC0yOfW5QkLDRIKuAuWJKEr?usp=sharing) |
-| **Self Discover**| [Github](https://colab.research.google.com/drive/1GkAQKmw1XQgg5UNzzy8OncRe79V6pADB?usp=sharing) |
+| **DSPy on BIG-Bench Hard Example** | [GitHub](https://drchrislevy.github.io/posts/dspy/dspy.html) |
+| **Building a chess playing agent using DSPy** | [GitHub](https://medium.com/thoughts-on-machine-learning/building-a-chess-playing-agent-using-dspy-9b87c868f71e) |
+| **Ittia Research Fact Checking** | [GitHub](https://github.com/ittia-research/check) |
+| **Strategic Debate via Tree-of-Thought** | [GitHub](https://github.com/zbambergerNLP/strategic-debate-tot) |
+| **Sanskrit to English Translation App**| [GitHub](https://github.com/ganarajpr/sanskrit-translator-dspy) |
+| **DSPy for extracting features from PDFs on arXiv**| [GitHub](https://github.com/S1M0N38/dspy-arxiv) |
+| **DSPygen: DSPy in Ruby on Rails**| [GitHub](https://github.com/seanchatmangpt/dspygen) |
+| **DSPy Inspector**| [GitHub](https://github.com/Neoxelox/dspy-inspector) |
+| **DSPy with FastAPI**| [GitHub](https://github.com/diicellman/dspy-rag-fastapi) |
+| **DSPy for Indian Languages**| [GitHub](https://github.com/saifulhaq95/DSPy-Indic) |
+| **Hurricane: Blog Posts with Generative Feedback Loops!**| [GitHub](https://github.com/weaviate-tutorials/Hurricane) |
+| **RAG example using DSPy, Gradio, FastAPI, and Ollama**| [GitHub](https://github.com/diicellman/dspy-gradio-rag) |
+| **Synthetic Data Generation**| [GitHub](https://colab.research.google.com/drive/1CweVOu0qhTC0yOfW5QkLDRIKuAuWJKEr?usp=sharing) |
+| **Self Discover**| [GitHub](https://colab.research.google.com/drive/1GkAQKmw1XQgg5UNzzy8OncRe79V6pADB?usp=sharing) |
TODO: This list in particular is highly incomplete. There are a couple dozen other good ones.
diff --git a/docs/docs/quick-start/getting-started-01.md b/docs/docs/quick-start/getting-started-01.md
index b276210ed..bd65cd5f2 100644
--- a/docs/docs/quick-start/getting-started-01.md
+++ b/docs/docs/quick-start/getting-started-01.md
@@ -167,7 +167,7 @@ pred = cot(**example.inputs())
score = metric(example, pred)
print(f"Question: \t {example.question}\n")
-print(f"Gold Reponse: \t {example.response}\n")
+print(f"Gold Response: \t {example.response}\n")
print(f"Predicted Response: \t {pred.response}\n")
print(f"Semantic F1 Score: {score:.2f}")
```
@@ -176,7 +176,7 @@ print(f"Semantic F1 Score: {score:.2f}")
```
Question: what are high memory and low memory on linux?
-Gold Reponse: "High Memory" refers to the application or user space, the memory that user programs can use and which isn't permanently mapped in the kernel's space, while "Low Memory" is the kernel's space, which the kernel can address directly and is permanently mapped.
+Gold Response: "High Memory" refers to the application or user space, the memory that user programs can use and which isn't permanently mapped in the kernel's space, while "Low Memory" is the kernel's space, which the kernel can address directly and is permanently mapped.
The user cannot access the Low Memory as it is set aside for the required kernel programs.
Predicted Response: In Linux, "low memory" refers to the memory that is directly accessible by the kernel and user processes, typically the first 4GB on a 32-bit system. "High memory" refers to memory above this limit, which is not directly accessible by the kernel in a 32-bit environment. This distinction is crucial for memory management, particularly in systems with large amounts of RAM, as it influences how memory is allocated and accessed.
diff --git a/docs/mkdocs.yml b/docs/mkdocs.yml
index a98084ba6..e487236b9 100644
--- a/docs/mkdocs.yml
+++ b/docs/mkdocs.yml
@@ -33,7 +33,7 @@ nav:
- MultiChainComparison: deep-dive/modules/multi-chain-comparison.md
- ProgramOfThought: deep-dive/modules/program-of-thought.md
- Assertions: deep-dive/assertions.md
- - Retreive: deep-dive/modules/retrieve.md
+ - Retrieve: deep-dive/modules/retrieve.md
- Modules Guide: deep-dive/modules/guide.md
- Metrics and Assertions:
- Overview: building-blocks/7-assertions.md
diff --git a/dsp/adapters/utils.py b/dsp/adapters/utils.py
index 4bc15c058..1cbe604d4 100644
--- a/dsp/adapters/utils.py
+++ b/dsp/adapters/utils.py
@@ -53,7 +53,7 @@ def format_answers(answers: Union[str, list]) -> Optional[str]:
ValueError: when is not of type list or str
Returns:
- _type_: Optiona[str]
+ _type_: Optional[str]
"""
if isinstance(answers, list):
if len(answers) >= 1:
diff --git a/dsp/modules/dummy_lm.py b/dsp/modules/dummy_lm.py
index 49f35fa72..0e8c25fda 100644
--- a/dsp/modules/dummy_lm.py
+++ b/dsp/modules/dummy_lm.py
@@ -87,5 +87,5 @@ class DSPDummyLM(LM):
return [choice["text"] for choice in choices]
def get_convo(self, index) -> str:
- """Get the prompt + anwer from the ith message."""
+ """Get the prompt + answer from the ith message."""
return self.history[index]["prompt"] + " " + self.history[index]["response"]["choices"][0]["text"]
diff --git a/dsp/modules/groq_client.py b/dsp/modules/groq_client.py
index e22c505ac..196a2b726 100644
--- a/dsp/modules/groq_client.py
+++ b/dsp/modules/groq_client.py
@@ -96,7 +96,7 @@ class GroqLM(LM):
on_backoff=backoff_hdlr,
)
def request(self, prompt: str, **kwargs):
- """Handles retreival of model completions whilst handling rate limiting and caching."""
+ """Handles retrieval of model completions whilst handling rate limiting and caching."""
if "model_type" in kwargs:
del kwargs["model_type"]
@@ -106,7 +106,7 @@ class GroqLM(LM):
return choice.message.content
def chat_request(self, **kwargs):
- """Handles retreival of model completions whilst handling rate limiting and caching."""
+ """Handles retrieval of model completions whilst handling rate limiting and caching."""
response = self.client.chat.completions.create(**kwargs)
return response
diff --git a/dsp/modules/hf.py b/dsp/modules/hf.py
index 57f57d259..fc16c2ae9 100644
--- a/dsp/modules/hf.py
+++ b/dsp/modules/hf.py
@@ -49,7 +49,7 @@ class HFModel(LM):
checkpoint (str, optional): load specific checkpoints of the model. Defaults to None.
is_client (bool, optional): whether to access models via client. Defaults to False.
hf_device_map (str, optional): HF config strategy to load the model.
- Recommeded to use "auto", which will help loading large models using accelerate. Defaults to "auto".
+ Recommended to use "auto", which will help loading large models using accelerate. Defaults to "auto".
model_kwargs (dict, optional): additional kwargs to pass to the model constructor. Defaults to empty dict.
"""
diff --git a/dsp/modules/premai.py b/dsp/modules/premai.py
index 17e9d151f..15968c7c5 100644
--- a/dsp/modules/premai.py
+++ b/dsp/modules/premai.py
@@ -63,7 +63,7 @@ class PremAI(LM):
api_key: Optional[str]
Prem AI API key, to connect with the API. If not provided then it will check from env var by the name
PREMAI_API_KEY
- kwargs: Optional[dict] For any additional paramters
+ kwargs: Optional[dict] For any additional parameters
"""
self.model = "default" if model is None else model
super().__init__(self.model)
diff --git a/dsp/modules/sentence_vectorizer.py b/dsp/modules/sentence_vectorizer.py
index 84be68420..61788eb98 100644
--- a/dsp/modules/sentence_vectorizer.py
+++ b/dsp/modules/sentence_vectorizer.py
@@ -211,7 +211,7 @@ class OpenAIVectorizer(BaseSentenceVectorizer):
class FastEmbedVectorizer(BaseSentenceVectorizer):
- """Sentence vectorizer implementaion using FastEmbed - https://qdrant.github.io/fastembed."""
+ """Sentence vectorizer implementation using FastEmbed - https://qdrant.github.io/fastembed."""
def __init__(
self,
diff --git a/dsp/primitives/demonstrate.py b/dsp/primitives/demonstrate.py
index 331c31517..9b3ff9584 100644
--- a/dsp/primitives/demonstrate.py
+++ b/dsp/primitives/demonstrate.py
@@ -155,7 +155,7 @@ def knn(
Args:
train: a bunch of questions to put in index & search later
- cast: function that contructs text before vectorization. By default,
+ cast: function that constructs text before vectorization. By default,
it uses only question. Check `cast_naive_get_question_and_answer` for more details.
n_probe: number of closest IVF-clusters to check for neighbours.
Doesn't affect bruteforce-based search.
diff --git a/dsp/utils/ann_utils.py b/dsp/utils/ann_utils.py
index dcd3f09ce..a5378d3f7 100644
--- a/dsp/utils/ann_utils.py
+++ b/dsp/utils/ann_utils.py
@@ -100,12 +100,12 @@ def create_faiss_index(
the difference between a vector and the reconstruction that can be
decoded from its representation in the index.
in_list_dist_type: type of distance to calculate simmilarities within one IVF.
- Can be `IP` (for inner product) or `L2` distance. Case insensetive.
+ Can be `IP` (for inner product) or `L2` distance. Case insensitive.
If the index type is bruteforce (`n_objects` < 20_000), this variable will define
- the distane type for that bruteforce index. `centroid_dist_type` will be ignored.
+ the distance type for that bruteforce index. `centroid_dist_type` will be ignored.
centroid_dist_type: type of distance to calculate simmilarities between a query
and cluster centroids. Can be `IP` (for inner product) or `L2` distance.
- Case insensetive.
+ Case insensitive.
Returns: untrained FAISS-index
"""
if n_objects < 20_000:
diff --git a/dspy/clients/openai.py b/dspy/clients/openai.py
index 0522475f8..f5d08c283 100644
--- a/dspy/clients/openai.py
+++ b/dspy/clients/openai.py
@@ -33,7 +33,7 @@ def is_openai_model(model: str) -> bool:
if model in valid_model_names:
return True
- # Check if the model is a fine-tuned OpneAI model. Fine-tuned OpenAI models
+ # Check if the model is a fine-tuned OpenAI model. Fine-tuned OpenAI models
# have the prefix "ft:<BASE_MODEL_NAME>:", followed by a string specifying
# the fine-tuned model. The following RegEx pattern is used to match the
# base model name.
diff --git a/dspy/experimental/synthesizer/synthesizer.py b/dspy/experimental/synthesizer/synthesizer.py
index bd70e6d7a..2fea0f85d 100644
--- a/dspy/experimental/synthesizer/synthesizer.py
+++ b/dspy/experimental/synthesizer/synthesizer.py
@@ -244,17 +244,17 @@ class Synthesizer:
return data
def export(self, data: List[dspy.Example], path: str, mode: str = None, **kwargs):
- extention = mode or path.split(".")[-1]
+ extension = mode or path.split(".")[-1]
dataset = Dataset.from_list(
[example.toDict() for example in data],
)
- if extention == "csv":
+ if extension == "csv":
dataset.to_csv(path_or_buf=path, **kwargs)
- elif extention == "json":
+ elif extension == "json":
dataset.to_json(path_or_buf=path, **kwargs)
- elif extention == "arrow" or extention == "hf":
+ elif extension == "arrow" or extension == "hf":
dataset.save_to_disk(path)
diff --git a/dspy/experimental/synthetic_data.py b/dspy/experimental/synthetic_data.py
index 16bc5e797..9f965bcba 100644
--- a/dspy/experimental/synthetic_data.py
+++ b/dspy/experimental/synthetic_data.py
@@ -41,7 +41,7 @@ class SyntheticDataGenerator:
def _define_or_infer_fields(self):
"""Define fields to generate if a schema class is provided.
- Infer fields to generate if an inital sample of examples is provided.
+ Infer fields to generate if an initial sample of examples is provided.
Returns:
dict: dictionary of fields to generate
diff --git a/dspy/functional/functional.py b/dspy/functional/functional.py
index 8bb826044..03d859763 100644
--- a/dspy/functional/functional.py
+++ b/dspy/functional/functional.py
@@ -62,7 +62,7 @@ class _StripOutput(dspy.Module):
class FunctionalModule(dspy.Module):
- """To use the @cot and @predictor decorators, your module needs to inheret form this class."""
+ """To use the @cot and @predictor decorators, your module needs to inherit form this class."""
def __init__(self):
super().__init__()
@@ -208,7 +208,7 @@ class TypedPredictor(dspy.Module):
task_description: str = dspy.InputField(desc="What I asked the model to do")
language_model_output: str = dspy.InputField(desc="The output of the model")
- error: str = dspy.InputField(desc="The validation error trigged by the models output")
+ error: str = dspy.InputField(desc="The validation error triggered by the models output")
explanation: str = dspy.OutputField(desc="Explain what the model did wrong")
advice: str = dspy.OutputField(
desc="Instructions for the model to do better next time. A single paragraph.",
diff --git a/dspy/retrieve/snowflake_rm.py b/dspy/retrieve/snowflake_rm.py
index f58e205b7..7485a1954 100644
--- a/dspy/retrieve/snowflake_rm.py
+++ b/dspy/retrieve/snowflake_rm.py
@@ -17,14 +17,14 @@ except ImportError:
class SnowflakeRM(dspy.Retrieve):
- """A retrieval module that uses Snowlfake's Cortex Search service to return the top relevant passages for a given query.
+ """A retrieval module that uses Snowflake's Cortex Search service to return the top relevant passages for a given query.
Assumes that a Snowflake Cortex Search endpoint has been configured by the use.
For more information on configuring the Cortex Search service, visit: https://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-search/cortex-search-overview
Args:
- snowflake_sesssion (object): Snowflake Snowpark session for accessing the service.
+ snowflake_session (object): Snowflake Snowpark session for accessing the service.
cortex_search_service(str): Name of the Cortex Search service to be used.
snowflake_database (str): The name of the Snowflake table containing document embeddings.
snowflake_schema (str): The name of the Snowflake table containing document embeddings.
@@ -241,7 +241,7 @@ class GenerateFilter(dspy.Signature):
Sample Values: {"industry":["biotechnology","healthcare","agriculture"],"HQ":["NY, US","CA,US","FL,US"],"date":["01/01,1999","01/01/2024"]}
Answer: {"@or":[{"@eq":{"year":"2021"}},{"@eq":{"year":"2022"}},{"@eq":{"year":"2023"}},{"@eq":{"year":"2024"}}]}
- Query: Wha is the sentiment of Biotech CEO's of companies based in New York?
+ Query: What is the sentiment of Biotech CEO's of companies based in New York?
Attributes: industry,hq,date
Sample Values: {"industry":["biotechnology","healthcare","agriculture"],"HQ":["NY, US","CA,US","FL,US"],"date":["01/01,1999","01/01/2024"]}
Answer: {"@and": [ { "@eq": { "industry"": "biotechnology" } }, { "@eq": { "HQ": "NY,US" } }]}
diff --git a/dspy/signatures/signature.py b/dspy/signatures/signature.py
index 4ad7bab1b..88812be0c 100644
--- a/dspy/signatures/signature.py
+++ b/dspy/signatures/signature.py
@@ -141,7 +141,7 @@ class SignatureMeta(type(BaseModel)):
return cls.insert(-1, name, field, type_)
def insert(cls, index: int, name: str, field, type_: Type = None) -> Type["Signature"]:
- # It's posisble to set the type as annotation=type in pydantic.Field(...)
+ # It's possible to set the type as annotation=type in pydantic.Field(...)
# But this may be annoying for users, so we allow them to pass the type
if type_ is None:
type_ = field.annotation
diff --git a/dspy/teleprompt/finetune_teleprompter.py b/dspy/teleprompt/finetune_teleprompter.py
index 315434aba..57246fea6 100644
--- a/dspy/teleprompt/finetune_teleprompter.py
+++ b/dspy/teleprompt/finetune_teleprompter.py
@@ -62,10 +62,10 @@ def convert_to_module_level_message_data(
prompt_completion_data = []
for data_dict in data:
trace = data_dict["trace"]
- trace_prompt_comletion_data = build_messages_from_trace(
+ trace_prompt_completion_data = build_messages_from_trace(
trace=trace, exclude_demos=exclude_demos, try_to_record_lm_kwargs=try_to_record_lm_kwargs, program=program
)
- for prompt_completion_dict in trace_prompt_comletion_data:
+ for prompt_completion_dict in trace_prompt_completion_data:
if keep_data_keys:
prompt_completion_dict = {**data_dict, **prompt_completion_dict}
prompt_completion_data.append(prompt_completion_dict)
diff --git a/dspy/teleprompt/mipro_optimizer.py b/dspy/teleprompt/mipro_optimizer.py
index f0d96cc13..128ede37e 100644
--- a/dspy/teleprompt/mipro_optimizer.py
+++ b/dspy/teleprompt/mipro_optimizer.py
@@ -419,7 +419,7 @@ class MIPRO(Teleprompter):
module = student.deepcopy()
evaluate = Evaluate(devset=trainset, metric=self.metric, **eval_kwargs)
- # In the case where the bootstrapped and labeled demos are set to 0, we'll stil bootstrap examples to use in our meta prompt
+ # In the case where the bootstrapped and labeled demos are set to 0, we'll still bootstrap examples to use in our meta prompt
if (
max_bootstrapped_demos == 0 and max_labeled_demos == 0
): # TODO: address case when max_bootstrapped alone is 0
diff --git a/dspy/teleprompt/signature_opt_typed.py b/dspy/teleprompt/signature_opt_typed.py
index 3e5c7587f..cc1bb341c 100644
--- a/dspy/teleprompt/signature_opt_typed.py
+++ b/dspy/teleprompt/signature_opt_typed.py
@@ -19,7 +19,7 @@ from dspy.signatures.field import InputField, OutputField
def make_info(signature: type[Signature]) -> BaseModel:
"""Creates a SignatureInfo pydantic type, that describes the Signature.
- Returns an instnce of this type, with the instructions and field descriptions of the input type.
+ Returns an instance of this type, with the instructions and field descriptions of the input type.
"""
# First, create the SignatureInfo type
fields = {
@@ -82,7 +82,7 @@ def make_initial_signature(n_prompts: int) -> type[Signature]:
- You are an expert mathematician.
- You are a professor of mathematics.
Task Descriptions:
- - Be consise in your answer.
+ - Be concise in your answer.
- Be as clear as possible.
- Use lots of creativity.
Closers:
diff --git a/dspy/utils/dummies.py b/dspy/utils/dummies.py
index 90f5def66..39ea37c61 100644
--- a/dspy/utils/dummies.py
+++ b/dspy/utils/dummies.py
@@ -95,7 +95,7 @@ class DSPDummyLM(DSPLM):
return [choice["text"] for choice in choices]
def get_convo(self, index) -> str:
- """Get the prompt + anwer from the ith message."""
+ """Get the prompt + answer from the ith message."""
return self.history[index]["prompt"] + " " + self.history[index]["response"]["choices"][0]["text"]
@@ -209,7 +209,7 @@ class DummyLM(LM):
return outputs
def get_convo(self, index):
- """Get the prompt + anwer from the ith message."""
+ """Get the prompt + answer from the ith message."""
return self.history[index]["messages"], self.history[index]["outputs"]
diff --git a/internals/build-and-release.md b/internals/build-and-release.md
index 76b3b400d..fb000ed61 100644
--- a/internals/build-and-release.md
+++ b/internals/build-and-release.md
@@ -56,4 +56,4 @@ Builds and publishes the package to pypi.
1. Publishes the package to pypi.
-\* The package name is updated by the worfklow to allow the same files to be used to build both the pypi and test-pypi packages.
\ No newline at end of file
+\* The package name is updated by the workflow to allow the same files to be used to build both the pypi and test-pypi packages.
\ No newline at end of file
diff --git a/internals/release-checklist.md b/internals/release-checklist.md
index 862ab7a5c..0213a52b7 100644
--- a/internals/release-checklist.md
+++ b/internals/release-checklist.md
@@ -9,10 +9,10 @@
* [ ] Confirm the tests pass and the package has been published to pypi.
* If the tests fail, you can remove the tag from your local and github repo using:
```bash
- git push origin --delete X.Y.Z # Delete on Github
+ git push origin --delete X.Y.Z # Delete on GitHub
git tag -d X.Y.Z # Delete locally
```
- * Fix the errors and then repeat the steps above to recreate the tag locally and push to Github to restart the process.
+ * Fix the errors and then repeat the steps above to recreate the tag locally and push to GitHub to restart the process.
* Note that the github action takes care of incrementing the release version on test-pypi automatically by adding a pre-release identifier in the scenario where the tests fail and you need to delete and push the same tag again.
* [ ] [Create a release](https://docs.github.com/en/repositories/releasing-projects-on-github/managing-releases-in-a-repository)
* [ ] Add release notes. You can make use of [automatically generated release notes](https://docs.github.com/en/repositories/releasing-projects-on-github/automatically-generated-release-notes)
diff --git a/testing/README.md b/testing/README.md
index 6fc730072..78441af70 100644
--- a/testing/README.md
+++ b/testing/README.md
@@ -12,7 +12,7 @@ from optimizer_tester import OptimizerTester
tester = OptimizerTester()
```
-The default verison (no parameters) expects a llama model hosted on ports [7140, 7141, 7142, 7143] and OpenAI keys stored in a .env file (OPENAI_API_KEY and OPENAI_API_BASE).
+The default version (no parameters) expects a llama model hosted on ports [7140, 7141, 7142, 7143] and OpenAI keys stored in a .env file (OPENAI_API_KEY and OPENAI_API_BASE).
If you prefer to specify your own model parameters then you can pass models into the OptimizerTester
diff --git a/testing/tasks/heart_disease.py b/testing/tasks/heart_disease.py
index f64026172..0885f475c 100644
--- a/testing/tasks/heart_disease.py
+++ b/testing/tasks/heart_disease.py
@@ -53,7 +53,7 @@ class HeartDiseaseInput(dspy.Signature):
trestbps = dspy.InputField(
desc="Resting blood pressure (in mm Hg on admission to the hospital)"
)
- chol = dspy.InputField(desc="Serum cholestoral in mg/dl")
+ chol = dspy.InputField(desc="Serum cholesterol in mg/dl")
fbs = dspy.InputField(desc="Fasting blood sugar > 120 mg/dl (true or false)")
restecg = dspy.InputField(
desc="Resting electrocardiographic results (normal, ST-T wave abnormality, left ventricular hypertrophy)"
diff --git a/tests/dsp_LM/functional/test_functional.py b/tests/dsp_LM/functional/test_functional.py
index 5e5274567..e71b41fc0 100644
--- a/tests/dsp_LM/functional/test_functional.py
+++ b/tests/dsp_LM/functional/test_functional.py
@@ -530,7 +530,7 @@ def test_literal():
assert f() == "2"
-def test_literal_missmatch():
+def test_literal_mismatch():
lm = DSPDummyLM([f'"{i}"' for i in range(5, 100)])
dspy.settings.configure(lm=lm)
@@ -555,7 +555,7 @@ def test_literal_int():
assert f() == 2
-def test_literal_int_missmatch():
+def test_literal_int_mismatch():
lm = DSPDummyLM([f"{i}" for i in range(5, 100)])
dspy.settings.configure(lm=lm)
@@ -893,7 +893,7 @@ def test_model_validator():
category: str = dspy.OutputField()
@model_validator(mode="after")
- def check_cateogry(self):
+ def check_category(self):
if self.category not in self.allowed_categories:
raise ValueError(f"category not in {self.allowed_categories}")
return self
diff --git a/tests/dsp_LM/predict/test_react.py b/tests/dsp_LM/predict/test_react.py
index 607105aba..6c8bbf70e 100644
--- a/tests/dsp_LM/predict/test_react.py
+++ b/tests/dsp_LM/predict/test_react.py
@@ -37,7 +37,7 @@ from dspy.utils.dummies import DSPDummyLM, dummy_rm
# def test_example_search():
-# # Createa a simple dataset which the model will use with the Retrieve tool.
+# # Create a simple dataset which the model will use with the Retrieve tool.
# lm = DSPDummyLM(
# [
# "Initial thoughts", # Thought_1
@@ -49,7 +49,7 @@ from dspy.utils.dummies import DSPDummyLM, dummy_rm
# rm = dummy_rm(
# [
# "We all know the color of the sky is blue.",
-# "Somethng about the sky colors",
+# "Something about the sky colors",
# "This sentence is completely irellevant to answer the question.",
# "Let's add some more sentences to act as summy passages.",
# "Let's add some more sentences to act as summy passages.",
@@ -79,7 +79,7 @@ from dspy.utils.dummies import DSPDummyLM, dummy_rm
# "Action 1: Search[the color of the sky]\n\n"
# "Observation 1:\n"
# "[1] «We all know the color of the sky is blue.»\n"
-# "[2] «Somethng about the sky colors»\n"
+# "[2] «Something about the sky colors»\n"
# "[3] «This sentence is completely irellevant to answer the question.»\n\n"
# "Thought 2: More thoughts\n\n"
# "Action 2: finish[blue]"
@@ -151,4 +151,4 @@ from dspy.utils.dummies import DSPDummyLM, dummy_rm
# "Observation 2: tool 2 output\n\n"
# "Thought 3: Even more thoughts\n\n"
# "Action 3: finish[baz]"
-# )
+# )
\ No newline at end of file
diff --git a/tests/dsp_LM/predict/test_retry.py b/tests/dsp_LM/predict/test_retry.py
index bd22984d4..89cac67c9 100644
--- a/tests/dsp_LM/predict/test_retry.py
+++ b/tests/dsp_LM/predict/test_retry.py
@@ -72,7 +72,7 @@ def test_retry_forward_with_typed_predictor():
dspy.settings.configure(lm=lm, trace=[])
class AnswerQuestion(dspy.Signature):
- """Answer questions with succint responses."""
+ """Answer questions with succinct responses."""
class Input(pydantic.BaseModel):
question: str
diff --git a/tests/dsp_LM/teleprompt/test_mipro_optimizer.py b/tests/dsp_LM/teleprompt/test_mipro_optimizer.py
index 86d8c00d0..c699be2eb 100644
--- a/tests/dsp_LM/teleprompt/test_mipro_optimizer.py
+++ b/tests/dsp_LM/teleprompt/test_mipro_optimizer.py
@@ -111,7 +111,7 @@ class ConditionalLM(LM):
return [choice["text"] for choice in response["choices"]]
def get_convo(self, index):
- """get the prompt + anwer from the ith message"""
+ """get the prompt + answer from the ith message"""
return self.history[index]["prompt"] + " " + self.history[index]["response"]["choices"][0]["text"]
diff --git a/tests/functional/test_functional.py b/tests/functional/test_functional.py
index 9674cca19..8623c6f55 100644
--- a/tests/functional/test_functional.py
+++ b/tests/functional/test_functional.py
@@ -502,7 +502,7 @@ def test_literal():
assert f() == "2"
-def test_literal_missmatch():
+def test_literal_mismatch():
lm = DummyLM([{"f": f"{i}"} for i in range(5, 100)])
dspy.settings.configure(lm=lm)
@@ -527,7 +527,7 @@ def test_literal_int():
assert f() == 2
-def test_literal_int_missmatch():
+def test_literal_int_mismatch():
lm = DummyLM([{"f": f"{i}"} for i in range(5, 100)])
dspy.settings.configure(lm=lm)
@@ -823,7 +823,7 @@ def test_model_validator():
category: str = dspy.OutputField()
@model_validator(mode="after")
- def check_cateogry(self):
+ def check_category(self):
if self.category not in self.allowed_categories:
raise ValueError(f"category not in {self.allowed_categories}")
return self
diff --git a/tests/predict/test_react.py b/tests/predict/test_react.py
index 1ba36b21b..8435f86a9 100644
--- a/tests/predict/test_react.py
+++ b/tests/predict/test_react.py
@@ -5,7 +5,7 @@ from dspy.utils.dummies import DummyLM, dummy_rm
# def test_example_no_tools():
-# # Createa a simple dataset which the model will use with the Retrieve tool.
+# # Create a simple dataset which the model will use with the Retrieve tool.
# lm = DummyLM(
# [
# {"Thought_1": "Initial thoughts", "Action_1": "Finish[blue]"},
@@ -25,7 +25,7 @@ from dspy.utils.dummies import DummyLM, dummy_rm
# def test_example_search():
-# # Createa a simple dataset which the model will use with the Retrieve tool.
+# # Create a simple dataset which the model will use with the Retrieve tool.
# lm = DummyLM(
# [
# {"Thought_1": "Initial thoughts", "Action_1": "Search[the color of the sky]"},
@@ -35,7 +35,7 @@ from dspy.utils.dummies import DummyLM, dummy_rm
# rm = dummy_rm(
# [
# "We all know the color of the sky is blue.",
-# "Somethng about the sky colors",
+# "Something about the sky colors",
# "This sentence is completely irellevant to answer the question.",
# "Let's add some more sentences to act as summy passages.",
# "Let's add some more sentences to act as summy passages.",
@@ -121,4 +121,4 @@ from dspy.utils.dummies import DummyLM, dummy_rm
# react = dspy.ReAct(ExampleSignature)
# assert react.react[0].signature.instructions is not None
-# assert react.react[0].signature.instructions.startswith("You are going to generate output based on input.")
+# assert react.react[0].signature.instructions.startswith("You are going to generate output based on input.")
\ No newline at end of file
diff --git a/tests/predict/test_retry.py b/tests/predict/test_retry.py
index 687a18dbf..4289ab75e 100644
--- a/tests/predict/test_retry.py
+++ b/tests/predict/test_retry.py
@@ -59,7 +59,7 @@ def test_retry_forward_with_typed_predictor():
dspy.settings.configure(lm=lm, trace=[])
class AnswerQuestion(dspy.Signature):
- """Answer questions with succint responses."""
+ """Answer questions with succinct responses."""
class Input(pydantic.BaseModel):
question: str
diff --git a/tests/primitives/test_program.py b/tests/primitives/test_program.py
index ea6633682..546fc5971 100644
--- a/tests/primitives/test_program.py
+++ b/tests/primitives/test_program.py
@@ -132,7 +132,7 @@ def test_complex_module_traversal():
"self.sub_module.nested_list[0]",
"self.sub_module.nested_list[1][key]", # NOTE: named_sub_modules allows recursive structures
"self.sub_module.nested_tuple[0]",
- "self.sub_module.nested_tuple[1][0]", # NEW: named_sub_modules allows recursive structures, but named_prameters does not
+ "self.sub_module.nested_tuple[1][0]", # NEW: named_sub_modules allows recursive structures, but named_parameters does not
# "self.sub_module.nested_tuple[1][1]", This should not be included, as it's the same module as the previous one
}
found_names = {name for name, _ in root.named_sub_modules()}