cat-llm 0.0.24__tar.gz → 0.0.25__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {cat_llm-0.0.24 → cat_llm-0.0.25}/PKG-INFO +1 -1
- {cat_llm-0.0.24 → cat_llm-0.0.25}/src/catllm/CERAD_functions.py +15 -0
- {cat_llm-0.0.24 → cat_llm-0.0.25}/src/catllm/__about__.py +1 -1
- {cat_llm-0.0.24 → cat_llm-0.0.25}/LICENSE +0 -0
- {cat_llm-0.0.24 → cat_llm-0.0.25}/README.md +0 -0
- {cat_llm-0.0.24 → cat_llm-0.0.25}/pyproject.toml +0 -0
- {cat_llm-0.0.24 → cat_llm-0.0.25}/src/catllm/__init__.py +0 -0
- {cat_llm-0.0.24 → cat_llm-0.0.25}/src/catllm/cat_llm.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: cat-llm
|
|
3
|
-
Version: 0.0.
|
|
3
|
+
Version: 0.0.25
|
|
4
4
|
Summary: A tool for categorizing text data and images using LLMs and vision models
|
|
5
5
|
Project-URL: Documentation, https://github.com/chrissoria/cat-llm#readme
|
|
6
6
|
Project-URL: Issues, https://github.com/chrissoria/cat-llm/issues
|
|
@@ -1,5 +1,20 @@
|
|
|
1
1
|
# a function specifically for CERAD Constructional Praxis test
|
|
2
2
|
# specifically for pictures of drawings of shapes like circles, diamonds, rectangles, and cubes
|
|
3
|
+
|
|
4
|
+
"""
|
|
5
|
+
Areas for improvement:
|
|
6
|
+
1. Prompt refinement: adjusting the prompt so that it produces a more accurate score.
|
|
7
|
+
2. Image preprocessing: adjusting the images so that they are easier to be analyzed by the models.
|
|
8
|
+
3. Model selection: using a different model that is better suited for image analysis.
|
|
9
|
+
4. Model Ensembling: using multiple models and combining their scores to produce a more accurate score.
|
|
10
|
+
5. Prompt ensembling: using multiple prompts and combining their scores to produce a more accurate score.
|
|
11
|
+
6. Post-processing: adjusting the way scores are calculated after the model has output its assessment.
|
|
12
|
+
7. Efficiency: optimizing the code to run faster, cheaper, and more efficiently.
|
|
13
|
+
8. Drawn-format versatility: making the function more versatile to handle different scenarios, such as shapes drawn on tablets.
|
|
14
|
+
9. Image input flexibility: allowing the function to accept images in various formats, such as URLs or raw image data.
|
|
15
|
+
10. Test variety: expanding or adding functions to handle score more tests relevant for cogntive assesment, such as the MMSE.
|
|
16
|
+
11. Error handling: improving error handling to better manage unexpected inputs or model failures.
|
|
17
|
+
"""
|
|
3
18
|
def cerad_drawn_score(
|
|
4
19
|
shape,
|
|
5
20
|
image_input,
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|