Update README.md
Browse files
README.md
CHANGED
|
@@ -28,18 +28,19 @@ Load in your favorite GGUF inference engine, or try with llmware as follows:
|
|
| 28 |
# this one line will download the model and run a series of tests
|
| 29 |
ModelCatalog().tool_test_run("slim-sentiment-tool", verbose=True)
|
| 30 |
|
| 31 |
-
Note: please review [**config.json**](https://huggingface.co/llmware/slim-sentiment-tool/blob/main/config.json) in the repository for prompt wrapping information, details on the model, and full test set.
|
| 32 |
|
| 33 |
-
|
| 34 |
Slim models can also be loaded even more simply as part of a multi-model, multi-step LLMfx calls:
|
| 35 |
|
| 36 |
from llmware.agents import LLMfx
|
| 37 |
|
| 38 |
llm_fx = LLMfx()
|
| 39 |
llm_fx.load_tool("sentiment")
|
| 40 |
-
response = llm_fx.sentiment(text)
|
| 41 |
|
| 42 |
|
|
|
|
|
|
|
|
|
|
| 43 |
### Model Description
|
| 44 |
|
| 45 |
<!-- Provide a longer summary of what this model is. -->
|
|
@@ -50,27 +51,6 @@ Slim models can also be loaded even more simply as part of a multi-model, multi-
|
|
| 50 |
- **License:** Apache 2.0
|
| 51 |
- **Quantized from model:** llmware/slim-sentiment (finetuned tiny llama 1b)
|
| 52 |
|
| 53 |
-
## Uses
|
| 54 |
-
|
| 55 |
-
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 56 |
-
|
| 57 |
-
SLIM models provide a fast, flexible, intuitive way to integrate classifiers and structured function calls into RAG and LLM application workflows.
|
| 58 |
-
|
| 59 |
-
Model instructions, details and test samples have been packaged into the config.json file in the repository, along with the GGUF file.
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
Example:
|
| 63 |
-
|
| 64 |
-
text = "The stock market declined yesterday as investors worried increasingly about the slowing economy."
|
| 65 |
-
|
| 66 |
-
model generation - {"sentiment": ["negative"]}
|
| 67 |
-
|
| 68 |
-
keys = "sentiment"
|
| 69 |
-
|
| 70 |
-
All of the SLIM models use a novel prompt instruction structured as follows:
|
| 71 |
-
|
| 72 |
-
"<human> " + text + "<classify> " + keys + "</classify>" + "/n<bot>: "
|
| 73 |
-
|
| 74 |
|
| 75 |
## Model Card Contact
|
| 76 |
|
|
|
|
| 28 |
# this one line will download the model and run a series of tests
|
| 29 |
ModelCatalog().tool_test_run("slim-sentiment-tool", verbose=True)
|
| 30 |
|
|
|
|
| 31 |
|
|
|
|
| 32 |
Slim models can also be loaded even more simply as part of a multi-model, multi-step LLMfx calls:
|
| 33 |
|
| 34 |
from llmware.agents import LLMfx
|
| 35 |
|
| 36 |
llm_fx = LLMfx()
|
| 37 |
llm_fx.load_tool("sentiment")
|
| 38 |
+
response = llm_fx.sentiment(text)
|
| 39 |
|
| 40 |
|
| 41 |
+
Note: please review [**config.json**](https://huggingface.co/llmware/slim-sentiment-tool/blob/main/config.json) in the repository for prompt wrapping information, details on the model, and full test set.
|
| 42 |
+
|
| 43 |
+
|
| 44 |
### Model Description
|
| 45 |
|
| 46 |
<!-- Provide a longer summary of what this model is. -->
|
|
|
|
| 51 |
- **License:** Apache 2.0
|
| 52 |
- **Quantized from model:** llmware/slim-sentiment (finetuned tiny llama 1b)
|
| 53 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
|
| 55 |
## Model Card Contact
|
| 56 |
|