File size: 2,791 Bytes
83b2998 ca7713e 83b2998 ca7713e e87260e ca7713e 14d296d ca7713e e87260e ca7713e 14d296d ca7713e e87260e ca7713e 14d296d ca7713e 14d296d ca7713e 14d296d ca7713e 14d296d ca7713e 1c00e94 ca7713e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 | ---
license: apache-2.0
inference: false
---
# SLIM-CATEGORY
<!-- Provide a quick summary of what the model is/does. -->
**slim-category** is part of the SLIM ("**S**tructured **L**anguage **I**nstruction **M**odel") model series, consisting of small, specialized decoder-based models, fine-tuned for function-calling.
slim-category has been fine-tuned for **category topic analysis** function calls, generating output consisting of a python dictionary corresponding to specified keys, e.g.:
`{"category": ["markets"]}`
SLIM models are designed to generate structured outputs that can be used programmatically as part of a multi-step, multi-model LLM-based automation workflow.
Each slim model has a 'quantized tool' version, e.g., [**'slim-category-tool'**](https://huggingface.co/llmware/slim-category-tool).
## Prompt format:
`function = "classify"`
`params = "category"`
`prompt = "<human> " + {text} + "\n" + `
`"<{function}> " + {params} + "</{function}>" + "\n<bot>:"`
<details>
<summary>Transformers Script </summary>
model = AutoModelForCausalLM.from_pretrained("llmware/slim-category")
tokenizer = AutoTokenizer.from_pretrained("llmware/slim-category")
function = "classify"
params = "category"
text = "The stock market declined yesterday as investors worried increasingly about the slowing economy."
prompt = "<human>: " + text + "\n" + f"<{function}> {params} </{function}>\n<bot>:"
inputs = tokenizer(prompt, return_tensors="pt")
start_of_input = len(inputs.input_ids[0])
outputs = model.generate(
inputs.input_ids.to('cpu'),
eos_token_id=tokenizer.eos_token_id,
pad_token_id=tokenizer.eos_token_id,
do_sample=True,
temperature=0.3,
max_new_tokens=100
)
output_only = tokenizer.decode(outputs[0][start_of_input:], skip_special_tokens=True)
print("output only: ", output_only)
# here's the fun part
try:
output_only = ast.literal_eval(llm_string_output)
print("success - converted to python dictionary automatically")
except:
print("fail - could not convert to python dictionary automatically - ", llm_string_output)
</details>
<details>
<summary>Using as Function Call in LLMWare</summary>
from llmware.models import ModelCatalog
slim_model = ModelCatalog().load_model("llmware/slim-category")
response = slim_model.function_call(text,params=["category"], function="classify")
print("llmware - llm_response: ", response)
</details>
## Model Card Contact
Darren Oberst & llmware team
[Join us on Discord](https://discord.gg/MhZn5Nc39h)
|