Instructions to use Chaotic-Transsion/CMMI-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use Chaotic-Transsion/CMMI-models with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('feature-extraction', 'Chaotic-Transsion/CMMI-models');
CMMI models
Models for CMMI (Consolidated Media and Market Intelligence), a tool that assigns themes and sentiment to social media comments about phones, in English, Urdu and Roman Urdu. Everything here runs in the browser with Transformers.js.
| Path | Contents |
|---|---|
config.json, tokenizer.json, onnx/model_quantized.onnx |
int8 ONNX build of multilingual-e5-base, mirrored from Xenova/multilingual-e5-base |
classifier/classifier.json |
Logistic-regression weights, per-theme thresholds and text-cleaning rules, trained on hand-labelled comments |
Loading the embedder:
import { pipeline } from "@huggingface/transformers";
const embed = await pipeline("feature-extraction", "Chaotic-Transsion/CMMI-models", { dtype: "q8" });
Built by the scripts in the CMMI-inference repo.
- Downloads last month
- 63
Model tree for Chaotic-Transsion/CMMI-models
Base model
intfloat/multilingual-e5-base