fix eval ts

Merge pull request #6 from akashkolte/akash/m3-changes

fixed affect

tighten affect classifier: less twitchy FRUSTRATED, smile beats brow

revert ink vision endpoint from Ollama Cloud back to Gemini

add live HF Space URL to README

add Dockerfile + HF Space frontmatter for hosted deploy

bug fixes

relevance + diversity evals, batched scoring, math tooltips

Prompt imporved

updated readme

sensing calibration + ink via ollama

Revert "added docker file"

updated prompt

head gestures

air-writing

added affect and gesture fix

added docker file

-

Add debounce + tighten sensing thresholds

Fix affect and gesture misclassification

-

refactored and added synthetic data

updated gitignore

Add voice + air-writing conflict resolution

Stop blocking /chat on evals; let the UI poll for them

Streaming candidate picker + side-index feedback loops

Session priors (P(bucket) + P(type)) + inline eval UI

Add faithfulness + authenticity evals; run scoring off the response path

Add MMR reranking with conversation-context query fusion

imultimodal sensing → real stylistic constraints

Add PRESENT_STATE intent + head-shake turnaround

adding lessons

Route sub-intents to their own pools, rip out the LLM intent router

added deeper memories and new users.

added favicon

dropped more bloat

Modified UI + Readme Update

Remove library bloat

adding detailed TODOs to ReadME

Added scaffolding for evals

Merge pull request #4 from akashkolte/shwetangi/m2-changes

Add todos to Readme

Code refactor for extensibility

added a setup script

Merge pull request #3 from akashkolte/akash/v1

added supress

updated settings

added basic pipeline

added basic pipeline