Optimizing text classification for small datasets

orenmel · November 28, 2018, 1:56pm

Hi.

I’m trying to optimize a text classifier for small subsets of the IMDB (something like the IMDB sample or even smaller) based on this starting point:

github.com

fastai/course-v3/blob/master/nbs/dl1/lesson3-imdb.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# IMDB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%reload_ext autoreload\n",
    "%autoreload 2\n",
    "%matplotlib inline"
   ]
  },

This file has been truncated. show original

Has anybody tried something like that and can share recommended hyperparameters?

Thanks!