[week5 dogscats-ensemble]

MTAU · November 16, 2017, 2:54am

During the build ensemble block of code within dogscate-ensemble

for i in range(5):
i = str(i)
model = train_last_layer(i)
train_dense_layers(i, model)

I am seeing this error.

ValueError: You called set_weights(weights) on layer “batchnormalization_3” with a weight list of length 0, but the layer was expecting 4 weights. Provided weights: []…

github link with all it’s embarrassing errors.

github.com

aspiringguru/fastai_bmt_reworking/blob/master/Week5/dogscats-ensemble_working.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from datetime import datetime\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#set FIXED variables\n",
    "HOMEPATH = \"/home/ubuntu/fastai/\"\n",

This file has been truncated. show original

ll_feat.shape = (23000, 4096)
ll_val_feat.shape = (2000, 4096)

Keras : 1.2.2
Theano : 0.9.0.dev-c697eeab84e5b8a74908da654b66ec9eca4f1291
Tensorflow : 1.3.0

using theano backend

MTAU · November 20, 2017, 4:41am

hrm… was hoping someone would call me on an obvious error.

no?

did everyone sail through this with no problems?