{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Important: This notebook will only work with fastai-0.7.x. Do not try to run any fastai-1.x code from this path in the repository because it will load fastai-0.7.x**"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Image classification with Convolutional Neural Networks"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Put these at the top of every notebook, to get automatic reloading and inline plotting\n",
"%reload_ext autoreload\n",
"%autoreload 2\n",
"%matplotlib inline"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# This file contains all the main external libs we'll use\n",
"from fastai.imports import *\n",
"\n",
"from fastai.transforms import *\n",
"from fastai.conv_learner import *\n",
"from fastai.model import *\n",
"from fastai.dataset import *\n",
"from fastai.sgdr import *\n",
"from fastai.plots import *"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"PATH = \"data/dogscats/\"\n",
"sz=224\n",
"arch=vgg16\n",
"bs=64"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Uncomment the below if you need to reset your precomputed activations\n",
"# !rm -rf {PATH}tmp"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = ImageClassifierData.from_paths(PATH, bs=bs, tfms=tfms_from_model(arch, sz))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"100%|██████████| 360/360 [01:21<00:00, 4.44it/s]\n",
"100%|██████████| 32/32 [00:12<00:00, 2.64it/s]\n"
]
}
],
"source": [
"learn = ConvLearner.pretrained(arch, data, precompute=True)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ea734bbc71264979961457aa35490d6e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"A Jupyter Widget"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[ 0. 0.05266 0.02898 0.99072] \n",
"[ 1. 0.04534 0.02512 0.99023] \n",
"[ 2. 0.04645 0.02377 0.99121] \n",
"\n"
]
}
],
"source": [
"learn.fit(0.01, 3, cycle_len=1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"tfms = tfms_from_model(arch, sz, aug_tfms=transforms_side_on, max_zoom=1.1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = ImageClassifierData.from_paths(PATH, tfms=tfms, bs=bs, num_workers=4)\n",
"learn = ConvLearner.pretrained(arch, data, precompute=True)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "26ab4ac9de2945efb08366756bc95727",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"A Jupyter Widget"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[ 0. 0.05622 0.0268 0.98926] \n",
"[ 1. 0.04222 0.02508 0.99072] \n",
"\n"
]
}
],
"source": [
"learn.fit(1e-2, 2)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.precompute=False"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "efe2b55c24c64e688ad6e68fe2fb2a35",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"A Jupyter Widget"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[ 0. 0.05511 0.02654 0.98909] \n",
"\n"
]
}
],
"source": [
"learn.fit(1e-2, 1, cycle_len=1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.unfreeze()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"lr=np.array([1e-4,1e-3,1e-2])"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d741a2efd6574cb0a97d641f7a95ea9c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"A Jupyter Widget"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[ 0. 0.05664 0.0245 0.99023] \n",
"\n"
]
}
],
"source": [
"learn.fit(lr, 1, cycle_len=1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "be834ba833ff43bdba628966c24311c0",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"A Jupyter Widget"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[ 0. 0.05955 0.02217 0.9917 ] \n",
"[ 1. 0.03902 0.0181 0.99414] \n",
"[ 2. 0.03676 0.0168 0.99414] \n",
"[ 3. 0.03097 0.01607 0.99316] \n",
"[ 4. 0.02823 0.01574 0.99316] \n",
"[ 5. 0.02663 0.01448 0.99316] \n",
"[ 6. 0.03113 0.01642 0.99414] \n",
"\n"
]
}
],
"source": [
"learn.fit(lr, 3, cycle_len=1, cycle_mult=2)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "bc193cd343b74190a24ccf132c98b680",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"A Jupyter Widget"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"[ 0. 0.02769 0.01896 0.99268] \n",
"[ 1. 0.01939 0.01417 0.99365] \n",
"[ 2. 0.01935 0.0142 0.99414] \n",
"[ 3. 0.02465 0.01641 0.99316] \n",
"[ 4. 0.02753 0.01376 0.99365] \n",
"[ 5. 0.01721 0.01413 0.99414] \n",
"[ 6. 0.0221 0.01317 0.99365] \n",
"[ 7. 0.01789 0.01379 0.99268] \n",
"[ 8. 0.01861 0.0139 0.99316] \n",
"\n"
]
}
],
"source": [
"learn.fit(lr, 3, cycle_len=3)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"0.995"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"log_preds,y = learn.TTA()\n",
"probs = np.mean(np.exp(log_preds),0);",
"accuracy_np(probs,y)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
}
},
"nbformat": 4,
"nbformat_minor": 2
}