2020-03-06 18:19:03 +00:00
{
"cells": [
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#hide\n",
"from utils import *\n",
"from fastai2.vision.widgets import *"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"# From Model to Production"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"## The Practice of Deep Learning"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### Starting Your Project"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### The State of Deep Learning"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Computer vision"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Text (natural language processing)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Combining text and images"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Tabular data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Recommendation systems"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"#### Other data types"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### The Drivetrain Approach"
2020-03-06 18:19:03 +00:00
]
},
2020-03-24 18:18:16 +00:00
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"## Gathering Data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"To download images with Bing Image Search, sign up at Microsoft for a free account. You will be given a key, which you can copy and enter in a cell as follows (replacing 'XXX' with your key and executing it):"
2020-03-24 18:18:16 +00:00
]
},
2020-03-06 18:19:03 +00:00
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"key = 'XXX'"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<function utils.search_images_bing(key, term, min_sz=128)>"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"search_images_bing"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"150"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"results = search_images_bing(key, 'grizzly bear')\n",
"ims = results.attrgot('content_url')\n",
"len(ims)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"hide_input": true
},
"outputs": [],
"source": [
"#hide\n",
"ims = ['http://3.bp.blogspot.com/-S1scRCkI3vY/UHzV2kucsPI/AAAAAAAAA-k/YQ5UzHEm9Ss/s1600/Grizzly%2BBear%2BWildlife.jpg']"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"dest = 'images/grizzly.jpg'\n",
"download_url(ims[0], dest)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<PIL.Image.Image image mode=RGB size=109x128 at 0x7F78BCDD5390>"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"im = Image.open(dest)\n",
"im.to_thumb(128,128)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"bear_types = 'grizzly','black','teddy'\n",
"path = Path('bears')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"if not path.exists():\n",
" path.mkdir()\n",
" for o in bear_types:\n",
" dest = (path/o)\n",
" dest.mkdir(exist_ok=True)\n",
" results = search_images_bing(key, f'{o} bear')\n",
" download_images(dest, urls=results.attrgot('content_url'))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(#421) [Path('bears/black/00000095.jpg'),Path('bears/black/00000133.jpg'),Path('bears/black/00000062.jpg'),Path('bears/black/00000023.jpg'),Path('bears/black/00000029.jpg'),Path('bears/black/00000094.jpg'),Path('bears/black/00000124.jpg'),Path('bears/black/00000056.jpeg'),Path('bears/black/00000046.jpg'),Path('bears/black/00000045.jpg')...]"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"fns = get_image_files(path)\n",
"fns"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"(#0) []"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"failed = verify_images(fns)\n",
"failed"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"failed.map(Path.unlink);"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### Sidebar: Getting Help in Jupyter Notebooks"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### End sidebar"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"## From Data to DataLoaders"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"bears = DataBlock(\n",
" blocks=(ImageBlock, CategoryBlock), \n",
" get_items=get_image_files, \n",
2020-05-14 12:18:31 +00:00
" splitter=RandomSplitter(valid_pct=0.2, seed=42),\n",
2020-03-06 18:19:03 +00:00
" get_y=parent_label,\n",
" item_tfms=Resize(128))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"dls = bears.dataloaders(path)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 864x216 with 4 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
2020-03-19 13:21:55 +00:00
"dls.valid.show_batch(max_n=4, nrows=1)"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 864x216 with 4 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"bears = bears.new(item_tfms=Resize(128, ResizeMethod.Squish))\n",
"dls = bears.dataloaders(path)\n",
2020-03-19 13:21:55 +00:00
"dls.valid.show_batch(max_n=4, nrows=1)"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 864x216 with 4 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"bears = bears.new(item_tfms=Resize(128, ResizeMethod.Pad, pad_mode='zeros'))\n",
"dls = bears.dataloaders(path)\n",
2020-03-19 13:21:55 +00:00
"dls.valid.show_batch(max_n=4, nrows=1)"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 864x216 with 4 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"bears = bears.new(item_tfms=RandomResizedCrop(128, min_scale=0.3))\n",
"dls = bears.dataloaders(path)\n",
2020-03-31 20:57:32 +00:00
"dls.train.show_batch(max_n=4, nrows=1, unique=True)"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### Data Augmentation"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 864x432 with 8 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"bears = bears.new(item_tfms=Resize(128), batch_tfms=aug_transforms(mult=2))\n",
"dls = bears.dataloaders(path)\n",
2020-03-31 20:57:32 +00:00
"dls.train.show_batch(max_n=8, nrows=2, unique=True)"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"## Training Your Model, and Using It to Clean Your Data"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"bears = bears.new(\n",
" item_tfms=RandomResizedCrop(224, min_scale=0.5),\n",
" batch_tfms=aug_transforms())\n",
"dls = bears.dataloaders(path)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: left;\">\n",
" <th>epoch</th>\n",
" <th>train_loss</th>\n",
" <th>valid_loss</th>\n",
" <th>error_rate</th>\n",
" <th>time</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <td>0</td>\n",
" <td>1.235733</td>\n",
" <td>0.212541</td>\n",
" <td>0.087302</td>\n",
" <td>00:05</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: left;\">\n",
" <th>epoch</th>\n",
" <th>train_loss</th>\n",
" <th>valid_loss</th>\n",
" <th>error_rate</th>\n",
" <th>time</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <td>0</td>\n",
" <td>0.213371</td>\n",
" <td>0.112450</td>\n",
" <td>0.023810</td>\n",
" <td>00:05</td>\n",
" </tr>\n",
" <tr>\n",
" <td>1</td>\n",
" <td>0.173855</td>\n",
" <td>0.072306</td>\n",
" <td>0.023810</td>\n",
" <td>00:06</td>\n",
" </tr>\n",
" <tr>\n",
" <td>2</td>\n",
" <td>0.147096</td>\n",
" <td>0.039068</td>\n",
" <td>0.015873</td>\n",
" <td>00:06</td>\n",
" </tr>\n",
" <tr>\n",
" <td>3</td>\n",
" <td>0.123984</td>\n",
" <td>0.026801</td>\n",
" <td>0.015873</td>\n",
" <td>00:06</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"learn = cnn_learner(dls, resnet18, metrics=error_rate)\n",
"learn.fine_tune(4)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"interp = ClassificationInterpretation.from_learner(learn)\n",
"interp.plot_confusion_matrix()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 1080x288 with 5 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
2020-03-19 13:21:55 +00:00
"interp.plot_top_losses(5, nrows=1)"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d547f14e0f7848f39627ebb88d457e64",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"VBox(children=(Dropdown(options=('black', 'grizzly', 'teddy'), value='black'), Dropdown(options=('Train', 'Val…"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"cleaner = ImageClassifierCleaner(learn)\n",
"cleaner"
]
},
2020-03-24 12:47:36 +00:00
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#hide\n",
"# for idx in cleaner.delete(): cleaner.fns[idx].unlink()\n",
2020-03-31 00:49:36 +00:00
"# for idx,cat in cleaner.change(): shutil.move(str(cleaner.fns[idx]), path/cat)"
2020-03-24 12:47:36 +00:00
]
},
2020-03-06 18:19:03 +00:00
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"## Turning Your Model into an Online Application"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### Using the Model for Inference"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.export()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(#1) [Path('export.pkl')]"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"path = Path()\n",
"path.ls(file_exts='.pkl')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn_inf = load_learner(path/'export.pkl')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"('grizzly', tensor(1), tensor([9.0767e-06, 9.9999e-01, 1.5748e-07]))"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"learn_inf.predict('images/grizzly.jpg')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(#3) ['black','grizzly','teddy']"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"learn_inf.dls.vocab"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### Creating a Notebook App from the Model"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "e0c4141e3c76425c98ae9994ccf9a748",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"FileUpload(value={}, description='Upload')"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"btn_upload = widgets.FileUpload()\n",
"btn_upload"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"hide_input": true
},
"outputs": [],
"source": [
"#hide\n",
"# For the book, we can't actually click an upload button, so we fake it\n",
"btn_upload = SimpleNamespace(data = ['images/grizzly.jpg'])"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"img = PILImage.create(btn_upload.data[-1])"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"out_pl = widgets.Output()\n",
"out_pl.clear_output()\n",
"with out_pl: display(img.to_thumb(128,128))\n",
"out_pl"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pred,pred_idx,probs = learn_inf.predict(img)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "08509e39d3454701b5fed10439970e84",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Label(value='Prediction: grizzly; Probability: 1.0000')"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"lbl_pred = widgets.Label()\n",
"lbl_pred.value = f'Prediction: {pred}; Probability: {probs[pred_idx]:.04f}'\n",
"lbl_pred"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5948c2dc026d43cb9afdce7dee8fa425",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Button(description='Classify', style=ButtonStyle())"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"btn_run = widgets.Button(description='Classify')\n",
"btn_run"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"def on_click_classify(change):\n",
" img = PILImage.create(btn_upload.data[-1])\n",
" out_pl.clear_output()\n",
" with out_pl: display(img.to_thumb(128,128))\n",
" pred,pred_idx,probs = learn_inf.predict(img)\n",
" lbl_pred.value = f'Prediction: {pred}; Probability: {probs[pred_idx]:.04f}'\n",
"\n",
"btn_run.on_click(on_click_classify)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#hide\n",
"#Putting back btn_upload to a widget for next cell\n",
"btn_upload = widgets.FileUpload()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "e9e7b05555a44125ac0e5365e17ea59d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"VBox(children=(Label(value='Select your bear!'), FileUpload(value={}, description='Upload'), Button(descriptio…"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"VBox([widgets.Label('Select your bear!'), \n",
" btn_upload, btn_run, out_pl, lbl_pred])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### Turning Your Notebook into a Real App"
2020-03-06 18:19:03 +00:00
]
},
2020-03-24 12:47:36 +00:00
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#hide\n",
"# !pip install voila\n",
2020-05-19 01:18:45 +00:00
"# !jupyter serverextension enable voila —sys-prefix"
2020-03-24 12:47:36 +00:00
]
},
2020-03-06 18:19:03 +00:00
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Deploying your app"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"## How to Avoid Disaster"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### Unforeseen Consequences and Feedback Loops"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"## Get Writing!"
2020-03-06 18:19:03 +00:00
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Questionnaire"
]
},
2020-03-18 00:34:07 +00:00
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"1. Provide an example of where the bear classification model might work poorly in production, due to structural or style differences in the training data.\n",
2020-03-18 00:34:07 +00:00
"1. Where do text models currently have a major deficiency?\n",
"1. What are possible negative societal implications of text generation models?\n",
"1. In situations where a model might make mistakes, and those mistakes could be harmful, what is a good alternative to automating a process?\n",
"1. What kind of tabular data is deep learning particularly good at?\n",
"1. What's a key downside of directly using a deep learning model for recommendation systems?\n",
2020-05-14 12:18:31 +00:00
"1. What are the steps of the Drivetrain Approach?\n",
"1. How do the steps of the Drivetrain Approach map to a recommendation system?\n",
2020-03-18 00:34:07 +00:00
"1. Create an image recognition model using data you curate, and deploy it on the web.\n",
"1. What is `DataLoaders`?\n",
"1. What four things do we need to tell fastai to create `DataLoaders`?\n",
"1. What does the `splitter` parameter to `DataBlock` do?\n",
"1. How do we ensure a random split always gives the same validation set?\n",
"1. What letters are often used to signify the independent and dependent variables?\n",
2020-05-14 12:18:31 +00:00
"1. What's the difference between the crop, pad, and squish resize approaches? When might you choose one over the others?\n",
2020-03-18 00:34:07 +00:00
"1. What is data augmentation? Why is it needed?\n",
"1. What is the difference between `item_tfms` and `batch_tfms`?\n",
"1. What is a confusion matrix?\n",
"1. What does `export` save?\n",
"1. What is it called when we use a model for getting predictions, instead of training?\n",
"1. What are IPython widgets?\n",
"1. When might you want to use CPU for deployment? When might GPU be better?\n",
"1. What are the downsides of deploying your app to a server, instead of to a client (or edge) device such as a phone or PC?\n",
2020-05-14 12:18:31 +00:00
"1. What are three examples of problems that could occur when rolling out a bear warning system in practice?\n",
"1. What is \"out-of-domain data\"?\n",
2020-03-18 00:34:07 +00:00
"1. What is \"domain shift\"?\n",
2020-05-14 12:18:31 +00:00
"1. What are the three steps in the deployment process?"
2020-03-18 00:34:07 +00:00
]
},
2020-03-06 18:19:03 +00:00
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"### Further Research"
2020-03-06 18:19:03 +00:00
]
},
2020-03-18 00:34:07 +00:00
{
"cell_type": "markdown",
"metadata": {},
"source": [
2020-05-14 12:18:31 +00:00
"1. Consider how the Drivetrain Approach maps to a project or problem you're interested in.\n",
"1. When might it be best to avoid certain types of data augmentation?\n",
"1. For a project you're interested in applying deep learning to, consider the thought experiment \"What would happen if it went really, really well?\"\n",
"1. Start a blog, and write your first blog post. For instance, write about what you think deep learning might be useful for in a domain you're interested in."
2020-03-18 00:34:07 +00:00
]
},
2020-03-06 18:19:03 +00:00
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
2020-03-19 13:21:55 +00:00
"jupytext": {
"split_at_heading": true
},
2020-03-06 18:19:03 +00:00
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
