diff --git "a/testing.ipynb" "b/testing.ipynb" new file mode 100644--- /dev/null +++ "b/testing.ipynb" @@ -0,0 +1,217 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "from fastai.vision.all import *\n", + "import gradio as gr\n", + "import timm\n", + "import requests\n", + "from bs4 import BeautifulSoup" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from PIL import Image" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "im = Image.open(\"leaf1.jpg\")\n", + "im.to_thumb(224,224)" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [], + "source": [ + "learn = load_learner(\"export_ConvNextLargeLargeDS.pkl\")" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "PIL.JpegImagePlugin.JpegImageFile" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "type(im)" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": {}, + "outputs": [], + "source": [ + "import torchvision.transforms as T\n", + "img_pil = Image.open(\"leaf1.jpg\")\n", + "\n", + "img_tensor = T.ToTensor()(img_pil)" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "\n", + "\n" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " 0.00% [0/1 00:00<?]\n", + "
\n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "ename": "TypeError", + "evalue": "Cannot handle this data type: (1, 1, 1540), 2953\u001b[0m mode, rawmode \u001b[39m=\u001b[39m _fromarray_typemap[typekey]\n\u001b[1;32m 2954\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mKeyError\u001b[39;00m \u001b[39mas\u001b[39;00m e:\n", + "\u001b[0;31mKeyError\u001b[0m: ((1, 1, 1540), '\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0m pred, idx, probs \u001b[39m=\u001b[39m learn\u001b[39m.\u001b[39;49mpredict(img_tensor)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/learner.py:303\u001b[0m, in \u001b[0;36mLearner.predict\u001b[0;34m(self, item, rm_type_tfms, with_input)\u001b[0m\n\u001b[1;32m 301\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mpredict\u001b[39m(\u001b[39mself\u001b[39m, item, rm_type_tfms\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, with_input\u001b[39m=\u001b[39m\u001b[39mFalse\u001b[39;00m):\n\u001b[1;32m 302\u001b[0m dl \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdls\u001b[39m.\u001b[39mtest_dl([item], rm_type_tfms\u001b[39m=\u001b[39mrm_type_tfms, num_workers\u001b[39m=\u001b[39m\u001b[39m0\u001b[39m)\n\u001b[0;32m--> 303\u001b[0m inp,preds,_,dec_preds \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mget_preds(dl\u001b[39m=\u001b[39;49mdl, with_input\u001b[39m=\u001b[39;49m\u001b[39mTrue\u001b[39;49;00m, with_decoded\u001b[39m=\u001b[39;49m\u001b[39mTrue\u001b[39;49;00m)\n\u001b[1;32m 304\u001b[0m i \u001b[39m=\u001b[39m \u001b[39mgetattr\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdls, \u001b[39m'\u001b[39m\u001b[39mn_inp\u001b[39m\u001b[39m'\u001b[39m, \u001b[39m-\u001b[39m\u001b[39m1\u001b[39m)\n\u001b[1;32m 305\u001b[0m inp \u001b[39m=\u001b[39m (inp,) \u001b[39mif\u001b[39;00m i\u001b[39m==\u001b[39m\u001b[39m1\u001b[39m \u001b[39melse\u001b[39;00m tuplify(inp)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/learner.py:290\u001b[0m, in \u001b[0;36mLearner.get_preds\u001b[0;34m(self, ds_idx, dl, with_input, with_decoded, with_loss, act, inner, reorder, cbs, **kwargs)\u001b[0m\n\u001b[1;32m 288\u001b[0m \u001b[39mif\u001b[39;00m with_loss: ctx_mgrs\u001b[39m.\u001b[39mappend(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mloss_not_reduced())\n\u001b[1;32m 289\u001b[0m \u001b[39mwith\u001b[39;00m ContextManagers(ctx_mgrs):\n\u001b[0;32m--> 290\u001b[0m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_do_epoch_validate(dl\u001b[39m=\u001b[39;49mdl)\n\u001b[1;32m 291\u001b[0m \u001b[39mif\u001b[39;00m act \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m: act \u001b[39m=\u001b[39m getcallable(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mloss_func, \u001b[39m'\u001b[39m\u001b[39mactivation\u001b[39m\u001b[39m'\u001b[39m)\n\u001b[1;32m 292\u001b[0m res \u001b[39m=\u001b[39m cb\u001b[39m.\u001b[39mall_tensors()\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/learner.py:236\u001b[0m, in \u001b[0;36mLearner._do_epoch_validate\u001b[0;34m(self, ds_idx, dl)\u001b[0m\n\u001b[1;32m 234\u001b[0m \u001b[39mif\u001b[39;00m dl \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m: dl \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdls[ds_idx]\n\u001b[1;32m 235\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdl \u001b[39m=\u001b[39m dl\n\u001b[0;32m--> 236\u001b[0m \u001b[39mwith\u001b[39;00m torch\u001b[39m.\u001b[39mno_grad(): \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_with_events(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mall_batches, \u001b[39m'\u001b[39;49m\u001b[39mvalidate\u001b[39;49m\u001b[39m'\u001b[39;49m, CancelValidException)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/learner.py:193\u001b[0m, in \u001b[0;36mLearner._with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m 192\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_with_events\u001b[39m(\u001b[39mself\u001b[39m, f, event_type, ex, final\u001b[39m=\u001b[39mnoop):\n\u001b[0;32m--> 193\u001b[0m \u001b[39mtry\u001b[39;00m: \u001b[39mself\u001b[39m(\u001b[39mf\u001b[39m\u001b[39m'\u001b[39m\u001b[39mbefore_\u001b[39m\u001b[39m{\u001b[39;00mevent_type\u001b[39m}\u001b[39;00m\u001b[39m'\u001b[39m); f()\n\u001b[1;32m 194\u001b[0m \u001b[39mexcept\u001b[39;00m ex: \u001b[39mself\u001b[39m(\u001b[39mf\u001b[39m\u001b[39m'\u001b[39m\u001b[39mafter_cancel_\u001b[39m\u001b[39m{\u001b[39;00mevent_type\u001b[39m}\u001b[39;00m\u001b[39m'\u001b[39m)\n\u001b[1;32m 195\u001b[0m \u001b[39mself\u001b[39m(\u001b[39mf\u001b[39m\u001b[39m'\u001b[39m\u001b[39mafter_\u001b[39m\u001b[39m{\u001b[39;00mevent_type\u001b[39m}\u001b[39;00m\u001b[39m'\u001b[39m); final()\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/learner.py:199\u001b[0m, in \u001b[0;36mLearner.all_batches\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 197\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mall_batches\u001b[39m(\u001b[39mself\u001b[39m):\n\u001b[1;32m 198\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mn_iter \u001b[39m=\u001b[39m \u001b[39mlen\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdl)\n\u001b[0;32m--> 199\u001b[0m \u001b[39mfor\u001b[39;00m o \u001b[39min\u001b[39;00m \u001b[39menumerate\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdl): \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mone_batch(\u001b[39m*\u001b[39mo)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/data/load.py:129\u001b[0m, in \u001b[0;36mDataLoader.__iter__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 127\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mbefore_iter()\n\u001b[1;32m 128\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m__idxs\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mget_idxs() \u001b[39m# called in context of main process (not workers/subprocesses)\u001b[39;00m\n\u001b[0;32m--> 129\u001b[0m \u001b[39mfor\u001b[39;00m b \u001b[39min\u001b[39;00m _loaders[\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mfake_l\u001b[39m.\u001b[39mnum_workers\u001b[39m==\u001b[39m\u001b[39m0\u001b[39m](\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mfake_l):\n\u001b[1;32m 130\u001b[0m \u001b[39m# pin_memory causes tuples to be converted to lists, so convert them back to tuples\u001b[39;00m\n\u001b[1;32m 131\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpin_memory \u001b[39mand\u001b[39;00m \u001b[39mtype\u001b[39m(b) \u001b[39m==\u001b[39m \u001b[39mlist\u001b[39m: b \u001b[39m=\u001b[39m \u001b[39mtuple\u001b[39m(b)\n\u001b[1;32m 132\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdevice \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m: b \u001b[39m=\u001b[39m to_device(b, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdevice)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/torch/utils/data/dataloader.py:681\u001b[0m, in \u001b[0;36m_BaseDataLoaderIter.__next__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 678\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_sampler_iter \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m 679\u001b[0m \u001b[39m# TODO(https://github.com/pytorch/pytorch/issues/76750)\u001b[39;00m\n\u001b[1;32m 680\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_reset() \u001b[39m# type: ignore[call-arg]\u001b[39;00m\n\u001b[0;32m--> 681\u001b[0m data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_next_data()\n\u001b[1;32m 682\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_num_yielded \u001b[39m+\u001b[39m\u001b[39m=\u001b[39m \u001b[39m1\u001b[39m\n\u001b[1;32m 683\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_dataset_kind \u001b[39m==\u001b[39m _DatasetKind\u001b[39m.\u001b[39mIterable \u001b[39mand\u001b[39;00m \\\n\u001b[1;32m 684\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_IterableDataset_len_called \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m \u001b[39mand\u001b[39;00m \\\n\u001b[1;32m 685\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_num_yielded \u001b[39m>\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_IterableDataset_len_called:\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/torch/utils/data/dataloader.py:721\u001b[0m, in \u001b[0;36m_SingleProcessDataLoaderIter._next_data\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 719\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_next_data\u001b[39m(\u001b[39mself\u001b[39m):\n\u001b[1;32m 720\u001b[0m index \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_next_index() \u001b[39m# may raise StopIteration\u001b[39;00m\n\u001b[0;32m--> 721\u001b[0m data \u001b[39m=\u001b[39m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_dataset_fetcher\u001b[39m.\u001b[39;49mfetch(index) \u001b[39m# may raise StopIteration\u001b[39;00m\n\u001b[1;32m 722\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_pin_memory:\n\u001b[1;32m 723\u001b[0m data \u001b[39m=\u001b[39m _utils\u001b[39m.\u001b[39mpin_memory\u001b[39m.\u001b[39mpin_memory(data, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_pin_memory_device)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/torch/utils/data/_utils/fetch.py:39\u001b[0m, in \u001b[0;36m_IterableDatasetFetcher.fetch\u001b[0;34m(self, possibly_batched_index)\u001b[0m\n\u001b[1;32m 37\u001b[0m \u001b[39mraise\u001b[39;00m \u001b[39mStopIteration\u001b[39;00m\n\u001b[1;32m 38\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[0;32m---> 39\u001b[0m data \u001b[39m=\u001b[39m \u001b[39mnext\u001b[39;49m(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdataset_iter)\n\u001b[1;32m 40\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mcollate_fn(data)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/data/load.py:140\u001b[0m, in \u001b[0;36mDataLoader.create_batches\u001b[0;34m(self, samps)\u001b[0m\n\u001b[1;32m 138\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m: \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mit \u001b[39m=\u001b[39m \u001b[39miter\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdataset)\n\u001b[1;32m 139\u001b[0m res \u001b[39m=\u001b[39m \u001b[39mfilter\u001b[39m(\u001b[39mlambda\u001b[39;00m o:o \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m, \u001b[39mmap\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdo_item, samps))\n\u001b[0;32m--> 140\u001b[0m \u001b[39myield from\u001b[39;00m \u001b[39mmap\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdo_batch, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mchunkify(res))\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastcore/basics.py:225\u001b[0m, in \u001b[0;36mchunked\u001b[0;34m(it, chunk_sz, drop_last, n_chunks)\u001b[0m\n\u001b[1;32m 223\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39misinstance\u001b[39m(it, Iterator): it \u001b[39m=\u001b[39m \u001b[39miter\u001b[39m(it)\n\u001b[1;32m 224\u001b[0m \u001b[39mwhile\u001b[39;00m \u001b[39mTrue\u001b[39;00m:\n\u001b[0;32m--> 225\u001b[0m res \u001b[39m=\u001b[39m \u001b[39mlist\u001b[39;49m(itertools\u001b[39m.\u001b[39;49mislice(it, chunk_sz))\n\u001b[1;32m 226\u001b[0m \u001b[39mif\u001b[39;00m res \u001b[39mand\u001b[39;00m (\u001b[39mlen\u001b[39m(res)\u001b[39m==\u001b[39mchunk_sz \u001b[39mor\u001b[39;00m \u001b[39mnot\u001b[39;00m drop_last): \u001b[39myield\u001b[39;00m res\n\u001b[1;32m 227\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mlen\u001b[39m(res)\u001b[39m<\u001b[39mchunk_sz: \u001b[39mreturn\u001b[39;00m\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/data/load.py:155\u001b[0m, in \u001b[0;36mDataLoader.do_item\u001b[0;34m(self, s)\u001b[0m\n\u001b[1;32m 154\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mdo_item\u001b[39m(\u001b[39mself\u001b[39m, s):\n\u001b[0;32m--> 155\u001b[0m \u001b[39mtry\u001b[39;00m: \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mafter_item(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mcreate_item(s))\n\u001b[1;32m 156\u001b[0m \u001b[39mexcept\u001b[39;00m SkipItemException: \u001b[39mreturn\u001b[39;00m \u001b[39mNone\u001b[39;00m\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/data/load.py:162\u001b[0m, in \u001b[0;36mDataLoader.create_item\u001b[0;34m(self, s)\u001b[0m\n\u001b[1;32m 161\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mcreate_item\u001b[39m(\u001b[39mself\u001b[39m, s):\n\u001b[0;32m--> 162\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mindexed: \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdataset[s \u001b[39mor\u001b[39;49;00m \u001b[39m0\u001b[39;49m]\n\u001b[1;32m 163\u001b[0m \u001b[39melif\u001b[39;00m s \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m: \u001b[39mreturn\u001b[39;00m \u001b[39mnext\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mit)\n\u001b[1;32m 164\u001b[0m \u001b[39melse\u001b[39;00m: \u001b[39mraise\u001b[39;00m \u001b[39mIndexError\u001b[39;00m(\u001b[39m\"\u001b[39m\u001b[39mCannot index an iterable dataset numerically - must use `None`.\u001b[39m\u001b[39m\"\u001b[39m)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/data/core.py:455\u001b[0m, in \u001b[0;36mDatasets.__getitem__\u001b[0;34m(self, it)\u001b[0m\n\u001b[1;32m 454\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__getitem__\u001b[39m(\u001b[39mself\u001b[39m, it):\n\u001b[0;32m--> 455\u001b[0m res \u001b[39m=\u001b[39m \u001b[39mtuple\u001b[39m([tl[it] \u001b[39mfor\u001b[39;00m tl \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mtls])\n\u001b[1;32m 456\u001b[0m \u001b[39mreturn\u001b[39;00m res \u001b[39mif\u001b[39;00m is_indexer(it) \u001b[39melse\u001b[39;00m \u001b[39mlist\u001b[39m(\u001b[39mzip\u001b[39m(\u001b[39m*\u001b[39mres))\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/data/core.py:455\u001b[0m, in \u001b[0;36m\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 454\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__getitem__\u001b[39m(\u001b[39mself\u001b[39m, it):\n\u001b[0;32m--> 455\u001b[0m res \u001b[39m=\u001b[39m \u001b[39mtuple\u001b[39m([tl[it] \u001b[39mfor\u001b[39;00m tl \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mtls])\n\u001b[1;32m 456\u001b[0m \u001b[39mreturn\u001b[39;00m res \u001b[39mif\u001b[39;00m is_indexer(it) \u001b[39melse\u001b[39;00m \u001b[39mlist\u001b[39m(\u001b[39mzip\u001b[39m(\u001b[39m*\u001b[39mres))\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/data/core.py:414\u001b[0m, in \u001b[0;36mTfmdLists.__getitem__\u001b[0;34m(self, idx)\u001b[0m\n\u001b[1;32m 412\u001b[0m res \u001b[39m=\u001b[39m \u001b[39msuper\u001b[39m()\u001b[39m.\u001b[39m\u001b[39m__getitem__\u001b[39m(idx)\n\u001b[1;32m 413\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_after_item \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m: \u001b[39mreturn\u001b[39;00m res\n\u001b[0;32m--> 414\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_after_item(res) \u001b[39mif\u001b[39;00m is_indexer(idx) \u001b[39melse\u001b[39;00m res\u001b[39m.\u001b[39mmap(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_after_item)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/data/core.py:374\u001b[0m, in \u001b[0;36mTfmdLists._after_item\u001b[0;34m(self, o)\u001b[0m\n\u001b[0;32m--> 374\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_after_item\u001b[39m(\u001b[39mself\u001b[39m, o): \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mtfms(o)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastcore/transform.py:208\u001b[0m, in \u001b[0;36mPipeline.__call__\u001b[0;34m(self, o)\u001b[0m\n\u001b[0;32m--> 208\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__call__\u001b[39m(\u001b[39mself\u001b[39m, o): \u001b[39mreturn\u001b[39;00m compose_tfms(o, tfms\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mfs, split_idx\u001b[39m=\u001b[39;49m\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49msplit_idx)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastcore/transform.py:158\u001b[0m, in \u001b[0;36mcompose_tfms\u001b[0;34m(x, tfms, is_enc, reverse, **kwargs)\u001b[0m\n\u001b[1;32m 156\u001b[0m \u001b[39mfor\u001b[39;00m f \u001b[39min\u001b[39;00m tfms:\n\u001b[1;32m 157\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m is_enc: f \u001b[39m=\u001b[39m f\u001b[39m.\u001b[39mdecode\n\u001b[0;32m--> 158\u001b[0m x \u001b[39m=\u001b[39m f(x, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n\u001b[1;32m 159\u001b[0m \u001b[39mreturn\u001b[39;00m x\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastcore/transform.py:81\u001b[0m, in \u001b[0;36mTransform.__call__\u001b[0;34m(self, x, **kwargs)\u001b[0m\n\u001b[0;32m---> 81\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__call__\u001b[39m(\u001b[39mself\u001b[39m, x, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs): \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_call(\u001b[39m'\u001b[39;49m\u001b[39mencodes\u001b[39;49m\u001b[39m'\u001b[39;49m, x, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastcore/transform.py:91\u001b[0m, in \u001b[0;36mTransform._call\u001b[0;34m(self, fn, x, split_idx, **kwargs)\u001b[0m\n\u001b[1;32m 89\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m_call\u001b[39m(\u001b[39mself\u001b[39m, fn, x, split_idx\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs):\n\u001b[1;32m 90\u001b[0m \u001b[39mif\u001b[39;00m split_idx\u001b[39m!=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39msplit_idx \u001b[39mand\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39msplit_idx \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m: \u001b[39mreturn\u001b[39;00m x\n\u001b[0;32m---> 91\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_do_call(\u001b[39mgetattr\u001b[39;49m(\u001b[39mself\u001b[39;49m, fn), x, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastcore/transform.py:97\u001b[0m, in \u001b[0;36mTransform._do_call\u001b[0;34m(self, f, x, **kwargs)\u001b[0m\n\u001b[1;32m 95\u001b[0m \u001b[39mif\u001b[39;00m f \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m: \u001b[39mreturn\u001b[39;00m x\n\u001b[1;32m 96\u001b[0m ret \u001b[39m=\u001b[39m f\u001b[39m.\u001b[39mreturns(x) \u001b[39mif\u001b[39;00m \u001b[39mhasattr\u001b[39m(f,\u001b[39m'\u001b[39m\u001b[39mreturns\u001b[39m\u001b[39m'\u001b[39m) \u001b[39melse\u001b[39;00m \u001b[39mNone\u001b[39;00m\n\u001b[0;32m---> 97\u001b[0m \u001b[39mreturn\u001b[39;00m retain_type(f(x, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs), x, ret)\n\u001b[1;32m 98\u001b[0m res \u001b[39m=\u001b[39m \u001b[39mtuple\u001b[39m(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_do_call(f, x_, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs) \u001b[39mfor\u001b[39;00m x_ \u001b[39min\u001b[39;00m x)\n\u001b[1;32m 99\u001b[0m \u001b[39mreturn\u001b[39;00m retain_type(res, x)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastcore/dispatch.py:120\u001b[0m, in \u001b[0;36mTypeDispatch.__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 118\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39minst \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m: f \u001b[39m=\u001b[39m MethodType(f, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39minst)\n\u001b[1;32m 119\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mowner \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m: f \u001b[39m=\u001b[39m MethodType(f, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mowner)\n\u001b[0;32m--> 120\u001b[0m \u001b[39mreturn\u001b[39;00m f(\u001b[39m*\u001b[39;49margs, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mkwargs)\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/fastai/vision/core.py:121\u001b[0m, in \u001b[0;36mPILBase.create\u001b[0;34m(cls, fn, **kwargs)\u001b[0m\n\u001b[1;32m 119\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(fn, TensorMask): fn \u001b[39m=\u001b[39m fn\u001b[39m.\u001b[39mtype(torch\u001b[39m.\u001b[39muint8)\n\u001b[1;32m 120\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(fn,Tensor): fn \u001b[39m=\u001b[39m fn\u001b[39m.\u001b[39mnumpy()\n\u001b[0;32m--> 121\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(fn,ndarray): \u001b[39mreturn\u001b[39;00m \u001b[39mcls\u001b[39m(Image\u001b[39m.\u001b[39;49mfromarray(fn))\n\u001b[1;32m 122\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(fn,\u001b[39mbytes\u001b[39m): fn \u001b[39m=\u001b[39m io\u001b[39m.\u001b[39mBytesIO(fn)\n\u001b[1;32m 123\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39mcls\u001b[39m(load_image(fn, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mmerge(\u001b[39mcls\u001b[39m\u001b[39m.\u001b[39m_open_args, kwargs)))\n", + "File \u001b[0;32m~/opt/anaconda3/envs/fastbook/lib/python3.9/site-packages/PIL/Image.py:2955\u001b[0m, in \u001b[0;36mfromarray\u001b[0;34m(obj, mode)\u001b[0m\n\u001b[1;32m 2953\u001b[0m mode, rawmode \u001b[39m=\u001b[39m _fromarray_typemap[typekey]\n\u001b[1;32m 2954\u001b[0m \u001b[39mexcept\u001b[39;00m \u001b[39mKeyError\u001b[39;00m \u001b[39mas\u001b[39;00m e:\n\u001b[0;32m-> 2955\u001b[0m \u001b[39mraise\u001b[39;00m \u001b[39mTypeError\u001b[39;00m(\u001b[39m\"\u001b[39m\u001b[39mCannot handle this data type: \u001b[39m\u001b[39m%s\u001b[39;00m\u001b[39m, \u001b[39m\u001b[39m%s\u001b[39;00m\u001b[39m\"\u001b[39m \u001b[39m%\u001b[39m typekey) \u001b[39mfrom\u001b[39;00m \u001b[39me\u001b[39;00m\n\u001b[1;32m 2956\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m 2957\u001b[0m rawmode \u001b[39m=\u001b[39m mode\n", + "\u001b[0;31mTypeError\u001b[0m: Cannot handle this data type: (1, 1, 1540),