@tvm.script.ir_module\n",
"class Module:\n",
" @T.prim_func\n",
" def main(A: T.Buffer[(128, 128), "float32"], B: T.Buffer[(128, 128), "float32"], C: T.Buffer[(128, 128), "float32"]) -> None:\n",
" # function attr dict\n",
" T.func_attr({"global_symbol": "main", "tir.noalias": True})\n",
" # body\n",
" # with T.block("root")\n",
" for i, j_0 in T.grid(128, 32):\n",
" for j_1_init in T.serial(4):\n",
" with T.block("C_init"):\n",
" vi = T.axis.spatial(128, i)\n",
" vj = T.axis.spatial(128, j_0 * 4 + j_1_init)\n",
" T.reads()\n",
" T.writes(C[vi, vj])\n",
" C[vi, vj] = T.float32(0)\n",
" for k, j_1 in T.grid(128, 4):\n",
" with T.block("C_update"):\n",
" vi = T.axis.spatial(128, i)\n",
" vj = T.axis.spatial(128, j_0 * 4 + j_1)\n",
" vk = T.axis.reduce(128, k)\n",
" T.reads(C[vi, vj], A[vi, vk], B[vk, vj])\n",
" T.writes(C[vi, vj])\n",
" C[vi, vj] = C[vi, vj] + A[vi, vk] * B[vk, vj]\n",
" \n",
"@tvm.script.ir_module\n",
"class Module:\n",
" @T.prim_func\n",
" def main(A: T.Buffer[(128, 128), "float32"], B: T.Buffer[(128, 128), "float32"], C: T.Buffer[(128, 128), "float32"]) -> None:\n",
" # function attr dict\n",
" T.func_attr({"global_symbol": "main", "tir.noalias": True})\n",
" # body\n",
" # with T.block("root")\n",
" for i, j_0 in T.grid(128, 32):\n",
" for j_1_init in T.serial(4):\n",
" with T.block("C_init"):\n",
" vi = T.axis.spatial(128, i)\n",
" vj = T.axis.spatial(128, j_0 * 4 + j_1_init)\n",
" T.reads()\n",
" T.writes(C[vi, vj])\n",
" C[vi, vj] = T.float32(0)\n",
" for k, j_1 in T.grid(128, 4):\n",
" with T.block("C_update"):\n",
" vi = T.axis.spatial(128, i)\n",
" vj = T.axis.spatial(128, j_0 * 4 + j_1)\n",
" vk = T.axis.reduce(128, k)\n",
" T.reads(C[vi, vj], A[vi, vk], B[vk, vj])\n",
" T.writes(C[vi, vj])\n",
" C[vi, vj] = C[vi, vj] + A[vi, vk] * B[vk, vj]\n",
" \n",
"@tvm.script.ir_module\n",
"class Module:\n",
" @T.prim_func\n",
" def main(A: T.Buffer[(128, 128), "float32"], B: T.Buffer[(128, 128), "float32"], C: T.Buffer[(128, 128), "float32"]) -> None:\n",
" # function attr dict\n",
" T.func_attr({"global_symbol": "main", "tir.noalias": True})\n",
" # body\n",
" # with T.block("root")\n",
" for i, j, k in T.grid(128, 128, 128):\n",
" with T.block("C"):\n",
" vi, vj, vk = T.axis.remap("SSR", [i, j, k])\n",
" T.reads(A[vi, vk], B[vk, vj])\n",
" T.writes(C[vi, vj])\n",
" with T.init():\n",
" C[vi, vj] = T.float32(0)\n",
" C[vi, vj] = C[vi, vj] + A[vi, vk] * B[vk, vj]\n",
" \n",
"@tvm.script.ir_module\n",
"class Module:\n",
" @T.prim_func\n",
" def main(A: T.Buffer[(128, 128), "float32"], B: T.Buffer[(128, 128), "float32"], C: T.Buffer[(128, 128), "float32"]) -> None:\n",
" # function attr dict\n",
" T.func_attr({"global_symbol": "main", "tir.noalias": True})\n",
" # body\n",
" # with T.block("root")\n",
" for i, j_0, k, j_1 in T.grid(128, 8, 128, 16):\n",
" with T.block("C"):\n",
" vi = T.axis.spatial(128, i)\n",
" vj = T.axis.spatial(128, j_0 * 16 + j_1)\n",
" vk = T.axis.reduce(128, k)\n",
" T.reads(A[vi, vk], B[vk, vj])\n",
" T.writes(C[vi, vj])\n",
" with T.init():\n",
" C[vi, vj] = T.float32(0)\n",
" C[vi, vj] = C[vi, vj] + A[vi, vk] * B[vk, vj]\n",
" \n",
"@tvm.script.ir_module\n",
"class Module:\n",
" @T.prim_func\n",
" def main(A: T.Buffer[(128, 128), "float32"], B: T.Buffer[(128, 128), "float32"], C: T.Buffer[(128, 128), "float32"]) -> None:\n",
" # function attr dict\n",
" T.func_attr({"global_symbol": "main", "tir.noalias": True})\n",
" # body\n",
" # with T.block("root")\n",
" for i, j_0 in T.grid(128, 8):\n",
" for j_1_init in T.serial(16):\n",
" with T.block("C_init"):\n",
" vi = T.axis.spatial(128, i)\n",
" vj = T.axis.spatial(128, j_0 * 16 + j_1_init)\n",
" T.reads()\n",
" T.writes(C[vi, vj])\n",
" C[vi, vj] = T.float32(0)\n",
" for k, j_1 in T.grid(128, 16):\n",
" with T.block("C_update"):\n",
" vi = T.axis.spatial(128, i)\n",
" vj = T.axis.spatial(128, j_0 * 16 + j_1)\n",
" vk = T.axis.reduce(128, k)\n",
" T.reads(C[vi, vj], A[vi, vk], B[vk, vj])\n",
" T.writes(C[vi, vj])\n",
" C[vi, vj] = C[vi, vj] + A[vi, vk] * B[vk, vj]\n",
" \n",
"\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[39mfrom\u001b[39;00m \u001b[39mtvm\u001b[39;00m \u001b[39mimport\u001b[39;00m meta_schedule \u001b[39mas\u001b[39;00m ms\n\u001b[0;32m----> 3\u001b[0m sch_tuned \u001b[39m=\u001b[39m ms\u001b[39m.\u001b[39;49mtune_tir(\n\u001b[1;32m 4\u001b[0m mod\u001b[39m=\u001b[39;49mMyModule,\n\u001b[1;32m 5\u001b[0m target\u001b[39m=\u001b[39;49m\u001b[39m\"\u001b[39;49m\u001b[39mllvm --num-cores=1\u001b[39;49m\u001b[39m\"\u001b[39;49m,\n\u001b[1;32m 6\u001b[0m config\u001b[39m=\u001b[39;49mms\u001b[39m.\u001b[39;49mTuneConfig(\n\u001b[1;32m 7\u001b[0m max_trials_global\u001b[39m=\u001b[39;49m\u001b[39m64\u001b[39;49m,\n\u001b[1;32m 8\u001b[0m num_trials_per_iter\u001b[39m=\u001b[39;49m\u001b[39m64\u001b[39;49m,\n\u001b[1;32m 9\u001b[0m ),\n\u001b[1;32m 10\u001b[0m space\u001b[39m=\u001b[39;49mms\u001b[39m.\u001b[39;49mspace_generator\u001b[39m.\u001b[39;49mScheduleFn(stochastic_schedule_mm),\n\u001b[1;32m 11\u001b[0m work_dir\u001b[39m=\u001b[39;49m\u001b[39m\"\u001b[39;49m\u001b[39m./tune_tmp\u001b[39;49m\u001b[39m\"\u001b[39;49m,\n\u001b[1;32m 12\u001b[0m task_name\u001b[39m=\u001b[39;49m\u001b[39m\"\u001b[39;49m\u001b[39mmain\u001b[39;49m\u001b[39m\"\u001b[39;49m\n\u001b[1;32m 13\u001b[0m )\n",
"File \u001b[0;32m/media/pc/data/4tb/lxw/libs/anaconda3/envs/mlc/lib/python3.10/site-packages/tvm/meta_schedule/tune.py:414\u001b[0m, in \u001b[0;36mtune_tir\u001b[0;34m(mod, target, config, work_dir, builder, runner, database, cost_model, measure_callbacks, space, sch_rules, postprocs, mutator_probs, task_name, num_threads)\u001b[0m\n\u001b[1;32m 412\u001b[0m target \u001b[39m=\u001b[39m default_config\u001b[39m.\u001b[39mtarget(target)\n\u001b[1;32m 413\u001b[0m \u001b[39m# pylint: enable=protected-access\u001b[39;00m\n\u001b[0;32m--> 414\u001b[0m database \u001b[39m=\u001b[39m tune_extracted_tasks(\n\u001b[1;32m 415\u001b[0m extracted_tasks\u001b[39m=\u001b[39;49m[\n\u001b[1;32m 416\u001b[0m ExtractedTask(\n\u001b[1;32m 417\u001b[0m task_name\u001b[39m=\u001b[39;49mtask_name,\n\u001b[1;32m 418\u001b[0m mod\u001b[39m=\u001b[39;49mmod,\n\u001b[1;32m 419\u001b[0m dispatched\u001b[39m=\u001b[39;49m[mod],\n\u001b[1;32m 420\u001b[0m target\u001b[39m=\u001b[39;49mtarget,\n\u001b[1;32m 421\u001b[0m weight\u001b[39m=\u001b[39;49m\u001b[39m1\u001b[39;49m,\n\u001b[1;32m 422\u001b[0m ),\n\u001b[1;32m 423\u001b[0m ],\n\u001b[1;32m 424\u001b[0m config\u001b[39m=\u001b[39;49mconfig,\n\u001b[1;32m 425\u001b[0m work_dir\u001b[39m=\u001b[39;49mwork_dir,\n\u001b[1;32m 426\u001b[0m builder\u001b[39m=\u001b[39;49mbuilder,\n\u001b[1;32m 427\u001b[0m runner\u001b[39m=\u001b[39;49mrunner,\n\u001b[1;32m 428\u001b[0m database\u001b[39m=\u001b[39;49mdatabase,\n\u001b[1;32m 429\u001b[0m cost_model\u001b[39m=\u001b[39;49mcost_model,\n\u001b[1;32m 430\u001b[0m measure_callbacks\u001b[39m=\u001b[39;49mmeasure_callbacks,\n\u001b[1;32m 431\u001b[0m space\u001b[39m=\u001b[39;49mspace,\n\u001b[1;32m 432\u001b[0m sch_rules\u001b[39m=\u001b[39;49msch_rules,\n\u001b[1;32m 433\u001b[0m postprocs\u001b[39m=\u001b[39;49mpostprocs,\n\u001b[1;32m 434\u001b[0m mutator_probs\u001b[39m=\u001b[39;49mmutator_probs,\n\u001b[1;32m 435\u001b[0m num_threads\u001b[39m=\u001b[39;49mnum_threads,\n\u001b[1;32m 436\u001b[0m )\n\u001b[1;32m 437\u001b[0m \u001b[39mwith\u001b[39;00m Profiler\u001b[39m.\u001b[39mtimeit(\u001b[39m\"\u001b[39m\u001b[39mApplyHistoryBest\u001b[39m\u001b[39m\"\u001b[39m):\n\u001b[1;32m 438\u001b[0m bests: List[TuningRecord] \u001b[39m=\u001b[39m database\u001b[39m.\u001b[39mget_top_k(database\u001b[39m.\u001b[39mcommit_workload(mod), top_k\u001b[39m=\u001b[39m\u001b[39m1\u001b[39m)\n",
"File \u001b[0;32m/media/pc/data/4tb/lxw/libs/anaconda3/envs/mlc/lib/python3.10/site-packages/tvm/meta_schedule/tune.py:350\u001b[0m, in \u001b[0;36mtune_extracted_tasks\u001b[0;34m(extracted_tasks, config, work_dir, builder, runner, database, cost_model, measure_callbacks, space, sch_rules, postprocs, mutator_probs, num_threads)\u001b[0m\n\u001b[1;32m 340\u001b[0m task_scheduler \u001b[39m=\u001b[39m config\u001b[39m.\u001b[39mcreate_task_scheduler(\n\u001b[1;32m 341\u001b[0m tasks\u001b[39m=\u001b[39mtune_contexts,\n\u001b[1;32m 342\u001b[0m task_weights\u001b[39m=\u001b[39m[\u001b[39mfloat\u001b[39m(t\u001b[39m.\u001b[39mweight) \u001b[39mfor\u001b[39;00m t \u001b[39min\u001b[39;00m extracted_tasks],\n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 347\u001b[0m measure_callbacks\u001b[39m=\u001b[39mmeasure_callbacks,\n\u001b[1;32m 348\u001b[0m )\n\u001b[1;32m 349\u001b[0m \u001b[39mif\u001b[39;00m config\u001b[39m.\u001b[39mmax_trials_global \u001b[39m>\u001b[39m \u001b[39m0\u001b[39m:\n\u001b[0;32m--> 350\u001b[0m task_scheduler\u001b[39m.\u001b[39;49mtune()\n\u001b[1;32m 351\u001b[0m cost_model\u001b[39m.\u001b[39msave(osp\u001b[39m.\u001b[39mjoin(work_dir, \u001b[39m\"\u001b[39m\u001b[39mcost_model.xgb\u001b[39m\u001b[39m\"\u001b[39m))\n\u001b[1;32m 352\u001b[0m \u001b[39mreturn\u001b[39;00m database\n",
"File \u001b[0;32m/media/pc/data/4tb/lxw/libs/anaconda3/envs/mlc/lib/python3.10/site-packages/tvm/meta_schedule/task_scheduler/task_scheduler.py:72\u001b[0m, in \u001b[0;36mTaskScheduler.tune\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 70\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mtune\u001b[39m(\u001b[39mself\u001b[39m) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m 71\u001b[0m \u001b[39m\"\"\"Auto-tuning.\"\"\"\u001b[39;00m\n\u001b[0;32m---> 72\u001b[0m _ffi_api\u001b[39m.\u001b[39;49mTaskSchedulerTune(\u001b[39mself\u001b[39;49m)\n",
"File \u001b[0;32m/media/pc/data/4tb/lxw/libs/anaconda3/envs/mlc/lib/python3.10/site-packages/tvm/_ffi/_ctypes/packed_func.py:237\u001b[0m, in \u001b[0;36mPackedFuncBase.__call__\u001b[0;34m(self, *args)\u001b[0m\n\u001b[1;32m 225\u001b[0m ret_tcode \u001b[39m=\u001b[39m ctypes\u001b[39m.\u001b[39mc_int()\n\u001b[1;32m 226\u001b[0m \u001b[39mif\u001b[39;00m (\n\u001b[1;32m 227\u001b[0m _LIB\u001b[39m.\u001b[39mTVMFuncCall(\n\u001b[1;32m 228\u001b[0m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mhandle,\n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 235\u001b[0m \u001b[39m!=\u001b[39m \u001b[39m0\u001b[39m\n\u001b[1;32m 236\u001b[0m ):\n\u001b[0;32m--> 237\u001b[0m \u001b[39mraise\u001b[39;00m get_last_ffi_error()\n\u001b[1;32m 238\u001b[0m _ \u001b[39m=\u001b[39m temp_args\n\u001b[1;32m 239\u001b[0m _ \u001b[39m=\u001b[39m args\n",
"\u001b[0;31mTVMError\u001b[0m: Traceback (most recent call last):\n 8: TVMFuncCall\n 7: tvm::runtime::PackedFuncObj::Extractor::AssignTypedLambda(void (tvm::meta_schedule::TaskSchedulerNode::*)())::{lambda(tvm::meta_schedule::TaskScheduler)#1}>(tvm::runtime::Registry::set_body_method(void (tvm::meta_schedule::TaskSchedulerNode::*)())::{lambda(tvm::meta_schedule::TaskScheduler)#1}, std::string)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*)\n 6: tvm::meta_schedule::TaskSchedulerNode::Tune()\n 5: tvm::meta_schedule::GradientBasedNode::NextTaskId()\n 4: tvm::meta_schedule::GradientBasedNode::JoinRunningTask(int)\n 3: tvm::meta_schedule::TaskSchedulerNode::JoinRunningTask(int)\n 2: tvm::meta_schedule::UpdateCostModelNode::Apply(tvm::meta_schedule::TaskScheduler const&, int, tvm::runtime::Array const&, tvm::runtime::Array const&, tvm::runtime::Array const&)\n 1: tvm::meta_schedule::PyCostModelNode::Update(tvm::meta_schedule::TuneContext const&, tvm::runtime::Array const&, tvm::runtime::Array const&)\n 0: tvm::runtime::PackedFuncObj::Extractor >::Call(tvm::runtime::PackedFuncObj const*, tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) [clone .cold]\n File \"/media/pc/data/4tb/lxw/libs/anaconda3/envs/mlc/lib/python3.10/site-packages/tvm/_ffi/_ctypes/packed_func.py\", line 81, in cfun\n rv = local_pyfunc(*pyargs)\n File \"/media/pc/data/4tb/lxw/libs/anaconda3/envs/mlc/lib/python3.10/site-packages/tvm/meta_schedule/utils.py\", line 77, in method\n return getattr(inst, name)(*args, **kwargs)\n File \"/media/pc/data/4tb/lxw/libs/anaconda3/envs/mlc/lib/python3.10/site-packages/tvm/meta_schedule/cost_model/xgb_model.py\", line 513, in update\n self._train(\n File \"/media/pc/data/4tb/lxw/libs/anaconda3/envs/mlc/lib/python3.10/site-packages/tvm/meta_schedule/cost_model/xgb_model.py\", line 563, in _train\n import xgboost as xgb # type: ignore # pylint: disable=import-outside-toplevel\nModuleNotFoundError: No module named 'xgboost'"
]
}
],
"source": [
"from tvm import meta_schedule as ms\n",
"\n",
"sch_tuned = ms.tune_tir(\n",
" mod=MyModule,\n",
" target=\"llvm --num-cores=1\",\n",
" config=ms.TuneConfig(\n",
" max_trials_global=64,\n",
" num_trials_per_iter=64,\n",
" ),\n",
" space=ms.space_generator.ScheduleFn(stochastic_schedule_mm),\n",
" work_dir=\"./tune_tmp\",\n",
" task_name=\"main\"\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "2au1M-kptntR"
},
"source": [
"`tune_tir` functions return an optimized schedule found during the tuning process."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "mCbI8ulnxnvN",
"outputId": "fe3796c3-821f-410e-9af1-93675a233994"
},
"outputs": [],
"source": [
"print(sch_tuned.trace)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 451
},
"id": "lHE8cU0Ww9W6",
"outputId": "3c70a9f0-5470-4b16-8f8b-b34cbf793a1c"
},
"outputs": [],
"source": [
"IPython.display.HTML(code2html(sch_tuned.mod.script()))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "CFngbVbjKrKd",
"outputId": "5c10905a-4a26-48dd-b68a-466e52b6b178"
},
"outputs": [],
"source": [
"lib = tvm.build(sch_tuned.mod, target=\"llvm\")\n",
"f_timer_after = lib.time_evaluator(\"main\", tvm.cpu())\n",
"print(\"Time cost of MyModule after tuning: %.3f ms\" % (f_timer_after(a_nd, b_nd, c_nd).mean * 1000))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "GvBUfh0RNFfJ"
},
"source": [
"### Leverage Default AutoScheduling \n",
"\n",
"In the last section, we showed how to tune a workload with stochastic transformations that we crafted. Metaschedule comes with its own built-in set of generic stochastic transformations that works for a broad set of TensorIR computations. This approach is also called auto-scheduling, as the search space is generated by the system. We can run it by removing the line `space=ms.space_generator.ScheduleFn(stochastic_schedule_mm)`.\n",
"\n",
"Under the hood, the meta-scheduler analyzes each block's data access and loop patterns and proposes stochastic transformations to the program. We won't go into these generic transformations in this chapter but want to note that they are also just stochastic transformations coupled with an analysis of the code. We can use the same mechanisms learned in the last section to enhance auto-scheduling. We will touch base on this topic in future chapters."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "dWHos7P_M_KJ",
"outputId": "23e2b866-41b1-4c51-cea4-c16bb34ab6fb"
},
"outputs": [],
"source": [
"sch_tuned = ms.tune_tir(\n",
" mod=MyModule,\n",
" target=\"llvm --num-cores=1\",\n",
" config=ms.TuneConfig(\n",
" max_trials_global=64,\n",
" num_trials_per_iter=64,\n",
" ),\n",
" work_dir=\"./tune_tmp\",\n",
" task_name=\"main\",\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "1YEebX59NobA",
"outputId": "9986b79c-445c-49ac-fba3-fff459e0abc4"
},
"outputs": [],
"source": [
"lib = tvm.build(sch_tuned.mod, target=\"llvm\")\n",
"f_timer_after = lib.time_evaluator(\"main\", tvm.cpu())\n",
"print(\"Time cost of MyModule after tuning: %.3f ms\" % (f_timer_after(a_nd, b_nd, c_nd).mean * 1000))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "6IEPOVDmArG5"
},
"source": [
"The result gets much faster than our original code. We can take a glimpse at the trace and the final code. For the purpose of this chapter, you do not need to understand all the transformations. At a high level, the trace involves:\n",
"\n",
"- More levels of loop tiling transformations.\n",
"- Vectorization of intermediate computations.\n",
"- Parallelization and unrolling of loops. \n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "-AskZWy_NkAL",
"outputId": "a6b41032-8867-4a2c-f9fd-985e25cf5044"
},
"outputs": [],
"source": [
"sch_tuned.trace"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 697
},
"id": "Q1f11ArzNLZy",
"outputId": "8034e8a6-038e-4e60-be42-860fe6dd866b"
},
"outputs": [],
"source": [
"IPython.display.HTML(code2html(sch_tuned.mod.script()))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "HCFjRzmBLyXB"
},
"source": [
"### Section Checkpoint\n",
"\n",
"Let us have a checkpoint about what we have learned so far.\n",
"\n",
"- Stochastic schedule allow us to express \"what are the possible transformations\".\n",
"- Metaschedule's `tune_tir` API helps to find a good solution within the space.\n",
"- Metaschedule comes with a default built-in set of stochastic transformations that covers a broad range of search space."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "wIIsUXGqpRqV"
},
"source": [
"## Putting Things Back to End to End Model Execution\n",
"\n",
"Up until now, we have learned to automate program optimization of a single tensor primitive function. How can we put it back and improve our end-to-end model execution?\n",
"\n",
"From the MLC perspective, the automated search is a modular step, and we just need to replace the original primitive function implementation with the new one provided by the tuned result.\n",
"\n",
"We will reuse the two-layer MLP example from the last chapter.\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 451,
"referenced_widgets": [
"0e91dc133cd440f9973acd52f341b532",
"de8f202890da499fa532e39882aa39ad",
"4cf4c23aaea549daa6ec02712c5229f7",
"42dbf79dd8804855a48940bd61adfd8c",
"9d928a84ddb2401eb8ac38b1905167b9",
"dd328252b2bf4ec994e13e8e6dba2875",
"808f5d3718ba40ad80882d8afcc6b70f",
"20e53fd4a49d4188a83fb152fac32a20",
"5edd0c12e56a4179be3775303eeecb5d",
"1dec5c156112404e84fd30a8f2bb6dda",
"2441c125a1b645b6bb7124fdc667a6ac",
"82820ce83cfb4c38bdf36e249b4c7d90",
"de6746f353614fa2b61e007c811dc6cf",
"46e69b009d8d419b854c789933bb31cc",
"c265c2e58744434a9896884b4ab8f555",
"49f847d7f2ae48d5a274bcba1b352527",
"f465c7c665874e06875e5564af85704e",
"f58599fc1d6c4308a21aafb5b15a7f7a",
"ab6012bbe9a54f23aeee9fc8985b0cd7",
"f2efbbf0de6f4245b5ed7b29d62aee7e",
"2afdb5134ffd48b5a3ea36fdf13f3ca9",
"5d03cf1d574f4319a78fd194c1802506",
"acfbdfba8a594540b7f809ad5a490bee",
"dfeb1006d236400cbb3d0e304c8208af",
"aabc8a2be63b4dcfabe841fb2b22fd95",
"fb7da5344fdd4e8d9f49da31fa32169a",
"dbb95703595642fbb2979a63e1391149",
"de6d869bae854d598d06dc7d17bb16b6",
"d6ec914f5ca44a7f884d4b20e901ecf8",
"401ac5195b714b7bae27caddb02d9dae",
"edb287c8f5e04532ad5c3760041491e2",
"418492022d584305b781ef243b35213a",
"abdf432283ea48409849a131e3c84538",
"463fc2365d084c03bb57e60310ef7797",
"5b5d9289df9b40a89e34ce048dfa0ece",
"7bd374159df44db9a117e97583427122",
"014e097bb2c540f6836c61175a9bcba5",
"7cb25b7ea7894bf999948823015a42c1",
"8b87c5c46b2a496e8fce50e8d94f0364",
"4b8f0c2e31fe4218902f572cdc73e9d2",
"8cdfa0916bf6417a91fb4b19c44d0021",
"eeb96b697cfa48b5add99d3931574ecb",
"b844d16570ce447598ddcf3ec084608a",
"dd2d6dfae7f34fa2af3a4f89bd7660bb"
]
},
"id": "NdWS5Jabq-DN",
"outputId": "26d68134-b807-4894-8bff-989724bba61e"
},
"outputs": [],
"source": [
"import torchvision\n",
"import torch\n",
"test_data = torchvision.datasets.FashionMNIST(\n",
" root=\"data\",\n",
" train=False,\n",
" download=True,\n",
" transform=torchvision.transforms.ToTensor()\n",
")\n",
"test_loader = torch.utils.data.DataLoader(test_data, batch_size=1, shuffle=True)\n",
"class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',\n",
" 'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']\n",
"\n",
"img, label = next(iter(test_loader))\n",
"img = img.reshape(1, 28, 28).numpy()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 287
},
"id": "Aht3W8MNkSf_",
"outputId": "efffedc0-1918-4cc1-8ea5-d67159eda774"
},
"outputs": [],
"source": [
"import matplotlib.pyplot as plt\n",
"\n",
"plt.figure()\n",
"plt.imshow(img[0])\n",
"plt.colorbar()\n",
"plt.grid(False)\n",
"plt.show()\n",
"\n",
"print(\"Class:\", class_names[label[0]])"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "dPR_WrTZglbh"
},
"source": [
"\n",
"We also download pre-packed model parameters that we will use in our examples."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "7Wl0niRLgqYb",
"outputId": "9485252e-f16c-4f08-823c-4c98c4b399eb"
},
"outputs": [],
"source": [
"!wget https://github.com/mlc-ai/web-data/raw/main/models/fasionmnist_mlp_params.pkl"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "hk64a3UllGIV"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "PUcCRU2IQPm-"
},
"source": [
"As a reminder, the above figure shows the model of interest."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "hU1NwPX7M0OA"
},
"outputs": [],
"source": [
"import pickle as pkl\n",
"mlp_params = pkl.load(open(\"fasionmnist_mlp_params.pkl\", \"rb\"))\n",
"\n",
"data_nd = tvm.nd.array(img.reshape(1, 784))\n",
"nd_params = {k: tvm.nd.array(v) for k, v in mlp_params.items()}"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "uVmYbaRPDSdr"
},
"source": [
"Let us use a mixture module where most of the components call into environment function and also come with one TensorIR function `linear0`."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "qkXWaLfzN4En"
},
"outputs": [],
"source": [
"@tvm.script.ir_module\n",
"class MyModuleMixture: \n",
" @T.prim_func\n",
" def linear0(X: T.Buffer[(1, 784), \"float32\"], \n",
" W: T.Buffer[(128, 784), \"float32\"], \n",
" B: T.Buffer[(128,), \"float32\"], \n",
" Z: T.Buffer[(1, 128), \"float32\"]):\n",
" T.func_attr({\"global_symbol\": \"linear0\", \"tir.noalias\": True})\n",
" Y = T.alloc_buffer((1, 128), \"float32\")\n",
" for i, j, k in T.grid(1, 128, 784):\n",
" with T.block(\"Y\"):\n",
" vi, vj, vk = T.axis.remap(\"SSR\", [i, j, k])\n",
" with T.init():\n",
" Y[vi, vj] = T.float32(0)\n",
" Y[vi, vj] = Y[vi, vj] + X[vi, vk] * W[vj, vk]\n",
" \n",
" for i, j in T.grid(1, 128):\n",
" with T.block(\"Z\"):\n",
" vi, vj = T.axis.remap(\"SS\", [i, j])\n",
" Z[vi, vj] = Y[vi, vj] + B[vj]\n",
"\n",
" @R.function\n",
" def main(x: Tensor((1, 784), \"float32\"), \n",
" w0: Tensor((128, 784), \"float32\"), \n",
" b0: Tensor((128,), \"float32\"), \n",
" w1: Tensor((10, 128), \"float32\"), \n",
" b1: Tensor((10,), \"float32\")):\n",
" with R.dataflow():\n",
" lv0 = R.call_tir(linear0, (x, w0, b0), (1, 128), dtype=\"float32\")\n",
" lv1 = R.call_tir(\"env.relu\", (lv0,), (1, 128), dtype=\"float32\")\n",
" out = R.call_tir(\"env.linear\", (lv1, w1, b1), (1, 10), dtype=\"float32\")\n",
" R.output(out)\n",
" return out\n",
" "
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "fCl3PnjWOvXf"
},
"outputs": [],
"source": [
"@tvm.register_func(\"env.linear\", override=True)\n",
"def torch_linear(x: tvm.nd.NDArray, \n",
" w: tvm.nd.NDArray, \n",
" b: tvm.nd.NDArray, \n",
" out: tvm.nd.NDArray):\n",
" x_torch = torch.from_dlpack(x)\n",
" w_torch = torch.from_dlpack(w)\n",
" b_torch = torch.from_dlpack(b)\n",
" out_torch = torch.from_dlpack(out)\n",
" torch.mm(x_torch, w_torch.T, out=out_torch)\n",
" torch.add(out_torch, b_torch, out=out_torch)\n",
"\n",
"@tvm.register_func(\"env.relu\", override=True)\n",
"def lnumpy_relu(x: tvm.nd.NDArray, \n",
" out: tvm.nd.NDArray):\n",
" x_torch = torch.from_dlpack(x)\n",
" out_torch = torch.from_dlpack(out)\n",
" torch.maximum(x_torch, torch.Tensor([0.0]), out=out_torch)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "cqCpSb2EDgUB"
},
"source": [
"We can bind the parameters and see if it gives the correct prediction."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "ONLqCQ3qOHlB"
},
"outputs": [],
"source": [
"MyModuleWithParams = relax.transform.BindParams(\"main\", nd_params)(MyModuleMixture)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "v-9mB9SEOatM",
"outputId": "c9ef0713-7896-445c-edfc-f5f8df20fa02"
},
"outputs": [],
"source": [
"ex = relax.vm.build(MyModuleWithParams, target=\"llvm\")\n",
"vm = relax.VirtualMachine(ex, tvm.cpu())\n",
"\n",
"nd_res = vm[\"main\"](data_nd)\n",
"\n",
"pred_kind = np.argmax(nd_res.numpy(), axis=1)\n",
"print(\"MyModuleWithParams Prediction:\", class_names[pred_kind[0]])"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "UAdTWYg3DkPU"
},
"source": [
"The following code evaluates the run time cost of the module before the transformation. Note that because this is a small model, the number can fluctuate a bit between runs, so we just need to read the overall magnitude."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "HrMoMP08OxYJ",
"outputId": "2a658837-67f5-4298-8828-becf5201392f"
},
"outputs": [],
"source": [
"ftimer = vm.module.time_evaluator(\"main\", tvm.cpu(), number=100)\n",
"\n",
"print(\"MyModuleWithParams time-cost: %g ms\" % (ftimer(data_nd).mean * 1000))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "y1ei5bAzD0O4"
},
"source": [
"We are now ready to tune the `linear0`. Our overall process is summarized in the following diagram."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Sxg7AOW3ZKH3"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Gd2Io-enZblB"
},
"source": [
"Currently, tune API only takes an IRModule with one `main` function, so we first get the `linear0` out into another module's main function and pass it to tune"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 454
},
"id": "lbl1l_TuU8Ug",
"outputId": "a335f2d0-f9eb-4b83-89b3-9e87343838b1"
},
"outputs": [],
"source": [
"mod_linear = tvm.IRModule.from_expr(MyModuleMixture[\"linear0\"].with_attr(\"global_symbol\", \"main\"))\n",
"IPython.display.HTML(code2html(mod_linear.script()))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "vdZwiQK8VSlG",
"outputId": "d8df5aa4-423b-4d80-e36b-03ac98d5bd64"
},
"outputs": [],
"source": [
"sch_tuned_linear = ms.tune_tir(\n",
" mod=mod_linear,\n",
" target=\"llvm --num-cores=1\",\n",
" config=ms.TuneConfig(\n",
" max_trials_global=64,\n",
" num_trials_per_iter=64,\n",
" ),\n",
" work_dir=\"./tune_tmp\",\n",
" task_name=\"main\",\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "WgfZZ2SdEhMV"
},
"source": [
"Now we need to replace the original `linear0` with the new function after tuning. We can do that by first getting a `global_var`, a `pointer` reference to the functions inside the IRModule, then calling `update_func` to replace the function with the new one."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 853
},
"id": "nZh9QOgSn013",
"outputId": "0f3580fd-b4f9-468a-d3e1-246e767420b1"
},
"outputs": [],
"source": [
"MyModuleWithParams2 = relax.transform.BindParams(\"main\", nd_params)(MyModuleMixture)\n",
"new_func = sch_tuned_linear.mod[\"main\"].with_attr(\"global_symbol\", \"linear0\")\n",
"gv = MyModuleWithParams2.get_global_var(\"linear0\")\n",
"MyModuleWithParams2.update_func(gv, new_func)\n",
"IPython.display.HTML(code2html(MyModuleWithParams2.script()))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "O6uHjCqMEyT_"
},
"source": [
"We can find that the `linear0` has been replaced in the above code."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "4DPNPP4DpKI6",
"outputId": "564fb257-a734-4e8b-843c-c4812ec13e4c"
},
"outputs": [],
"source": [
"ex = relax.vm.build(MyModuleWithParams2, target=\"llvm\")\n",
"vm = relax.VirtualMachine(ex, tvm.cpu())\n",
"\n",
"nd_res = vm[\"main\"](data_nd)\n",
"\n",
"pred_kind = np.argmax(nd_res.numpy(), axis=1)\n",
"print(\"MyModuleWithParams2 Prediction:\", class_names[pred_kind[0]])"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "HyAAkBx-FKgT"
},
"source": [
"Running the code again, we can find that we get an observable amount of time reduction, mainly thanks to the new `linear0` function."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "Oo87n_-HozOV",
"outputId": "0e327404-1507-4533-906c-88f9685bebbd"
},
"outputs": [],
"source": [
"ftimer = vm.module.time_evaluator(\"main\", tvm.cpu(), number=50)\n",
"\n",
"print(\"MyModuleWithParams2 time-cost: %g ms\" % (ftimer(data_nd).mean * 1000))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "bMHXI_ccFS-z"
},
"source": [
"## Discussions\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "VkB5oHttOY0U"
},
"source": [
"We might notice that our previous two chapters focused on **abstraction** while this chapter starts to focus on **transformation**. Stochastic transformations specify what can be possibly optimized without nailing down all the choices. The meta-schedule API helps us to search over the space of possible transformations and pick the best one.\n",
"\n",
"Importantly, putting the search result back into the end-to-end flow is just a matter of replacing the implementation of the original function with a new one that is informed by the tuning process. \n",
"\n",
"So we again are following the generic MLC process in the figure below. In future lectures, we will introduce more kinds of transformations on primitive functions and computational graph functions. A good MLC process composes these transformations together to form an end deployment form."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "y2KrBILMsNGf"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "pZGWq5EjJ0BA"
},
"source": [
"## Summary\n",
"\n",
"- Stochastic transformations help us to specify a search space of possible programs.\n",
"- MetaSchedule searches over the search space and finds an optimized one.\n",
"- We can use another transformation to replace the primitive tensor function with optimized ones and an updated end-to-end execution flow."
]
}
],
"metadata": {
"colab": {
"collapsed_sections": [],
"include_colab_link": true,
"name": "5 Automatic Program Optimization.ipynb",
"provenance": []
},
"kernelspec": {
"display_name": "Python 3.10.4 ('mlc': conda)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.4"
},
"vscode": {
"interpreter": {
"hash": "d8a760899c905ec5a15e0d212432af25d7f0b614c7ae634224dffa77837bb03c"
}
},
"widgets": {
"application/vnd.jupyter.widget-state+json": {
"014e097bb2c540f6836c61175a9bcba5": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HTMLModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HTMLModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HTMLView",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_b844d16570ce447598ddcf3ec084608a",
"placeholder": "",
"style": "IPY_MODEL_dd2d6dfae7f34fa2af3a4f89bd7660bb",
"value": " 5148/5148 [00:00<00:00, 134903.67it/s]"
}
},
"0e91dc133cd440f9973acd52f341b532": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HBoxModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HBoxModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HBoxView",
"box_style": "",
"children": [
"IPY_MODEL_de8f202890da499fa532e39882aa39ad",
"IPY_MODEL_4cf4c23aaea549daa6ec02712c5229f7",
"IPY_MODEL_42dbf79dd8804855a48940bd61adfd8c"
],
"layout": "IPY_MODEL_9d928a84ddb2401eb8ac38b1905167b9"
}
},
"1dec5c156112404e84fd30a8f2bb6dda": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"20e53fd4a49d4188a83fb152fac32a20": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"2441c125a1b645b6bb7124fdc667a6ac": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "DescriptionStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "DescriptionStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"description_width": ""
}
},
"2afdb5134ffd48b5a3ea36fdf13f3ca9": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"401ac5195b714b7bae27caddb02d9dae": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"418492022d584305b781ef243b35213a": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"42dbf79dd8804855a48940bd61adfd8c": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HTMLModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HTMLModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HTMLView",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_1dec5c156112404e84fd30a8f2bb6dda",
"placeholder": "",
"style": "IPY_MODEL_2441c125a1b645b6bb7124fdc667a6ac",
"value": " 26421880/26421880 [00:02<00:00, 18779588.66it/s]"
}
},
"463fc2365d084c03bb57e60310ef7797": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HBoxModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HBoxModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HBoxView",
"box_style": "",
"children": [
"IPY_MODEL_5b5d9289df9b40a89e34ce048dfa0ece",
"IPY_MODEL_7bd374159df44db9a117e97583427122",
"IPY_MODEL_014e097bb2c540f6836c61175a9bcba5"
],
"layout": "IPY_MODEL_7cb25b7ea7894bf999948823015a42c1"
}
},
"46e69b009d8d419b854c789933bb31cc": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "FloatProgressModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "FloatProgressModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "ProgressView",
"bar_style": "success",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_ab6012bbe9a54f23aeee9fc8985b0cd7",
"max": 29515,
"min": 0,
"orientation": "horizontal",
"style": "IPY_MODEL_f2efbbf0de6f4245b5ed7b29d62aee7e",
"value": 29515
}
},
"49f847d7f2ae48d5a274bcba1b352527": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"4b8f0c2e31fe4218902f572cdc73e9d2": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "DescriptionStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "DescriptionStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"description_width": ""
}
},
"4cf4c23aaea549daa6ec02712c5229f7": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "FloatProgressModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "FloatProgressModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "ProgressView",
"bar_style": "success",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_20e53fd4a49d4188a83fb152fac32a20",
"max": 26421880,
"min": 0,
"orientation": "horizontal",
"style": "IPY_MODEL_5edd0c12e56a4179be3775303eeecb5d",
"value": 26421880
}
},
"5b5d9289df9b40a89e34ce048dfa0ece": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HTMLModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HTMLModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HTMLView",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_8b87c5c46b2a496e8fce50e8d94f0364",
"placeholder": "",
"style": "IPY_MODEL_4b8f0c2e31fe4218902f572cdc73e9d2",
"value": "100%"
}
},
"5d03cf1d574f4319a78fd194c1802506": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "DescriptionStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "DescriptionStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"description_width": ""
}
},
"5edd0c12e56a4179be3775303eeecb5d": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "ProgressStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "ProgressStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"bar_color": null,
"description_width": ""
}
},
"7bd374159df44db9a117e97583427122": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "FloatProgressModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "FloatProgressModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "ProgressView",
"bar_style": "success",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_8cdfa0916bf6417a91fb4b19c44d0021",
"max": 5148,
"min": 0,
"orientation": "horizontal",
"style": "IPY_MODEL_eeb96b697cfa48b5add99d3931574ecb",
"value": 5148
}
},
"7cb25b7ea7894bf999948823015a42c1": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"808f5d3718ba40ad80882d8afcc6b70f": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "DescriptionStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "DescriptionStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"description_width": ""
}
},
"82820ce83cfb4c38bdf36e249b4c7d90": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HBoxModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HBoxModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HBoxView",
"box_style": "",
"children": [
"IPY_MODEL_de6746f353614fa2b61e007c811dc6cf",
"IPY_MODEL_46e69b009d8d419b854c789933bb31cc",
"IPY_MODEL_c265c2e58744434a9896884b4ab8f555"
],
"layout": "IPY_MODEL_49f847d7f2ae48d5a274bcba1b352527"
}
},
"8b87c5c46b2a496e8fce50e8d94f0364": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"8cdfa0916bf6417a91fb4b19c44d0021": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"9d928a84ddb2401eb8ac38b1905167b9": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"aabc8a2be63b4dcfabe841fb2b22fd95": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "FloatProgressModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "FloatProgressModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "ProgressView",
"bar_style": "success",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_401ac5195b714b7bae27caddb02d9dae",
"max": 4422102,
"min": 0,
"orientation": "horizontal",
"style": "IPY_MODEL_edb287c8f5e04532ad5c3760041491e2",
"value": 4422102
}
},
"ab6012bbe9a54f23aeee9fc8985b0cd7": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"abdf432283ea48409849a131e3c84538": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "DescriptionStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "DescriptionStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"description_width": ""
}
},
"acfbdfba8a594540b7f809ad5a490bee": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HBoxModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HBoxModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HBoxView",
"box_style": "",
"children": [
"IPY_MODEL_dfeb1006d236400cbb3d0e304c8208af",
"IPY_MODEL_aabc8a2be63b4dcfabe841fb2b22fd95",
"IPY_MODEL_fb7da5344fdd4e8d9f49da31fa32169a"
],
"layout": "IPY_MODEL_dbb95703595642fbb2979a63e1391149"
}
},
"b844d16570ce447598ddcf3ec084608a": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"c265c2e58744434a9896884b4ab8f555": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HTMLModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HTMLModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HTMLView",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_2afdb5134ffd48b5a3ea36fdf13f3ca9",
"placeholder": "",
"style": "IPY_MODEL_5d03cf1d574f4319a78fd194c1802506",
"value": " 29515/29515 [00:00<00:00, 208316.66it/s]"
}
},
"d6ec914f5ca44a7f884d4b20e901ecf8": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "DescriptionStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "DescriptionStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"description_width": ""
}
},
"dbb95703595642fbb2979a63e1391149": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"dd2d6dfae7f34fa2af3a4f89bd7660bb": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "DescriptionStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "DescriptionStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"description_width": ""
}
},
"dd328252b2bf4ec994e13e8e6dba2875": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"de6746f353614fa2b61e007c811dc6cf": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HTMLModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HTMLModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HTMLView",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_f465c7c665874e06875e5564af85704e",
"placeholder": "",
"style": "IPY_MODEL_f58599fc1d6c4308a21aafb5b15a7f7a",
"value": "100%"
}
},
"de6d869bae854d598d06dc7d17bb16b6": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"de8f202890da499fa532e39882aa39ad": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HTMLModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HTMLModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HTMLView",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_dd328252b2bf4ec994e13e8e6dba2875",
"placeholder": "",
"style": "IPY_MODEL_808f5d3718ba40ad80882d8afcc6b70f",
"value": "100%"
}
},
"dfeb1006d236400cbb3d0e304c8208af": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HTMLModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HTMLModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HTMLView",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_de6d869bae854d598d06dc7d17bb16b6",
"placeholder": "",
"style": "IPY_MODEL_d6ec914f5ca44a7f884d4b20e901ecf8",
"value": "100%"
}
},
"edb287c8f5e04532ad5c3760041491e2": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "ProgressStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "ProgressStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"bar_color": null,
"description_width": ""
}
},
"eeb96b697cfa48b5add99d3931574ecb": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "ProgressStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "ProgressStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"bar_color": null,
"description_width": ""
}
},
"f2efbbf0de6f4245b5ed7b29d62aee7e": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "ProgressStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "ProgressStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"bar_color": null,
"description_width": ""
}
},
"f465c7c665874e06875e5564af85704e": {
"model_module": "@jupyter-widgets/base",
"model_module_version": "1.2.0",
"model_name": "LayoutModel",
"state": {
"_model_module": "@jupyter-widgets/base",
"_model_module_version": "1.2.0",
"_model_name": "LayoutModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "LayoutView",
"align_content": null,
"align_items": null,
"align_self": null,
"border": null,
"bottom": null,
"display": null,
"flex": null,
"flex_flow": null,
"grid_area": null,
"grid_auto_columns": null,
"grid_auto_flow": null,
"grid_auto_rows": null,
"grid_column": null,
"grid_gap": null,
"grid_row": null,
"grid_template_areas": null,
"grid_template_columns": null,
"grid_template_rows": null,
"height": null,
"justify_content": null,
"justify_items": null,
"left": null,
"margin": null,
"max_height": null,
"max_width": null,
"min_height": null,
"min_width": null,
"object_fit": null,
"object_position": null,
"order": null,
"overflow": null,
"overflow_x": null,
"overflow_y": null,
"padding": null,
"right": null,
"top": null,
"visibility": null,
"width": null
}
},
"f58599fc1d6c4308a21aafb5b15a7f7a": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "DescriptionStyleModel",
"state": {
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "DescriptionStyleModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/base",
"_view_module_version": "1.2.0",
"_view_name": "StyleView",
"description_width": ""
}
},
"fb7da5344fdd4e8d9f49da31fa32169a": {
"model_module": "@jupyter-widgets/controls",
"model_module_version": "1.5.0",
"model_name": "HTMLModel",
"state": {
"_dom_classes": [],
"_model_module": "@jupyter-widgets/controls",
"_model_module_version": "1.5.0",
"_model_name": "HTMLModel",
"_view_count": null,
"_view_module": "@jupyter-widgets/controls",
"_view_module_version": "1.5.0",
"_view_name": "HTMLView",
"description": "",
"description_tooltip": null,
"layout": "IPY_MODEL_418492022d584305b781ef243b35213a",
"placeholder": "",
"style": "IPY_MODEL_abdf432283ea48409849a131e3c84538",
"value": " 4422102/4422102 [00:01<00:00, 6591138.43it/s]"
}
}
}
}
},
"nbformat": 4,
"nbformat_minor": 0
}
|