notebooks/assessments/bench_denoising.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Denoise Validation\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n"
     ]
    }
   ],
   "source": [
    "#! lamin init --storage ./lamin-intro --schema bionty \n",
    "\n",
    "import scanpy as sc\n",
    "from scprint import scPrint\n",
    "from scdataloader import Preprocessor\n",
    "import numpy as np\n",
    "import anndata as ad\n",
    "\n",
    "from scprint.tasks import Denoiser, withknn\n",
    "from scdataloader.utils import downsample_profile\n",
    "from scipy.stats import spearmanr\n",
    "from anndata import AnnData\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import torch\n",
    "torch.set_float32_matmul_precision('medium')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "prostate_combined = sc.read_h5ad(\"../data/prostate_combined_o2uniqsx.h5ad\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RuntimeError caught: scPrint is not attached to a `Trainer`.\n"
     ]
    }
   ],
   "source": [
    "# model = scPrint.load_from_checkpoint('../../data/tensorboard/scprint_test/urxxwu28/checkpoints/epoch=0-step=20000.ckpt') #chocolate-surf-8\n",
    "# '../../data/main/gene_embeddings.parquet')\n",
    "model = scPrint.load_from_checkpoint(\n",
    "#    '../data/temp/o2uniqsx/epoch=18-step=133000.ckpt', precpt_gene_emb=None)\n",
    "    '../../data/temp/last.ckpt',\n",
    "    precpt_gene_emb=None,\n",
    "    transformer=\"normal\")\n",
    "\n",
    "# ../../data/temp/epoch=4-step=12500.ckpt latest model maestro (fortuitous laugther)\n",
    "# ../../data/logs/scprint_test/f4vao230/checkpoints/epoch=10-step=27500.ckpt looked good as a training run (vermlillion noodle)\n",
    "# d1792z9k/epoch=3-step=10000.ckpt\n",
    "# '../../data/temp/d1792z9k/epoch=3-step=10000.ckpt'\n",
    "# '../../data/temp/6q3dzi3x/epoch=3-step=10000.ckpt'\n",
    "# '../../data/temp/grsachbg/epoch=2-step=15000.ckpt'\n",
    "# ../../data/temp/za9mg1w3/epoch=7-step=40000.ckpt\n",
    "# post debug\n",
    "# \"../lightning_logs/version_14/checkpoints/epoch=1-step=5000.ckpt\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#'/pasteur/appa/scratch/jkalfon/scprint/.lamindb/R4ZHoQegxXdSFNFY5LGe.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/SHV11AEetZOms4Wh7Ehb.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/V6DPJx8rP3wWRQ43LMHb.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/Gz5G2ETTEuuRDgwm7brA.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/YyBdEsN89p2aF4xJY1CW.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/SO5yBTUDBgkAmz0QbG8K.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/r4iCehg3Tw5IbCLiCIbl.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/SqvXr3i3PGXM8toXzUf9.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/REIyQZE6OMZm1S3W2Dxi.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/rYZ7gs0E0cqPOLONC8ia.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/FcwMDDbAQPNYIjcYNxoc.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/fvU5BAMJrm7vrgDmZM0z.h5ad', '/pasteur/appa/scratch/jkalfon/scprint/.lamindb/gNNpgpo6gATjuxTE7CCp.h5ad'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 12590 × 70116\n",
       "    obs: 'hash_label', 'hash_cluster', 'n_genes', 'hashed', 'group', 'treatment', 'n_genes_by_counts', 'total_counts', 'total_counts_mt', 'pct_counts_mt', 'pct_counts_hb', 'batch', 'S_score', 'G2M_score', 'phase', 'leiden', 'lineage', 'type', 'cds', 'donor_region', 'dpt_pseudotime', 'tissue_ontology_term_id', 'assay_ontology_term_id', 'disease_ontology_term_id', 'cell_type_ontology_term_id', 'sex_ontology_term_id', 'development_stage_ontology_term_id', 'is_primary_data', 'organism_ontology_term_id', 'donor_id', 'self_reported_ethnicity_ontology_term_id', 'suspension_type', 'BMI', 'cell_type', 'assay', 'disease', 'organism', 'sex', 'tissue', 'self_reported_ethnicity', 'development_stage', 'nnz', 'log1p_n_genes_by_counts', 'log1p_total_counts', 'pct_counts_in_top_20_genes', 'log1p_total_counts_mt', 'total_counts_ribo', 'log1p_total_counts_ribo', 'pct_counts_ribo', 'total_counts_hb', 'log1p_total_counts_hb', 'outlier', 'mt_outlier', 'cell_culture'\n",
       "    var: 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std', 's_phase', 'g2m_phase', 'feature_types', 'genome', 'mt', 'Donor_2-n_cells', 'Donor_1-n_cells', 'Donor_3-n_cells', 'Donor_2-n_cells_by_counts', 'Donor_1-n_cells_by_counts', 'Donor_3-n_cells_by_counts', 'Donor_2-mean_counts', 'Donor_1-mean_counts', 'Donor_3-mean_counts', 'Donor_2-pct_dropout_by_counts', 'Donor_1-pct_dropout_by_counts', 'Donor_3-pct_dropout_by_counts', 'Donor_2-total_counts', 'Donor_1-total_counts', 'Donor_3-total_counts', 'feature_is_filtered', 'feature_name', 'feature_reference', 'feature_biotype', 'uid', 'symbol', 'ncbi_gene_ids', 'biotype', 'description', 'synonyms', 'organism_id', 'public_source_id', 'created_by_id', 'ribo', 'hb', 'organism', 'n_cells_by_counts', 'mean_counts', 'log1p_mean_counts', 'pct_dropout_by_counts', 'total_counts', 'log1p_total_counts'\n",
       "    uns: 'unseen_genes'\n",
       "    obsm: 'X_pca', 'X_pca_harmony', 'X_umap'\n",
       "    layers: 'clean', 'true'\n",
       "    obsp: 'connectivities', 'distances'"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata = sc.read_h5ad('../../data/R4ZHoQegxXdSFNFY5LGe.h5ad')\n",
    "#adata = sc.read_h5ad('/home/ml4ig1/scprint/.lamindb/R4ZHoQegxXdSFNFY5LGe.h5ad')\n",
    "#adata = sc.read_h5ad('/home/ml4ig1/scprint/.lamindb/REIyQZE6OMZm1S3W2Dxi.h5ad')\n",
    "\n",
    "\n",
    "adata.layers['true'] = adata.X.copy()\n",
    "adata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sc.pl.embedding(adata, basis=\"X_umap\", color='cell_type')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "cell_type\n",
       "retinal rod cell          31858\n",
       "OFF-bipolar cell           3892\n",
       "Mueller cell               3668\n",
       "amacrine cell              2722\n",
       "ON-bipolar cell            2611\n",
       "rod bipolar cell           2319\n",
       "retinal cone cell          2211\n",
       "retina horizontal cell     1133\n",
       "retinal ganglion cell       493\n",
       "astrocyte                   232\n",
       "microglial cell             231\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata.obs.cell_type.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dropping layers:  KeysView(Layers with keys: clean, true)\n",
      "checking raw counts\n",
      "removed 0 non primary cells, 12590 renamining\n",
      "filtered out 0 cells, 12590 renamining\n",
      "Removed 175 genes.\n",
      "validating\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/jkobject/Documents/code/scPRINT/scDataLoader/scdataloader/preprocess.py:241: FutureWarning: Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`\n",
      "  data_utils.validate(adata, organism=adata.obs.organism_ontology_term_id[0])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "startin QC\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Exception ignored in: <function _MultiProcessingDataLoaderIter.__del__ at 0x1740f6b00>\n",
      "Traceback (most recent call last):\n",
      "  File \"/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/torch/utils/data/dataloader.py\", line 1479, in __del__\n",
      "    self._shutdown_workers()\n",
      "  File \"/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/torch/utils/data/dataloader.py\", line 1437, in _shutdown_workers\n",
      "    if self._persistent_workers or self._workers_status[worker_id]:\n",
      "AttributeError: '_MultiProcessingDataLoaderIter' object has no attribute '_workers_status'\n",
      "Exception ignored on calling ctypes callback function: <function ExecutionEngine._raw_object_cache_notify at 0x308d6aa70>\n",
      "Traceback (most recent call last):\n",
      "  File \"/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/llvmlite/binding/executionengine.py\", line 178, in _raw_object_cache_notify\n",
      "    def _raw_object_cache_notify(self, data):\n",
      "KeyboardInterrupt: \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Seeing 11526 outliers (91.55% of total dataset):\n",
      "done\n"
     ]
    }
   ],
   "source": [
    "adata.obs.drop(columns=\"is_primary_data\", inplace=True)\n",
    "preprocessor = Preprocessor(do_postp=False)\n",
    "adata = preprocessor(adata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "View of AnnData object with n_obs × n_vars = 12590 × 70611\n",
       "    obs: 'hash_label', 'hash_cluster', 'n_genes', 'hashed', 'group', 'treatment', 'n_genes_by_counts', 'total_counts', 'total_counts_mt', 'pct_counts_mt', 'pct_counts_hb', 'batch', 'S_score', 'G2M_score', 'phase', 'leiden', 'lineage', 'type', 'cds', 'donor_region', 'dpt_pseudotime', 'tissue_ontology_term_id', 'assay_ontology_term_id', 'disease_ontology_term_id', 'cell_type_ontology_term_id', 'sex_ontology_term_id', 'development_stage_ontology_term_id', 'organism_ontology_term_id', 'donor_id', 'self_reported_ethnicity_ontology_term_id', 'suspension_type', 'BMI', 'cell_type', 'assay', 'disease', 'organism', 'sex', 'tissue', 'self_reported_ethnicity', 'development_stage', 'nnz', 'log1p_n_genes_by_counts', 'log1p_total_counts', 'pct_counts_in_top_20_genes', 'log1p_total_counts_mt', 'total_counts_ribo', 'log1p_total_counts_ribo', 'pct_counts_ribo', 'total_counts_hb', 'log1p_total_counts_hb', 'outlier', 'mt_outlier', 'cell_culture'\n",
       "    var: 'highly_variable', 'means', 'dispersions', 'dispersions_norm', 'mean', 'std', 's_phase', 'g2m_phase', 'feature_types', 'genome', 'mt', 'Donor_2-n_cells', 'Donor_1-n_cells', 'Donor_3-n_cells', 'Donor_2-n_cells_by_counts', 'Donor_1-n_cells_by_counts', 'Donor_3-n_cells_by_counts', 'Donor_2-mean_counts', 'Donor_1-mean_counts', 'Donor_3-mean_counts', 'Donor_2-pct_dropout_by_counts', 'Donor_1-pct_dropout_by_counts', 'Donor_3-pct_dropout_by_counts', 'Donor_2-total_counts', 'Donor_1-total_counts', 'Donor_3-total_counts', 'feature_is_filtered', 'feature_name', 'feature_reference', 'feature_biotype', 'uid', 'symbol', 'ncbi_gene_ids', 'biotype', 'description', 'synonyms', 'organism_id', 'public_source_id', 'created_by_id', 'ribo', 'hb', 'organism', 'n_cells_by_counts', 'mean_counts', 'log1p_mean_counts', 'pct_dropout_by_counts', 'total_counts', 'log1p_total_counts', 'highly_variable_rank', 'variances', 'variances_norm', 'run_id'\n",
       "    uns: 'unseen_genes'\n",
       "    obsm: 'X_pca', 'X_pca_harmony', 'X_umap'\n",
       "    obsp: 'connectivities', 'distances'"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "adata"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## regular denoise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.attn_bias = \"none\"# = \"log\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No module named 'triton'\n",
      "FlashAttention is not installed, not using it..\n",
      "2000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/2000 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "💡 connected lamindb: jkobject/test-public-ontologies\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 2/2000 [01:56<32:13:23, 58.06s/it]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[27], line 13\u001b[0m\n\u001b[1;32m      1\u001b[0m denoise \u001b[38;5;241m=\u001b[39m Denoiser(\n\u001b[1;32m      2\u001b[0m     model,\n\u001b[1;32m      3\u001b[0m     batch_size\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m5\u001b[39m,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     11\u001b[0m     dtype\u001b[38;5;241m=\u001b[39mtorch\u001b[38;5;241m.\u001b[39mbfloat16,\n\u001b[1;32m     12\u001b[0m )\n\u001b[0;32m---> 13\u001b[0m metrics, idx, genes, expr \u001b[38;5;241m=\u001b[39m \u001b[43mdenoise\u001b[49m\u001b[43m(\u001b[49m\u001b[43madata\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     14\u001b[0m metrics\n",
      "File \u001b[0;32m~/Documents/code/scPRINT/scprint/tasks/denoise.py:127\u001b[0m, in \u001b[0;36mDenoiser.__call__\u001b[0;34m(self, adata)\u001b[0m\n\u001b[1;32m    121\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m batch \u001b[38;5;129;01min\u001b[39;00m tqdm(dataloader):\n\u001b[1;32m    122\u001b[0m         gene_pos, expression, depth \u001b[38;5;241m=\u001b[39m (\n\u001b[1;32m    123\u001b[0m             batch[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mgenes\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mto(device),\n\u001b[1;32m    124\u001b[0m             batch[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mx\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mto(device),\n\u001b[1;32m    125\u001b[0m             batch[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdepth\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mto(device),\n\u001b[1;32m    126\u001b[0m         )\n\u001b[0;32m--> 127\u001b[0m         \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_predict\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    128\u001b[0m \u001b[43m            \u001b[49m\u001b[43mgene_pos\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    129\u001b[0m \u001b[43m            \u001b[49m\u001b[43mexpression\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    130\u001b[0m \u001b[43m            \u001b[49m\u001b[43mdepth\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    131\u001b[0m \u001b[43m            \u001b[49m\u001b[43mpredict_mode\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mdenoise\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m    132\u001b[0m \u001b[43m            \u001b[49m\u001b[43mdepth_mult\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpredict_depth_mult\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    133\u001b[0m \u001b[43m        \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    134\u001b[0m torch\u001b[38;5;241m.\u001b[39mcuda\u001b[38;5;241m.\u001b[39mempty_cache()\n\u001b[1;32m    135\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mgenes \u001b[38;5;241m=\u001b[39m (\n\u001b[1;32m    136\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodel\u001b[38;5;241m.\u001b[39mpos\n\u001b[1;32m    137\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mhow \u001b[38;5;241m!=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmost var\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    138\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mlist\u001b[39m(\u001b[38;5;28mset\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodel\u001b[38;5;241m.\u001b[39mgenes) \u001b[38;5;241m&\u001b[39m \u001b[38;5;28mset\u001b[39m(genelist))\n\u001b[1;32m    139\u001b[0m )\n",
      "File \u001b[0;32m~/Documents/code/scPRINT/scprint/model/model.py:1216\u001b[0m, in \u001b[0;36mscPrint._predict\u001b[0;34m(self, gene_pos, expression, depth, predict_mode, pred_embedding, get_attention_layer, depth_mult, keep_output, max_size_in_mem)\u001b[0m\n\u001b[1;32m   1214\u001b[0m     cell_embs \u001b[38;5;241m=\u001b[39m output[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcell_embs\u001b[39m\u001b[38;5;124m\"\u001b[39m]\n\u001b[1;32m   1215\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m predict_mode \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdenoise\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[0;32m-> 1216\u001b[0m     output \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mforward\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1217\u001b[0m \u001b[43m        \u001b[49m\u001b[43mgene_pos\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1218\u001b[0m \u001b[43m        \u001b[49m\u001b[43mexpression\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1219\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdepth_mult\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mexpression\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msum\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mdepth_mult\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1220\u001b[0m \u001b[43m        \u001b[49m\u001b[43mfull_depth\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdepth\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mdepth_mult\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1221\u001b[0m \u001b[43m        \u001b[49m\u001b[43mget_attention_layer\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mget_attention_layer\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1222\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdo_class\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m   1223\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1224\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(get_attention_layer) \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m0\u001b[39m:\n\u001b[1;32m   1225\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mattn\u001b[38;5;241m.\u001b[39magg([i[:, :, :\u001b[38;5;241m2\u001b[39m, :] \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m output[\u001b[38;5;241m1\u001b[39m]], gene_pos)\n",
      "File \u001b[0;32m~/Documents/code/scPRINT/scprint/model/model.py:530\u001b[0m, in \u001b[0;36mscPrint.forward\u001b[0;34m(self, gene_pos, expression, mask, full_depth, timepoint, get_gene_emb, depth_mult, do_sample, do_mvc, do_class, get_attention_layer)\u001b[0m\n\u001b[1;32m    528\u001b[0m     bias[:, num:, :num] \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m10_000\u001b[39m  \u001b[38;5;66;03m# do not pay attention to the cls embeddings\u001b[39;00m\n\u001b[1;32m    529\u001b[0m     bias[:, num:, num:] \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mnbias[gene_pos[:, :, \u001b[38;5;28;01mNone\u001b[39;00m], gene_pos[:, \u001b[38;5;28;01mNone\u001b[39;00m, :]]\n\u001b[0;32m--> 530\u001b[0m transformer_output \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtransformer\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    531\u001b[0m \u001b[43m    \u001b[49m\u001b[43mencoding\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    532\u001b[0m \u001b[43m    \u001b[49m\u001b[43mreturn_qkv\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mget_attention_layer\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    533\u001b[0m \u001b[43m    \u001b[49m\u001b[43mbias\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbias\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mif\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mattn_bias\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m!=\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mnone\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01melse\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m    534\u001b[0m \u001b[43m    \u001b[49m\u001b[43mbias_layer\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mlist\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mrange\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnlayers\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m-\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    535\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    537\u001b[0m depth_mult \u001b[38;5;241m=\u001b[39m expression\u001b[38;5;241m.\u001b[39msum(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;28;01mif\u001b[39;00m depth_mult \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m depth_mult\n\u001b[1;32m    538\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(get_attention_layer) \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m0\u001b[39m:\n",
      "File \u001b[0;32m/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/Documents/code/scPRINT/scprint/model/flash_attn/flashformer.py:151\u001b[0m, in \u001b[0;36mFlashTransformerEncoder.forward\u001b[0;34m(self, hidden_states, mask, return_qkv, bias, bias_layer)\u001b[0m\n\u001b[1;32m    149\u001b[0m     bias \u001b[38;5;241m=\u001b[39m bias\u001b[38;5;241m.\u001b[39munsqueeze(\u001b[38;5;241m0\u001b[39m)\u001b[38;5;241m.\u001b[39munsqueeze(\u001b[38;5;241m0\u001b[39m)\n\u001b[1;32m    150\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m i, block \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mblocks):\n\u001b[0;32m--> 151\u001b[0m     hidden_states \u001b[38;5;241m=\u001b[39m \u001b[43mblock\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    152\u001b[0m \u001b[43m        \u001b[49m\u001b[43mhidden_states\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    153\u001b[0m \u001b[43m        \u001b[49m\u001b[43mresidual\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    154\u001b[0m \u001b[43m        \u001b[49m\u001b[43mreturn_qkv\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mi\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01min\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mreturn_qkv\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    155\u001b[0m \u001b[43m        \u001b[49m\u001b[43mbias\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbias\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mif\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mi\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01min\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mbias_layer\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01melse\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m    156\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    157\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m return_qkv:\n\u001b[1;32m    158\u001b[0m         qkvs\u001b[38;5;241m.\u001b[39mappend(hidden_states[\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m])\n",
      "File \u001b[0;32m/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/Documents/code/scPRINT/scprint/model/flash_attn/block.py:240\u001b[0m, in \u001b[0;36mBlock.forward\u001b[0;34m(self, hidden_states, residual, bias, src_mask, is_causal, src_key_padding_mask, mixer_subset, mixer_kwargs, return_qkv)\u001b[0m\n\u001b[1;32m    221\u001b[0m                 rowscale2 \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mdrop_path2(\n\u001b[1;32m    222\u001b[0m                     torch\u001b[38;5;241m.\u001b[39mones(\n\u001b[1;32m    223\u001b[0m                         hidden_states\u001b[38;5;241m.\u001b[39mshape[:\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m],\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    226\u001b[0m                     )\n\u001b[1;32m    227\u001b[0m                 )\n\u001b[1;32m    228\u001b[0m             hidden_states, residual \u001b[38;5;241m=\u001b[39m layer_norm_fn(\n\u001b[1;32m    229\u001b[0m                 hidden_states,\n\u001b[1;32m    230\u001b[0m                 \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mnorm2\u001b[38;5;241m.\u001b[39mweight,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    238\u001b[0m                 is_rms_norm\u001b[38;5;241m=\u001b[39m\u001b[38;5;28misinstance\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mnorm2, RMSNorm),\n\u001b[1;32m    239\u001b[0m             )\n\u001b[0;32m--> 240\u001b[0m         hidden_states \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmlp\u001b[49m\u001b[43m(\u001b[49m\u001b[43mhidden_states\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    241\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m (\n\u001b[1;32m    242\u001b[0m         (hidden_states, residual)\n\u001b[1;32m    243\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m return_qkv\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    248\u001b[0m         )\n\u001b[1;32m    249\u001b[0m     )\n\u001b[1;32m    250\u001b[0m \u001b[38;5;66;03m# if not prenorm (disregard for scPrint)\u001b[39;00m\n\u001b[1;32m    251\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n",
      "File \u001b[0;32m/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m~/Documents/code/scPRINT/scprint/model/flash_attn/mlp.py:40\u001b[0m, in \u001b[0;36mMlp.forward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m     38\u001b[0m y \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mfc1(x)\n\u001b[1;32m     39\u001b[0m y \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mactivation(y)\n\u001b[0;32m---> 40\u001b[0m y \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfc2\u001b[49m\u001b[43m(\u001b[49m\u001b[43my\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     41\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m y \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mreturn_residual \u001b[38;5;28;01melse\u001b[39;00m (y, x)\n",
      "File \u001b[0;32m/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/torch/nn/modules/module.py:1501\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1496\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[1;32m   1497\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[1;32m   1498\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[1;32m   1499\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[1;32m   1500\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[0;32m-> 1501\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mforward_call\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1502\u001b[0m \u001b[38;5;66;03m# Do not call functions when jit is used\u001b[39;00m\n\u001b[1;32m   1503\u001b[0m full_backward_hooks, non_full_backward_hooks \u001b[38;5;241m=\u001b[39m [], []\n",
      "File \u001b[0;32m/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/torch/nn/modules/linear.py:114\u001b[0m, in \u001b[0;36mLinear.forward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    113\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mforward\u001b[39m(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;28minput\u001b[39m: Tensor) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Tensor:\n\u001b[0;32m--> 114\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mF\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlinear\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43minput\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mweight\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbias\u001b[49m\u001b[43m)\u001b[49m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "denoise = Denoiser(\n",
    "    model,\n",
    "    batch_size=5,\n",
    "    max_len=2000,\n",
    "    plot_corr_size=10_000,\n",
    "    doplot=False,\n",
    "    num_workers=1,\n",
    "    predict_depth_mult=10,\n",
    "    downsample = 0.7,\n",
    "    how=\"most var\",\n",
    "    dtype=torch.bfloat16,\n",
    ")\n",
    "metrics, idx, genes, expr = denoise(adata)\n",
    "metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<QuerySet []>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bt.CellLine.filter()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scdataloader.utils import populate_my_ontology"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "❗ records with similar names exist! did you mean to load one of them?\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>uid</th>\n",
       "      <th>name</th>\n",
       "      <th>ontology_id</th>\n",
       "      <th>abbr</th>\n",
       "      <th>synonyms</th>\n",
       "      <th>description</th>\n",
       "      <th>public_source_id</th>\n",
       "      <th>run_id</th>\n",
       "      <th>created_by_id</th>\n",
       "      <th>updated_at</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>839</th>\n",
       "      <td>7CKxLDlu</td>\n",
       "      <td>obsolete congenital abnormality</td>\n",
       "      <td>MONDO:0000839</td>\n",
       "      <td>None</td>\n",
       "      <td>congenital anomaly or birth defect|congenital ...</td>\n",
       "      <td>Obsolete. Any Abnormality, Anatomical Or Bioch...</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:15.779843+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>881</th>\n",
       "      <td>5pHdg2Ic</td>\n",
       "      <td>obsolete myeloid and lymphoid neoplasms with e...</td>\n",
       "      <td>MONDO:0000881</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:15.781607+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>884</th>\n",
       "      <td>5lzQM8ub</td>\n",
       "      <td>myeloid and lymphoid neoplasms associated with...</td>\n",
       "      <td>MONDO:0000884</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:15.781729+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1303</th>\n",
       "      <td>2f3luB0s</td>\n",
       "      <td>abnormal pupillary function</td>\n",
       "      <td>MONDO:0001303</td>\n",
       "      <td>None</td>\n",
       "      <td>abnormal pupillary function (disease)|abnormal...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:15.804950+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1396</th>\n",
       "      <td>2slEWzUl</td>\n",
       "      <td>abnormal threshold of rods</td>\n",
       "      <td>MONDO:0001396</td>\n",
       "      <td>None</td>\n",
       "      <td>abnormal dark adaptation curve</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:15.810059+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1565</th>\n",
       "      <td>6ZqhPFj2</td>\n",
       "      <td>abnormal retinal correspondence</td>\n",
       "      <td>MONDO:0001565</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:15.819207+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1770</th>\n",
       "      <td>3DvVEahq</td>\n",
       "      <td>gastrin secretion abnormality</td>\n",
       "      <td>MONDO:0001770</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:15.829891+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1962</th>\n",
       "      <td>6FITWYCe</td>\n",
       "      <td>abnormality of glucagon secretion</td>\n",
       "      <td>MONDO:0001962</td>\n",
       "      <td>None</td>\n",
       "      <td>glucagon secretion abnormality</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:15.840150+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6314</th>\n",
       "      <td>6NFTwVkD</td>\n",
       "      <td>normal breast-like subtype of breast carcinoma</td>\n",
       "      <td>MONDO:0006324</td>\n",
       "      <td>None</td>\n",
       "      <td>normal breast-like subtype of breast cancer|no...</td>\n",
       "      <td>A Biologic Subset Of Breast Carcinoma Defined ...</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.083450+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7106</th>\n",
       "      <td>5nAdOPol</td>\n",
       "      <td>isolated anhidrosis with normal sweat glands</td>\n",
       "      <td>MONDO:0007118</td>\n",
       "      <td>None</td>\n",
       "      <td>Dann-Epstein-Sohar syndrome|anhidrosis caused ...</td>\n",
       "      <td>Any Anhidrosis In Which The Cause Of The Disea...</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.127705+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7168</th>\n",
       "      <td>3Oh1TC3K</td>\n",
       "      <td>Axenfeld-Rieger anomaly with partially absent ...</td>\n",
       "      <td>MONDO:0007180</td>\n",
       "      <td>None</td>\n",
       "      <td>Axenfeld-Rieger anomaly with partially absent ...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.131743+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7234</th>\n",
       "      <td>boyQgFKV</td>\n",
       "      <td>calcific aortic disease with immunologic abnor...</td>\n",
       "      <td>MONDO:0007246</td>\n",
       "      <td>None</td>\n",
       "      <td>calcific aortic disease with immunologic abnor...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.135803+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7468</th>\n",
       "      <td>2ufJZwGL</td>\n",
       "      <td>dwarfism with stiff joints and ocular abnormal...</td>\n",
       "      <td>MONDO:0007480</td>\n",
       "      <td>None</td>\n",
       "      <td>dwarfism with stiff joints and ocular abnormal...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.147956+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7969</th>\n",
       "      <td>1Ckl83dO</td>\n",
       "      <td>obsolete metachromatic leukodystrophy, adult-o...</td>\n",
       "      <td>MONDO:0007981</td>\n",
       "      <td>None</td>\n",
       "      <td>metachromatic leukodystrophy, adult-onset, wit...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.176289+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8079</th>\n",
       "      <td>5X1tbYMG</td>\n",
       "      <td>obsolete abnormal neutrophil chemotactic response</td>\n",
       "      <td>MONDO:0008091</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.182012+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8091</th>\n",
       "      <td>TNWQnFgm</td>\n",
       "      <td>noduli Cutanei, multiple, with urinary tract a...</td>\n",
       "      <td>MONDO:0008103</td>\n",
       "      <td>None</td>\n",
       "      <td>noduli Cutanei, multiple, with urinary tract a...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.182514+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8123</th>\n",
       "      <td>4yOtJm1j</td>\n",
       "      <td>optic atrophy 13 with retinal and foveal abnor...</td>\n",
       "      <td>MONDO:0008135</td>\n",
       "      <td>None</td>\n",
       "      <td>optic atrophy 13 with retinal and foveal abnor...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.185195+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8337</th>\n",
       "      <td>3zhNWHjz</td>\n",
       "      <td>pulmonic stenosis, atrial septal defect, and u...</td>\n",
       "      <td>MONDO:0008349</td>\n",
       "      <td>None</td>\n",
       "      <td>pulmonic stenosis, atrial septal defect, and u...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.196442+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8777</th>\n",
       "      <td>4cwPdoaU</td>\n",
       "      <td>anemia, nonspherocytic hemolytic, associated w...</td>\n",
       "      <td>MONDO:0008789</td>\n",
       "      <td>None</td>\n",
       "      <td>anemia, nonspherocytic hemolytic, associated w...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.221628+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8782</th>\n",
       "      <td>7WIoz0mt</td>\n",
       "      <td>anhidrosis, familial generalized, with abnorma...</td>\n",
       "      <td>MONDO:0008794</td>\n",
       "      <td>None</td>\n",
       "      <td>anhidrosis, familial generalized, with abnorma...</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.221838+00:00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           uid                                               name  \\\n",
       "id                                                                  \n",
       "839   7CKxLDlu                    obsolete congenital abnormality   \n",
       "881   5pHdg2Ic  obsolete myeloid and lymphoid neoplasms with e...   \n",
       "884   5lzQM8ub  myeloid and lymphoid neoplasms associated with...   \n",
       "1303  2f3luB0s                        abnormal pupillary function   \n",
       "1396  2slEWzUl                         abnormal threshold of rods   \n",
       "1565  6ZqhPFj2                    abnormal retinal correspondence   \n",
       "1770  3DvVEahq                      gastrin secretion abnormality   \n",
       "1962  6FITWYCe                  abnormality of glucagon secretion   \n",
       "6314  6NFTwVkD     normal breast-like subtype of breast carcinoma   \n",
       "7106  5nAdOPol       isolated anhidrosis with normal sweat glands   \n",
       "7168  3Oh1TC3K  Axenfeld-Rieger anomaly with partially absent ...   \n",
       "7234  boyQgFKV  calcific aortic disease with immunologic abnor...   \n",
       "7468  2ufJZwGL  dwarfism with stiff joints and ocular abnormal...   \n",
       "7969  1Ckl83dO  obsolete metachromatic leukodystrophy, adult-o...   \n",
       "8079  5X1tbYMG  obsolete abnormal neutrophil chemotactic response   \n",
       "8091  TNWQnFgm  noduli Cutanei, multiple, with urinary tract a...   \n",
       "8123  4yOtJm1j  optic atrophy 13 with retinal and foveal abnor...   \n",
       "8337  3zhNWHjz  pulmonic stenosis, atrial septal defect, and u...   \n",
       "8777  4cwPdoaU  anemia, nonspherocytic hemolytic, associated w...   \n",
       "8782  7WIoz0mt  anhidrosis, familial generalized, with abnorma...   \n",
       "\n",
       "        ontology_id  abbr                                           synonyms  \\\n",
       "id                                                                             \n",
       "839   MONDO:0000839  None  congenital anomaly or birth defect|congenital ...   \n",
       "881   MONDO:0000881  None                                               None   \n",
       "884   MONDO:0000884  None                                               None   \n",
       "1303  MONDO:0001303  None  abnormal pupillary function (disease)|abnormal...   \n",
       "1396  MONDO:0001396  None                     abnormal dark adaptation curve   \n",
       "1565  MONDO:0001565  None                                               None   \n",
       "1770  MONDO:0001770  None                                               None   \n",
       "1962  MONDO:0001962  None                     glucagon secretion abnormality   \n",
       "6314  MONDO:0006324  None  normal breast-like subtype of breast cancer|no...   \n",
       "7106  MONDO:0007118  None  Dann-Epstein-Sohar syndrome|anhidrosis caused ...   \n",
       "7168  MONDO:0007180  None  Axenfeld-Rieger anomaly with partially absent ...   \n",
       "7234  MONDO:0007246  None  calcific aortic disease with immunologic abnor...   \n",
       "7468  MONDO:0007480  None  dwarfism with stiff joints and ocular abnormal...   \n",
       "7969  MONDO:0007981  None  metachromatic leukodystrophy, adult-onset, wit...   \n",
       "8079  MONDO:0008091  None                                               None   \n",
       "8091  MONDO:0008103  None  noduli Cutanei, multiple, with urinary tract a...   \n",
       "8123  MONDO:0008135  None  optic atrophy 13 with retinal and foveal abnor...   \n",
       "8337  MONDO:0008349  None  pulmonic stenosis, atrial septal defect, and u...   \n",
       "8777  MONDO:0008789  None  anemia, nonspherocytic hemolytic, associated w...   \n",
       "8782  MONDO:0008794  None  anhidrosis, familial generalized, with abnorma...   \n",
       "\n",
       "                                            description  public_source_id  \\\n",
       "id                                                                          \n",
       "839   Obsolete. Any Abnormality, Anatomical Or Bioch...                39   \n",
       "881                                                None                39   \n",
       "884                                                None                39   \n",
       "1303                                               None                39   \n",
       "1396                                               None                39   \n",
       "1565                                               None                39   \n",
       "1770                                               None                39   \n",
       "1962                                               None                39   \n",
       "6314  A Biologic Subset Of Breast Carcinoma Defined ...                39   \n",
       "7106  Any Anhidrosis In Which The Cause Of The Disea...                39   \n",
       "7168                                               None                39   \n",
       "7234                                               None                39   \n",
       "7468                                               None                39   \n",
       "7969                                               None                39   \n",
       "8079                                               None                39   \n",
       "8091                                               None                39   \n",
       "8123                                               None                39   \n",
       "8337                                               None                39   \n",
       "8777                                               None                39   \n",
       "8782                                               None                39   \n",
       "\n",
       "     run_id  created_by_id                       updated_at  \n",
       "id                                                           \n",
       "839    None              1 2024-07-26 17:51:15.779843+00:00  \n",
       "881    None              1 2024-07-26 17:51:15.781607+00:00  \n",
       "884    None              1 2024-07-26 17:51:15.781729+00:00  \n",
       "1303   None              1 2024-07-26 17:51:15.804950+00:00  \n",
       "1396   None              1 2024-07-26 17:51:15.810059+00:00  \n",
       "1565   None              1 2024-07-26 17:51:15.819207+00:00  \n",
       "1770   None              1 2024-07-26 17:51:15.829891+00:00  \n",
       "1962   None              1 2024-07-26 17:51:15.840150+00:00  \n",
       "6314   None              1 2024-07-26 17:51:16.083450+00:00  \n",
       "7106   None              1 2024-07-26 17:51:16.127705+00:00  \n",
       "7168   None              1 2024-07-26 17:51:16.131743+00:00  \n",
       "7234   None              1 2024-07-26 17:51:16.135803+00:00  \n",
       "7468   None              1 2024-07-26 17:51:16.147956+00:00  \n",
       "7969   None              1 2024-07-26 17:51:16.176289+00:00  \n",
       "8079   None              1 2024-07-26 17:51:16.182012+00:00  \n",
       "8091   None              1 2024-07-26 17:51:16.182514+00:00  \n",
       "8123   None              1 2024-07-26 17:51:16.185195+00:00  \n",
       "8337   None              1 2024-07-26 17:51:16.196442+00:00  \n",
       "8777   None              1 2024-07-26 17:51:16.221628+00:00  \n",
       "8782   None              1 2024-07-26 17:51:16.221838+00:00  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "❗ records with similar names exist! did you mean to load one of them?\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>uid</th>\n",
       "      <th>name</th>\n",
       "      <th>ontology_id</th>\n",
       "      <th>abbr</th>\n",
       "      <th>synonyms</th>\n",
       "      <th>description</th>\n",
       "      <th>public_source_id</th>\n",
       "      <th>run_id</th>\n",
       "      <th>created_by_id</th>\n",
       "      <th>updated_at</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7705</th>\n",
       "      <td>1wpjcCa5</td>\n",
       "      <td>hemoglobin--variants for which the chain carry...</td>\n",
       "      <td>MONDO:0007717</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.161605+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16575</th>\n",
       "      <td>2C05jMg9</td>\n",
       "      <td>sporadic adult-onset ataxia of unknown etiology</td>\n",
       "      <td>MONDO:0016591</td>\n",
       "      <td>None</td>\n",
       "      <td>idiopathic late-onset cerebellar ataxia|SAOA</td>\n",
       "      <td>Sporadic Adult-Onset Ataxia Of Unknown Etiolog...</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.850901+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16688</th>\n",
       "      <td>4sqESY3D</td>\n",
       "      <td>obsolete glial tumor of neuroepithelial tissue...</td>\n",
       "      <td>MONDO:0016704</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:16.857050+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19372</th>\n",
       "      <td>5KqE5Q0L</td>\n",
       "      <td>unknown leukodystrophy</td>\n",
       "      <td>MONDO:0019397</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:17.020791+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23882</th>\n",
       "      <td>4XOYQjIL</td>\n",
       "      <td>chronic intervillositis of unknown etiology</td>\n",
       "      <td>MONDO:0035882</td>\n",
       "      <td>None</td>\n",
       "      <td>CIUE</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:17.292482+00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26179</th>\n",
       "      <td>6GRrR6D6</td>\n",
       "      <td>cancer of unknown primary site</td>\n",
       "      <td>MONDO:0858997</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>39</td>\n",
       "      <td>None</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-07-26 17:51:17.435467+00:00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            uid                                               name  \\\n",
       "id                                                                   \n",
       "7705   1wpjcCa5  hemoglobin--variants for which the chain carry...   \n",
       "16575  2C05jMg9    sporadic adult-onset ataxia of unknown etiology   \n",
       "16688  4sqESY3D  obsolete glial tumor of neuroepithelial tissue...   \n",
       "19372  5KqE5Q0L                             unknown leukodystrophy   \n",
       "23882  4XOYQjIL        chronic intervillositis of unknown etiology   \n",
       "26179  6GRrR6D6                     cancer of unknown primary site   \n",
       "\n",
       "         ontology_id  abbr                                      synonyms  \\\n",
       "id                                                                         \n",
       "7705   MONDO:0007717  None                                          None   \n",
       "16575  MONDO:0016591  None  idiopathic late-onset cerebellar ataxia|SAOA   \n",
       "16688  MONDO:0016704  None                                          None   \n",
       "19372  MONDO:0019397  None                                          None   \n",
       "23882  MONDO:0035882  None                                          CIUE   \n",
       "26179  MONDO:0858997  None                                          None   \n",
       "\n",
       "                                             description  public_source_id  \\\n",
       "id                                                                           \n",
       "7705                                                None                39   \n",
       "16575  Sporadic Adult-Onset Ataxia Of Unknown Etiolog...                39   \n",
       "16688                                               None                39   \n",
       "19372                                               None                39   \n",
       "23882                                               None                39   \n",
       "26179                                               None                39   \n",
       "\n",
       "      run_id  created_by_id                       updated_at  \n",
       "id                                                            \n",
       "7705    None              1 2024-07-26 17:51:16.161605+00:00  \n",
       "16575   None              1 2024-07-26 17:51:16.850901+00:00  \n",
       "16688   None              1 2024-07-26 17:51:16.857050+00:00  \n",
       "19372   None              1 2024-07-26 17:51:17.020791+00:00  \n",
       "23882   None              1 2024-07-26 17:51:17.292482+00:00  \n",
       "26179   None              1 2024-07-26 17:51:17.435467+00:00  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n",
      "/opt/homebrew/Caskroom/miniconda/base/envs/ntest/lib/python3.10/site-packages/pandas/core/frame.py:717: DeprecationWarning: Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "populate_my_ontology()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "💡 connected lamindb: jkobject/test-public-ontologies\n"
     ]
    }
   ],
   "source": [
    "import bionty as bt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<QuerySet []>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bt.Organism.filter(ontology_id=\"NCBI:9606\")#.filter()#)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# aggregate with KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {},
   "outputs": [],
   "source": [
    "subadata = AnnData(X=expr.cpu().numpy(), obs=adata.obs.iloc[idx], var=adata.var.loc[genes])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performing kNN-smoothing v2.1 with k=10, d=10, and dither=0.030...\n",
      "Step 1/4: Smooth using k=2\n",
      "Step 1/4: Smooth using k=2\n",
      "\tPCA took 0.6 s.\n",
      "\tThe fraction of variance explained by the top 10 PCs is 57.9 %.\n",
      "\tCalculating pair-wise distance matrix took 0.9 s.\n",
      "\tCalculating the smoothed expression matrix took 6.4 s.\n",
      "Step 2/4: Smooth using k=4\n",
      "\tPCA took 0.6 s.\n",
      "\tThe fraction of variance explained by the top 10 PCs is 72.2 %.\n",
      "\tCalculating pair-wise distance matrix took 0.8 s.\n",
      "\tCalculating the smoothed expression matrix took 6.4 s.\n",
      "Step 3/4: Smooth using k=8\n",
      "\tPCA took 0.5 s.\n",
      "\tThe fraction of variance explained by the top 10 PCs is 81.9 %.\n",
      "\tCalculating pair-wise distance matrix took 0.9 s.\n",
      "\tCalculating the smoothed expression matrix took 6.7 s.\n",
      "Step 4/4: Smooth using k=10\n",
      "\tPCA took 0.7 s.\n",
      "\tThe fraction of variance explained by the top 10 PCs is 88.0 %.\n",
      "\tCalculating pair-wise distance matrix took 0.8 s.\n",
      "\tCalculating the smoothed expression matrix took 6.5 s.\n",
      "kNN-smoothing finished in 32.7 s.\n"
     ]
    }
   ],
   "source": [
    "nadata = withknn(subadata, seed=42, k=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'reco2noisy': 0.137829846030993,\n",
       " 'reco2full': 0.3709222773794356,\n",
       " 'noisy2full': 0.3458865169667319}"
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reco = nadata.layers['denoised']\n",
    "true = adata.X[idx][:, adata.var.index.isin(genes)].toarray()\n",
    "noisy = np.loadtxt(\"collator_output.txt\")\n",
    "corr_coef, p_value = spearmanr(\n",
    "    np.vstack(\n",
    "        [reco[true != 0], noisy[true != 0], true[true != 0]]).T\n",
    ")\n",
    "{\n",
    "    \"reco2noisy\": corr_coef[0, 1],\n",
    "    \"reco2full\": corr_coef[0, 2],\n",
    "    \"noisy2full\": corr_coef[1, 2],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# on random expr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 125/125 [00:45<00:00,  2.72it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'reco2noisy': 0.7151437850641974,\n",
       " 'reco2full': 0.3239858026438798,\n",
       " 'noisy2full': 0.1896732525309344}"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "denoise = Denoiser(\n",
    "    model,\n",
    "    batch_size=40,\n",
    "    max_len=2200,\n",
    "    plot_corr_size=5_000,\n",
    "    doplot=False,\n",
    "    predict_depth_mult=10,\n",
    "    downsample = 0.7,\n",
    "    how=\"random expr\"\n",
    ")\n",
    "metrics, idx, genes, expr = denoise(adata)\n",
    "metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## with MAGIC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "genes = torch.load('../../scPRINT/data/temp/vbd8bavn/epoch=17-step=90000.ckpt', map_location=torch.device('cpu'))['hyper_parameters']['genes']\n",
    "sc.pp.highly_variable_genes(\n",
    "    adata, flavor=\"seurat_v3\", n_top_genes=5000, span=0.99)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_indices = np.random.choice(\n",
    "    adata.shape[0], size=10_000, replace=False\n",
    ")\n",
    "subadata = adata[random_indices, adata.var.index.isin(genes) & adata.var.highly_variable].copy()\n",
    "subadata.X = downsample_profile(torch.Tensor(subadata.X.toarray()), 0.7)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ml4ig1/.local/lib/python3.10/site-packages/magic/magic.py:650: DeprecationWarning: Call to deprecated method task. (Use TaskLogger.log_task instead) -- Deprecated since version 1.1.0.\n",
      "  with _logger.task(\"MAGIC\"):\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Calculating MAGIC...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ml4ig1/.local/lib/python3.10/site-packages/magic/magic.py:381: DeprecationWarning: Call to deprecated method info. (Use TaskLogger.log_info instead) -- Deprecated since version 1.1.0.\n",
      "  _logger.info(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Running MAGIC on 10000 cells and 3215 genes.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ml4ig1/.local/lib/python3.10/site-packages/magic/magic.py:436: DeprecationWarning: Call to deprecated method task. (Use TaskLogger.log_task instead) -- Deprecated since version 1.1.0.\n",
      "  with _logger.task(\"graph and diffusion operator\"):\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Calculating graph and diffusion operator...\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Calculating PCA...\n",
      "    Calculated PCA in 3.88 seconds.\n",
      "    Calculating KNN search...\n",
      "    Calculated KNN search in 4.74 seconds.\n",
      "    Calculating affinities...\n",
      "    Calculated affinities in 4.80 seconds.\n",
      "  Calculated graph and diffusion operator in 13.47 seconds.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ml4ig1/.local/lib/python3.10/site-packages/magic/magic.py:748: DeprecationWarning: Call to deprecated method task. (Use TaskLogger.log_task instead) -- Deprecated since version 1.1.0.\n",
      "  with _logger.task(\"imputation\"):\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Calculating imputation...\n",
      "  Calculated imputation in 0.11 seconds.\n",
      "  Calculated imputation in 0.11 seconds.\n",
      "Calculated MAGIC in 14.12 seconds.\n"
     ]
    }
   ],
   "source": [
    "denoised_adata = sc.external.pp.magic(subadata.copy(), copy=True, n_jobs=10, solver='approximate', verbose=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'reco2noisy': 0.2332537420283647,\n",
       " 'reco2full': 0.5902201434290297,\n",
       " 'noisy2full': 0.36316565975829096}"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reco = denoised_adata.X\n",
    "true = subadata.layers['true'].toarray()\n",
    "noisy = subadata.X\n",
    "corr_coef, p_value = spearmanr(\n",
    "    np.vstack(\n",
    "        [reco[true != 0], noisy[true != 0], true[true != 0]]).T\n",
    ")\n",
    "{\n",
    "    \"reco2noisy\": corr_coef[0, 1],\n",
    "    \"reco2full\": corr_coef[0, 2],\n",
    "    \"noisy2full\": corr_coef[1, 2],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## with MAGIC rare cells"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AnnData object with n_obs × n_vars = 10000 × 3107\n",
       "    obs: 'biosample_id', 'donor_id', 'cell_type_ontology_term_id', 'organism_ontology_term_id', 'disease_ontology_term_id', 'tissue_ontology_term_id', 'assay_ontology_term_id', 'cell_type__custom', 'development_stage_ontology_term_id', 'sex_ontology_term_id', 'suspension_type', 'is_primary_data', 'age', 'self_reported_ethnicity_ontology_term_id', 'cell_type', 'assay', 'disease', 'organism', 'sex', 'tissue', 'self_reported_ethnicity', 'development_stage', 'cell_culture', 'nnz', 'n_genes_by_counts', 'log1p_n_genes_by_counts', 'total_counts', 'log1p_total_counts', 'pct_counts_in_top_20_genes', 'total_counts_mt', 'log1p_total_counts_mt', 'pct_counts_mt', 'total_counts_ribo', 'log1p_total_counts_ribo', 'pct_counts_ribo', 'total_counts_hb', 'log1p_total_counts_hb', 'pct_counts_hb', 'outlier', 'mt_outlier'\n",
       "    var: 'feature_is_filtered', 'feature_name', 'feature_reference', 'feature_biotype', 'uid', 'symbol', 'ncbi_gene_ids', 'biotype', 'description', 'synonyms', 'organism_id', 'public_source_id', 'created_by_id', 'mt', 'ribo', 'hb', 'organism', 'n_cells_by_counts', 'mean_counts', 'log1p_mean_counts', 'pct_dropout_by_counts', 'total_counts', 'log1p_total_counts', 'highly_variable', 'highly_variable_rank', 'means', 'variances', 'variances_norm'\n",
       "    uns: 'unseen_genes', 'hvg'\n",
       "    obsm: 'X_UMAP'\n",
       "    layers: 'true'"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ct = \"microglial cell\"\n",
    "subadata\n",
    "#microglial cell, microfold cell of epithelium of small intestine, mesangial cell, pericyte"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'reco2noisy': 0.16739547347403197,\n",
       " 'reco2full': 0.40762284286749056,\n",
       " 'noisy2full': 0.3610326755364074}"
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reco = denoised_adata[subadata.obs.cell_type==ct].X\n",
    "true = subadata[subadata.obs.cell_type==ct].layers['true'].toarray()\n",
    "noisy = subadata[subadata.obs.cell_type==ct].X\n",
    "corr_coef, p_value = spearmanr(\n",
    "    np.vstack(\n",
    "        [reco[true != 0], noisy[true != 0], true[true != 0]]).T\n",
    ")\n",
    "{\n",
    "    \"reco2noisy\": corr_coef[0, 1],\n",
    "    \"reco2full\": corr_coef[0, 2],\n",
    "    \"noisy2full\": corr_coef[1, 2],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## same, with KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Performing kNN-smoothing v2.1 with k=10, d=10, and dither=0.030...\n",
      "\n",
      "Step 1/4: Smooth using k=2\n",
      "\tPCA took 0.5 s.\n",
      "\tThe fraction of variance explained by the top 10 PCs is 25.8 %.\n",
      "\tCalculating pair-wise distance matrix took 1.6 s.\n",
      "\tCalculating the smoothed expression matrix took 6.2 s.\n",
      "Step 2/4: Smooth using k=4\n",
      "\tPCA took 0.4 s.\n",
      "\tThe fraction of variance explained by the top 10 PCs is 35.0 %.\n",
      "\tCalculating pair-wise distance matrix took 1.5 s.\n",
      "\tCalculating the smoothed expression matrix took 6.2 s.\n",
      "Step 3/4: Smooth using k=8\n",
      "\tPCA took 0.4 s.\n",
      "\tThe fraction of variance explained by the top 10 PCs is 46.9 %.\n",
      "\tCalculating pair-wise distance matrix took 1.5 s.\n",
      "\tCalculating the smoothed expression matrix took 6.3 s.\n",
      "Step 4/4: Smooth using k=10\n",
      "\tPCA took 0.4 s.\n",
      "\tThe fraction of variance explained by the top 10 PCs is 59.5 %.\n",
      "\tCalculating pair-wise distance matrix took 1.5 s.\n",
      "\tCalculating the smoothed expression matrix took 6.3 s.\n",
      "kNN-smoothing finished in 33.6 s.\n"
     ]
    }
   ],
   "source": [
    "nadata = withknn(subadata, seed=42, k=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'reco2noisy': 0.41386242619767566,\n",
       " 'reco2full': 0.5781715062761915,\n",
       " 'noisy2full': 0.3685645948380719}"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reco = nadata.layers['denoised']#.toarray()\n",
    "true = subadata.layers['true'].toarray()\n",
    "noisy = subadata.X\n",
    "corr_coef, p_value = spearmanr(\n",
    "    np.vstack(\n",
    "        [reco[true != 0], noisy[true != 0], true[true != 0]]).T\n",
    ")\n",
    "{\n",
    "    \"reco2noisy\": corr_coef[0, 1],\n",
    "    \"reco2full\": corr_coef[0, 2],\n",
    "    \"noisy2full\": corr_coef[1, 2],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## with KNN rare cells"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'reco2noisy': 0.44956156837292166,\n",
       " 'reco2full': 0.5479113692785387,\n",
       " 'noisy2full': 0.3724976804482466}"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reco = nadata.layers['denoised'][nadata.obs.cell_type==ct]\n",
    "true = subadata[subadata.obs.cell_type==ct].layers['true'].toarray()\n",
    "noisy = subadata[subadata.obs.cell_type==ct].X\n",
    "corr_coef, p_value = spearmanr(\n",
    "    np.vstack(\n",
    "        [reco[true != 0], noisy[true != 0], true[true != 0]]).T\n",
    ")\n",
    "{\n",
    "    \"reco2noisy\": corr_coef[0, 1],\n",
    "    \"reco2full\": corr_coef[0, 2],\n",
    "    \"noisy2full\": corr_coef[1, 2],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MAGIC Rare cell without support"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [],
   "source": [
    "subadata = adata[adata.obs.cell_type==ct, adata.var.index.isin(genes)]\n",
    "sc.pp.highly_variable_genes(\n",
    "    subadata, flavor=\"seurat_v3\", n_top_genes=5000, span=0.99)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [],
   "source": [
    "subadata.X = downsample_profile(torch.Tensor(subadata.X.toarray()), 0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ml4ig1/.local/lib/python3.10/site-packages/magic/magic.py:650: DeprecationWarning: Call to deprecated method task. (Use TaskLogger.log_task instead) -- Deprecated since version 1.1.0.\n",
      "  with _logger.task(\"MAGIC\"):\n",
      "/home/ml4ig1/.local/lib/python3.10/site-packages/magic/magic.py:381: DeprecationWarning: Call to deprecated method info. (Use TaskLogger.log_info instead) -- Deprecated since version 1.1.0.\n",
      "  _logger.info(\n",
      "/home/ml4ig1/.local/lib/python3.10/site-packages/magic/magic.py:436: DeprecationWarning: Call to deprecated method task. (Use TaskLogger.log_task instead) -- Deprecated since version 1.1.0.\n",
      "  with _logger.task(\"graph and diffusion operator\"):\n",
      "/home/ml4ig1/.local/lib/python3.10/site-packages/graphtools/base.py:165: RuntimeWarning: Cannot perform PCA to 100 dimensions on data with min(n_samples, n_features) = 30\n",
      "  warnings.warn(\n",
      "/home/ml4ig1/.local/lib/python3.10/site-packages/magic/magic.py:748: DeprecationWarning: Call to deprecated method task. (Use TaskLogger.log_task instead) -- Deprecated since version 1.1.0.\n",
      "  with _logger.task(\"imputation\"):\n"
     ]
    }
   ],
   "source": [
    "denoised_adata = sc.external.pp.magic(subadata[:,subadata.var.highly_variable][:30], knn=5, n_jobs=20, solver='approximate', copy=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'reco2noisy': 0.3116307487080801,\n",
       " 'reco2full': 0.4477594810317442,\n",
       " 'noisy2full': 0.3117786337560895}"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reco = denoised_adata.X\n",
    "true = subadata[subadata.obs.cell_type==ct, subadata.var.highly_variable].layers['true'][:30].toarray()\n",
    "noisy = subadata[subadata.obs.cell_type==ct, subadata.var.highly_variable][:30].X\n",
    "corr_coef, p_value = spearmanr(\n",
    "    np.vstack(\n",
    "        [reco[true != 0], noisy[true != 0], true[true != 0]]).T\n",
    ")\n",
    "{\n",
    "    \"reco2noisy\": corr_coef[0, 1],\n",
    "    \"reco2full\": corr_coef[0, 2],\n",
    "    \"noisy2full\": corr_coef[1, 2],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## scPRINT rare cells"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1/1 [00:00<00:00,  1.44it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'reco2noisy': 0.7122362370625847,\n",
       " 'reco2full': 0.3170816393649402,\n",
       " 'noisy2full': 0.20527492811994658}"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "denoise = Denoiser(\n",
    "    model,\n",
    "    batch_size=40,\n",
    "    max_len=5000,\n",
    "    plot_corr_size=30,\n",
    "    doplot=False,\n",
    "    predict_depth_mult=10,\n",
    "    downsample = 0.7,\n",
    "    how=\"most var\"\n",
    ")\n",
    "metrics, idx, genes, expr = denoise(adata[adata.obs.cell_type==ct])\n",
    "metrics"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "scprint",
   "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": "3.10.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}