diff --git a/docs/tutorials/mouse_biccn.ipynb b/docs/tutorials/mouse_biccn.ipynb index 786a41d3..986d5c8e 100644 --- a/docs/tutorials/mouse_biccn.ipynb +++ b/docs/tutorials/mouse_biccn.ipynb @@ -18,11 +18,11 @@ "name": "stderr", "output_type": "stream", "text": [ - "2024-06-25 17:33:01.205298: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n", - "2024-06-25 17:33:01.244866: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", + "2024-06-26 14:46:47.842156: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n", + "2024-06-26 14:46:47.881644: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", "To enable the following instructions: AVX2 AVX512F AVX512_VNNI AVX512_BF16 AVX512_FP16 AVX_VNNI AMX_TILE AMX_INT8 AMX_BF16 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", - "2024-06-25 17:33:02.721837: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n", - "2024-06-25 17:33:04.704132: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1928] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 78790 MB memory: -> device: 0, name: NVIDIA H100 80GB HBM3, pci bus id: 0000:55:00.0, compute capability: 9.0\n" + "2024-06-26 14:46:49.415214: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n", + "2024-06-26 14:46:51.642221: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1928] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 78790 MB memory: -> device: 0, name: NVIDIA H100 80GB HBM3, pci bus id: 0000:55:00.0, compute capability: 9.0\n" ] } ], @@ -30,13 +30,16 @@ "import sys\n", "sys.path.insert(0, '/home/VIB.LOCAL/niklas.kempynck/.conda/envs/crested/lib/python3.11/site-packages')\n", "sys.path.insert(0,'/data/projects/c04/cbd-saerts/nkemp/software/CREsted/src')\n", - "#sys.path.remove('/mnt/modules/easybuild/software/SciPy-bundle/2023.07-gfbf-2023a/lib/python3.11/site-packages')\n", - "import crested\n" + "sys.path.remove('/mnt/modules/easybuild/software/SciPy-bundle/2023.07-gfbf-2023a/lib/python3.11/site-packages')\n", + "import crested\n", + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "%matplotlib inline" ] }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 2, "metadata": {}, "outputs": [ { @@ -45,7 +48,7 @@ "" ] }, - "execution_count": 19, + "execution_count": 2, "metadata": {}, "output_type": "execute_result" } @@ -64,7 +67,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "2024-06-25T17:33:11.288334+0200 INFO Extracting values from 19 bigWig files...\n" + "2024-06-26T14:46:57.920142+0200 INFO Extracting values from 19 bigWig files...\n" ] }, { @@ -99,9 +102,9 @@ "name": "stdout", "output_type": "stream", "text": [ - "2024-06-25T17:33:38.278601+0200 INFO Filtering on top k Gini scores...\n", - "2024-06-25T17:33:44.150386+0200 INFO Added normalization weights to adata.obsm['weights']...\n", - "2024-06-25T17:33:52.507340+0200 INFO After specificity filtering, kept 86887 out of 546993 regions.\n" + "2024-06-26T14:47:26.453591+0200 INFO Filtering on top k Gini scores...\n", + "2024-06-26T14:47:32.314144+0200 INFO Added normalization weights to adata.obsm['weights']...\n", + "2024-06-26T14:47:40.624157+0200 INFO After specificity filtering, kept 86887 out of 546993 regions.\n" ] } ], @@ -119,7 +122,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "Mon Jun 24 17:36:30 2024 \n", + "Wed Jun 26 11:35:03 2024 \n", "+-----------------------------------------------------------------------------------------+\n", "| NVIDIA-SMI 555.42.02 Driver Version: 555.42.02 CUDA Version: 12.5 |\n", "|-----------------------------------------+------------------------+----------------------+\n", @@ -137,8 +140,7 @@ "| GPU GI CI PID Type Process name GPU Memory |\n", "| ID ID Usage |\n", "|=========================================================================================|\n", - "| 0 N/A N/A 539601 C ...ynck/.conda/envs/crested/bin/python 72556MiB |\n", - "| 0 N/A N/A 542471 C ...ynck/.conda/envs/crested/bin/python 6746MiB |\n", + "| 0 N/A N/A 559712 C ...ynck/.conda/envs/crested/bin/python 79310MiB |\n", "+-----------------------------------------------------------------------------------------+\n" ] } @@ -167,7 +169,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -308,7 +310,7 @@ "[86887 rows x 4 columns]" ] }, - "execution_count": 6, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" } @@ -330,7 +332,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -460,7 +462,7 @@ "[7705 rows x 4 columns]" ] }, - "execution_count": 7, + "execution_count": 5, "metadata": {}, "output_type": "execute_result" } @@ -501,14 +503,14 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "2024-06-25T17:33:52.563735+0200 WARNING Chromsizes file not provided when shifting. Will not check if shifted regions are within chromosomes\n" + "2024-06-26T14:47:40.728963+0200 WARNING Chromsizes file not provided when shifting. Will not check if shifted regions are within chromosomes\n" ] } ], @@ -526,27 +528,7 @@ }, { "cell_type": "code", - "execution_count": 37, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "AnnDataLoader(dataset=AnnDataset(anndata_shape=(19, 440993), n_samples=881986, num_outputs=19, split=train, in_memory=False), batch_size=256, shuffle=True, one_hot_encode=True, drop_remainder=True)" - ] - }, - "execution_count": 37, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "datamodule.train_dataloader" - ] - }, - { - "cell_type": "code", - "execution_count": 10, + "execution_count": 8, "metadata": {}, "outputs": [], "source": [ @@ -568,7 +550,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 9, "metadata": {}, "outputs": [ { @@ -577,7 +559,7 @@ "" ] }, - "execution_count": 11, + "execution_count": 9, "metadata": {}, "output_type": "execute_result" } @@ -603,14 +585,14 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "TaskConfig(optimizer=, loss=, metrics=[, , , , , , ])\n" + "TaskConfig(optimizer=, loss=, metrics=[, , , , , , , ])\n" ] } ], @@ -618,7 +600,7 @@ "# Load the default configuration for training a topic classication model\n", "from crested.tl import default_configs, TaskConfig\n", "\n", - "config = default_configs(\"peak_regression\")\n", + "config = default_configs(\"peak_regression\", num_classes=19)\n", "print(config)\n", "\n", "# If you want to change some small parameters to an existing config, you can do it like this\n", @@ -647,7 +629,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 11, "metadata": {}, "outputs": [], "source": [ @@ -658,7 +640,7 @@ " config=config,\n", " project_name=\"deeppeak_benchmarking\",\n", " logger=\"wandb\",\n", - " run_name='crested_norm_TL'\n", + " run_name='crested_norm_TL_spearman'\n", ")\n" ] }, @@ -671,7 +653,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "Fri Jun 21 11:05:50 2024 \n", + "Wed Jun 26 11:18:17 2024 \n", "+-----------------------------------------------------------------------------------------+\n", "| NVIDIA-SMI 555.42.02 Driver Version: 555.42.02 CUDA Version: 12.5 |\n", "|-----------------------------------------+------------------------+----------------------+\n", @@ -680,7 +662,7 @@ "| | | MIG M. |\n", "|=========================================+========================+======================|\n", "| 0 NVIDIA H100 80GB HBM3 On | 00000000:55:00.0 Off | 0 |\n", - "| N/A 48C P0 123W / 700W | 79323MiB / 81559MiB | 0% Default |\n", + "| N/A 42C P0 118W / 700W | 79323MiB / 81559MiB | 0% Default |\n", "| | | Disabled |\n", "+-----------------------------------------+------------------------+----------------------+\n", " \n", @@ -689,7 +671,7 @@ "| GPU GI CI PID Type Process name GPU Memory |\n", "| ID ID Usage |\n", "|=========================================================================================|\n", - "| 0 N/A N/A 516332 C ...ynck/.conda/envs/crested/bin/python 70340MiB |\n", + "| 0 N/A N/A 558471 C ...ynck/.conda/envs/crested/bin/python 79312MiB |\n", "+-----------------------------------------------------------------------------------------+\n" ] } @@ -700,9 +682,545 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 12, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Failed to detect the name of this notebook, you can set it manually with the WANDB_NOTEBOOK_NAME environment variable to enable code saving.\n", + "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mkemp\u001b[0m. Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n" + ] + }, + { + "data": { + "text/html": [ + "wandb version 0.17.3 is available! To upgrade, please run:\n", + " $ pip install wandb --upgrade" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "Tracking run with wandb version 0.17.0" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "Run data is saved locally in /data/projects/c04/cbd-saerts/nkemp/software/CREsted/docs/tutorials/wandb/run-20240626_113547-70xblei4" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "Syncing run crested_norm_TL_spearman to Weights & Biases (docs)
" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + " View project at https://wandb.ai/kemp/deeppeak_benchmarking" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + " View run at https://wandb.ai/kemp/deeppeak_benchmarking/runs/70xblei4" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
Model: \"functional_1\"\n",
+       "
\n" + ], + "text/plain": [ + "\u001b[1mModel: \"functional_1\"\u001b[0m\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n",
+       "┃ Layer (type)        ┃ Output Shape      ┃    Param # ┃ Connected to      ┃\n",
+       "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n",
+       "│ sequence            │ (None, 2114, 4)   │          0 │ -                 │\n",
+       "│ (InputLayer)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ conv1d (Conv1D)     │ (None, 2114, 512) │     10,240 │ sequence[0][0]    │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ batch_normalization │ (None, 2114, 512) │      2,048 │ conv1d[0][0]      │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ activation          │ (None, 2114, 512) │          0 │ batch_normalizat… │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ dropout (Dropout)   │ (None, 2114, 512) │          0 │ activation[0][0]  │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_1conv         │ (None, 2110, 512) │    786,432 │ dropout[0][0]     │\n",
+       "│ (Conv1D)            │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_1bn           │ (None, 2110, 512) │      2,048 │ bpnet_1conv[0][0] │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_1activation   │ (None, 2110, 512) │          0 │ bpnet_1bn[0][0]   │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_1crop         │ (None, 2110, 512) │          0 │ dropout[0][0]     │\n",
+       "│ (Cropping1D)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ add (Add)           │ (None, 2110, 512) │          0 │ bpnet_1activatio… │\n",
+       "│                     │                   │            │ bpnet_1crop[0][0] │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_1dropout      │ (None, 2110, 512) │          0 │ add[0][0]         │\n",
+       "│ (Dropout)           │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_2conv         │ (None, 2102, 512) │    786,432 │ bpnet_1dropout[0… │\n",
+       "│ (Conv1D)            │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_2bn           │ (None, 2102, 512) │      2,048 │ bpnet_2conv[0][0] │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_2activation   │ (None, 2102, 512) │          0 │ bpnet_2bn[0][0]   │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_2crop         │ (None, 2102, 512) │          0 │ bpnet_1dropout[0… │\n",
+       "│ (Cropping1D)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ add_1 (Add)         │ (None, 2102, 512) │          0 │ bpnet_2activatio… │\n",
+       "│                     │                   │            │ bpnet_2crop[0][0] │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_2dropout      │ (None, 2102, 512) │          0 │ add_1[0][0]       │\n",
+       "│ (Dropout)           │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_3conv         │ (None, 2086, 512) │    786,432 │ bpnet_2dropout[0… │\n",
+       "│ (Conv1D)            │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_3bn           │ (None, 2086, 512) │      2,048 │ bpnet_3conv[0][0] │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_3activation   │ (None, 2086, 512) │          0 │ bpnet_3bn[0][0]   │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_3crop         │ (None, 2086, 512) │          0 │ bpnet_2dropout[0… │\n",
+       "│ (Cropping1D)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ add_2 (Add)         │ (None, 2086, 512) │          0 │ bpnet_3activatio… │\n",
+       "│                     │                   │            │ bpnet_3crop[0][0] │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_3dropout      │ (None, 2086, 512) │          0 │ add_2[0][0]       │\n",
+       "│ (Dropout)           │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_4conv         │ (None, 2054, 512) │    786,432 │ bpnet_3dropout[0… │\n",
+       "│ (Conv1D)            │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_4bn           │ (None, 2054, 512) │      2,048 │ bpnet_4conv[0][0] │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_4activation   │ (None, 2054, 512) │          0 │ bpnet_4bn[0][0]   │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_4crop         │ (None, 2054, 512) │          0 │ bpnet_3dropout[0… │\n",
+       "│ (Cropping1D)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ add_3 (Add)         │ (None, 2054, 512) │          0 │ bpnet_4activatio… │\n",
+       "│                     │                   │            │ bpnet_4crop[0][0] │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_4dropout      │ (None, 2054, 512) │          0 │ add_3[0][0]       │\n",
+       "│ (Dropout)           │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_5conv         │ (None, 1990, 512) │    786,432 │ bpnet_4dropout[0… │\n",
+       "│ (Conv1D)            │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_5bn           │ (None, 1990, 512) │      2,048 │ bpnet_5conv[0][0] │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_5activation   │ (None, 1990, 512) │          0 │ bpnet_5bn[0][0]   │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_5crop         │ (None, 1990, 512) │          0 │ bpnet_4dropout[0… │\n",
+       "│ (Cropping1D)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ add_4 (Add)         │ (None, 1990, 512) │          0 │ bpnet_5activatio… │\n",
+       "│                     │                   │            │ bpnet_5crop[0][0] │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_5dropout      │ (None, 1990, 512) │          0 │ add_4[0][0]       │\n",
+       "│ (Dropout)           │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_6conv         │ (None, 1862, 512) │    786,432 │ bpnet_5dropout[0… │\n",
+       "│ (Conv1D)            │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_6bn           │ (None, 1862, 512) │      2,048 │ bpnet_6conv[0][0] │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_6activation   │ (None, 1862, 512) │          0 │ bpnet_6bn[0][0]   │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_6crop         │ (None, 1862, 512) │          0 │ bpnet_5dropout[0… │\n",
+       "│ (Cropping1D)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ add_5 (Add)         │ (None, 1862, 512) │          0 │ bpnet_6activatio… │\n",
+       "│                     │                   │            │ bpnet_6crop[0][0] │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_6dropout      │ (None, 1862, 512) │          0 │ add_5[0][0]       │\n",
+       "│ (Dropout)           │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_7conv         │ (None, 1606, 512) │    786,432 │ bpnet_6dropout[0… │\n",
+       "│ (Conv1D)            │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_7bn           │ (None, 1606, 512) │      2,048 │ bpnet_7conv[0][0] │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_7activation   │ (None, 1606, 512) │          0 │ bpnet_7bn[0][0]   │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_7crop         │ (None, 1606, 512) │          0 │ bpnet_6dropout[0… │\n",
+       "│ (Cropping1D)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ add_6 (Add)         │ (None, 1606, 512) │          0 │ bpnet_7activatio… │\n",
+       "│                     │                   │            │ bpnet_7crop[0][0] │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_7dropout      │ (None, 1606, 512) │          0 │ add_6[0][0]       │\n",
+       "│ (Dropout)           │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_8conv         │ (None, 1094, 512) │    786,432 │ bpnet_7dropout[0… │\n",
+       "│ (Conv1D)            │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_8bn           │ (None, 1094, 512) │      2,048 │ bpnet_8conv[0][0] │\n",
+       "│ (BatchNormalizatio… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_8activation   │ (None, 1094, 512) │          0 │ bpnet_8bn[0][0]   │\n",
+       "│ (Activation)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_8crop         │ (None, 1094, 512) │          0 │ bpnet_7dropout[0… │\n",
+       "│ (Cropping1D)        │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ add_7 (Add)         │ (None, 1094, 512) │          0 │ bpnet_8activatio… │\n",
+       "│                     │                   │            │ bpnet_8crop[0][0] │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ bpnet_8dropout      │ (None, 1094, 512) │          0 │ add_7[0][0]       │\n",
+       "│ (Dropout)           │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ global_average_poo… │ (None, 512)       │          0 │ bpnet_8dropout[0… │\n",
+       "│ (GlobalAveragePool… │                   │            │                   │\n",
+       "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n",
+       "│ dense (Dense)       │ (None, 19)        │      9,747 │ global_average_p… │\n",
+       "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n",
+       "
\n" + ], + "text/plain": [ + "┏━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━┓\n", + "┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mConnected to \u001b[0m\u001b[1m \u001b[0m┃\n", + "┡━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━┩\n", + "│ sequence │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2114\u001b[0m, \u001b[38;5;34m4\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ - │\n", + "│ (\u001b[38;5;33mInputLayer\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ conv1d (\u001b[38;5;33mConv1D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2114\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m10,240\u001b[0m │ sequence[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ batch_normalization │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2114\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ conv1d[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2114\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ batch_normalizat… │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2114\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ activation[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_1conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2110\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ dropout[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mConv1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_1bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2110\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ bpnet_1conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_1activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2110\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_1bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_1crop │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2110\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ dropout[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mCropping1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ add (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2110\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_1activatio… │\n", + "│ │ │ │ bpnet_1crop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_1dropout │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2110\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_2conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2102\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ bpnet_1dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mConv1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_2bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2102\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ bpnet_2conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_2activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2102\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_2bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_2crop │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2102\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_1dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mCropping1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ add_1 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2102\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_2activatio… │\n", + "│ │ │ │ bpnet_2crop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_2dropout │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2102\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_1[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_3conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2086\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ bpnet_2dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mConv1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_3bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2086\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ bpnet_3conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_3activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2086\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_3bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_3crop │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2086\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_2dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mCropping1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ add_2 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2086\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_3activatio… │\n", + "│ │ │ │ bpnet_3crop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_3dropout │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2086\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_2[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_4conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2054\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ bpnet_3dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mConv1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_4bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2054\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ bpnet_4conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_4activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2054\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_4bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_4crop │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2054\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_3dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mCropping1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ add_3 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2054\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_4activatio… │\n", + "│ │ │ │ bpnet_4crop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_4dropout │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2054\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_3[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_5conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1990\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ bpnet_4dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mConv1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_5bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1990\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ bpnet_5conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_5activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1990\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_5bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_5crop │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1990\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_4dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mCropping1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ add_4 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1990\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_5activatio… │\n", + "│ │ │ │ bpnet_5crop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_5dropout │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1990\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_4[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_6conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1862\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ bpnet_5dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mConv1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_6bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1862\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ bpnet_6conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_6activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1862\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_6bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_6crop │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1862\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_5dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mCropping1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ add_5 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1862\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_6activatio… │\n", + "│ │ │ │ bpnet_6crop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_6dropout │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1862\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_5[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_7conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1606\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ bpnet_6dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mConv1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_7bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1606\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ bpnet_7conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_7activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1606\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_7bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_7crop │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1606\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_6dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mCropping1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ add_6 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1606\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_7activatio… │\n", + "│ │ │ │ bpnet_7crop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_7dropout │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1606\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_6[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_8conv │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1094\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m786,432\u001b[0m │ bpnet_7dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mConv1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_8bn │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1094\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m2,048\u001b[0m │ bpnet_8conv[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mBatchNormalizatio…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_8activation │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1094\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_8bn[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mActivation\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_8crop │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1094\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_7dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mCropping1D\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ add_7 (\u001b[38;5;33mAdd\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1094\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_8activatio… │\n", + "│ │ │ │ bpnet_8crop[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ bpnet_8dropout │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1094\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ add_7[\u001b[38;5;34m0\u001b[0m][\u001b[38;5;34m0\u001b[0m] │\n", + "│ (\u001b[38;5;33mDropout\u001b[0m) │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ global_average_poo… │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │ bpnet_8dropout[\u001b[38;5;34m0\u001b[0m… │\n", + "│ (\u001b[38;5;33mGlobalAveragePool…\u001b[0m │ │ │ │\n", + "├─────────────────────┼───────────────────┼────────────┼───────────────────┤\n", + "│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m19\u001b[0m) │ \u001b[38;5;34m9,747\u001b[0m │ global_average_p… │\n", + "└─────────────────────┴───────────────────┴────────────┴───────────────────┘\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
 Total params: 6,329,875 (24.15 MB)\n",
+       "
\n" + ], + "text/plain": [ + "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m6,329,875\u001b[0m (24.15 MB)\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
 Trainable params: 6,320,659 (24.11 MB)\n",
+       "
\n" + ], + "text/plain": [ + "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m6,320,659\u001b[0m (24.11 MB)\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "
 Non-trainable params: 9,216 (36.00 KB)\n",
+       "
\n" + ], + "text/plain": [ + "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m9,216\u001b[0m (36.00 KB)\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "None\n", + "2024-06-26T11:35:49.840921+0200 INFO Number of GPUs in use: 1\n", + "Epoch 1/50\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "VBox(children=(Label(value='0.030 MB of 0.030 MB uploaded\\r'), FloatProgress(value=1.0, max=1.0)))" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + " View run crested_norm_TL_spearman at: https://wandb.ai/kemp/deeppeak_benchmarking/runs/70xblei4
View project at: https://wandb.ai/kemp/deeppeak_benchmarking
Synced 6 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/html": [ + "Find logs at: ./wandb/run-20240626_113547-70xblei4/logs" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "ename": "AttributeError", + "evalue": "'SymbolicTensor' object has no attribute 'assign_add'", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mAttributeError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[12], line 13\u001b[0m\n\u001b[1;32m 10\u001b[0m \u001b[38;5;66;03m# Optionally, configure Python's logging module to suppress INFO logs\u001b[39;00m\n\u001b[1;32m 11\u001b[0m logging\u001b[38;5;241m.\u001b[39mgetLogger(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow\u001b[39m\u001b[38;5;124m'\u001b[39m)\u001b[38;5;241m.\u001b[39msetLevel(logging\u001b[38;5;241m.\u001b[39mWARNING)\n\u001b[0;32m---> 13\u001b[0m \u001b[43mtrainer\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m50\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mearly_stopping_patience\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m6\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmixed_precision\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlearning_rate_reduce\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlearning_rate_reduce_patience\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m3\u001b[39;49m\u001b[43m)\u001b[49m\n", + "File \u001b[0;32m/data/projects/c04/cbd-saerts/nkemp/software/CREsted/src/crested/tl/_crested.py:314\u001b[0m, in \u001b[0;36mCrested.fit\u001b[0;34m(self, epochs, mixed_precision, model_checkpointing, model_checkpointing_best_only, early_stopping, early_stopping_patience, learning_rate_reduce, learning_rate_reduce_patience, custom_callbacks)\u001b[0m\n\u001b[1;32m 290\u001b[0m run\u001b[38;5;241m.\u001b[39mconfig\u001b[38;5;241m.\u001b[39mupdate(\n\u001b[1;32m 291\u001b[0m {\n\u001b[1;32m 292\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mepochs\u001b[39m\u001b[38;5;124m\"\u001b[39m: epochs,\n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 310\u001b[0m }\n\u001b[1;32m 311\u001b[0m )\n\u001b[1;32m 313\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 314\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[43mfit\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 315\u001b[0m \u001b[43m \u001b[49m\u001b[43mtrain_loader\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdata\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 316\u001b[0m \u001b[43m \u001b[49m\u001b[43mvalidation_data\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mval_loader\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdata\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 317\u001b[0m \u001b[43m \u001b[49m\u001b[43mepochs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mepochs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 318\u001b[0m \u001b[43m \u001b[49m\u001b[43msteps_per_epoch\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mn_train_steps_per_epoch\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 319\u001b[0m \u001b[43m \u001b[49m\u001b[43mvalidation_steps\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mn_val_steps_per_epoch\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 320\u001b[0m \u001b[43m \u001b[49m\u001b[43mcallbacks\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcallbacks\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 321\u001b[0m \u001b[43m \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 322\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyboardInterrupt\u001b[39;00m:\n\u001b[1;32m 323\u001b[0m logger\u001b[38;5;241m.\u001b[39mwarning(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mTraining interrupted by user.\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n", + "File \u001b[0;32m~/.conda/envs/crested/lib/python3.11/site-packages/keras/src/utils/traceback_utils.py:122\u001b[0m, in \u001b[0;36mfilter_traceback..error_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 119\u001b[0m filtered_tb \u001b[38;5;241m=\u001b[39m _process_traceback_frames(e\u001b[38;5;241m.\u001b[39m__traceback__)\n\u001b[1;32m 120\u001b[0m \u001b[38;5;66;03m# To get the full stack trace, call:\u001b[39;00m\n\u001b[1;32m 121\u001b[0m \u001b[38;5;66;03m# `keras.config.disable_traceback_filtering()`\u001b[39;00m\n\u001b[0;32m--> 122\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m e\u001b[38;5;241m.\u001b[39mwith_traceback(filtered_tb) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m 123\u001b[0m \u001b[38;5;28;01mfinally\u001b[39;00m:\n\u001b[1;32m 124\u001b[0m \u001b[38;5;28;01mdel\u001b[39;00m filtered_tb\n", + "File \u001b[0;32m/data/projects/c04/cbd-saerts/nkemp/software/CREsted/src/crested/tl/metrics/_spearmancorr.py:34\u001b[0m, in \u001b[0;36mSpearmanCorrelationPerClass.update_state\u001b[0;34m(self, y_true, y_pred, sample_weight)\u001b[0m\n\u001b[1;32m 31\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;241m0.0\u001b[39m\n\u001b[1;32m 33\u001b[0m correlation \u001b[38;5;241m=\u001b[39m tf\u001b[38;5;241m.\u001b[39mcond(proceed, compute, skip)\n\u001b[0;32m---> 34\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcorrelation_sums\u001b[49m\u001b[43m[\u001b[49m\u001b[43mi\u001b[49m\u001b[43m]\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43massign_add\u001b[49m(correlation)\n\u001b[1;32m 35\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mupdate_counts[i]\u001b[38;5;241m.\u001b[39massign_add(tf\u001b[38;5;241m.\u001b[39mcast(proceed, tf\u001b[38;5;241m.\u001b[39mfloat32))\n", + "\u001b[0;31mAttributeError\u001b[0m: 'SymbolicTensor' object has no attribute 'assign_add'" + ] + } + ], "source": [ "# train the model\n", "\n", @@ -734,27 +1252,9 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 7, "metadata": {}, "outputs": [], - "source": [ - "classes = ['Astro','Endo','L2_3IT','L5ET','L5IT','L5_6NP','L6CT','L6IT','L6b','Lamp5','Micro_PVM','OPC','Oligo','Pvalb','Sncg','Sst','Sst-Chodl','VLMC','Vip']" - ] - }, - { - "cell_type": "code", - "execution_count": 13, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "The autoreload extension is already loaded. To reload it, use:\n", - " %reload_ext autoreload\n" - ] - } - ], "source": [ "%load_ext autoreload\n", "%autoreload 2\n", @@ -779,7 +1279,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 8, "metadata": {}, "outputs": [], "source": [ @@ -787,7 +1287,7 @@ "\n", "# load an existing model\n", "evaluator.load_model(\n", - " \"deeppeak_benchmarking/crested_norm_TL/checkpoints/05.keras\", compile=True\n", + " \"deeppeak_benchmarking/crested_norm_TL/checkpoints/05.keras\", compile=False\n", ")" ] }, @@ -849,6 +1349,36 @@ "evaluator.test()" ] }, + { + "cell_type": "code", + "execution_count": 43, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Index(['chr1:3093998-3096112', 'chr1:3094663-3096777', 'chr1:3111367-3113481',\n", + " 'chr1:3112727-3114841', 'chr1:3133779-3135893', 'chr1:3164901-3167015',\n", + " 'chr1:3166116-3168230', 'chr1:3180288-3182402', 'chr1:3209410-3211524',\n", + " 'chr1:3210016-3212130',\n", + " ...\n", + " 'chrX:169716892-169719006', 'chrX:169812832-169814946',\n", + " 'chrX:169824148-169826262', 'chrX:169826845-169828959',\n", + " 'chrX:169837427-169839541', 'chrX:169838199-169840313',\n", + " 'chrX:169843232-169845346', 'chrX:169862011-169864125',\n", + " 'chrX:169924670-169926784', 'chrX:169947743-169949857'],\n", + " dtype='object', length=86887)" + ] + }, + "execution_count": 43, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "adata.var_names" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -864,127 +1394,137 @@ }, { "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": 55, + "execution_count": 9, "metadata": {}, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n", + "I0000 00:00:1719406062.105265 563997 service.cc:145] XLA service 0x7f6b8400c560 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n", + "I0000 00:00:1719406062.105293 563997 service.cc:153] StreamExecutor device (0): NVIDIA H100 80GB HBM3, Compute Capability 9.0\n", + "2024-06-26 14:47:42.120878: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:268] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n", + "2024-06-26 14:47:42.210696: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:465] Loaded cuDNN version 8907\n" + ] + }, { "name": "stdout", "output_type": "stream", "text": [ - "Converted call: .flat_map_fn at 0x7fdc3f2b0220>\n", - " args: (,)\n", - " kwargs: {}\n", - "\n", - "Converted call: .get_iterator_id_fn at 0x7fdc3f2b3060>\n", - " args: (,)\n", - " kwargs: {}\n", - "\n", - "Converted call: .generator_next_fn at 0x7fdc3f2b0180>\n", - " args: ( dtype=int64>,)\n", - " kwargs: {}\n", - "\n", - "Converted call: .finalize_fn at 0x7fdc3f2b02c0>\n", - " args: ( dtype=int64>,)\n", - " kwargs: {}\n", - "\n", - "Converted call: . at 0x7fdc3f2b0360>\n", - " args: (, )\n", - " kwargs: {}\n", - "\n", - "Converted call: \n", - " args: (, )\n", - " kwargs: None\n", - "\n", - "Converted call: >\n", - " args: (, )\n", - " kwargs: {}\n", - "\n", - "Converted call: \n", - " args: (, )\n", - " kwargs: None\n", - "\n", - "Converted call: \n", - " args: (, )\n", - " kwargs: None\n", - "\n", - "Converted call: \n", - " args: (, 0)\n", - " kwargs: None\n", - "\n", - "Converted call: \n", - " args: ()\n", - " kwargs: {'shape': (2114, 4), 'dtype': tf.float32}\n", - "\n", - "Converted call: \n", - " args: (.inner_factory..tf___map_one_hot_encode..one_hot_encode at 0x7fdc714444a0>, )\n", - " kwargs: {'fn_output_signature': TensorSpec(shape=(2114, 4), dtype=tf.float32, name=None)}\n", - "\n", - "Converted call: \n", - " args: (, 'UTF-8')\n", - " kwargs: None\n", - "\n", - "Converted call: >\n", - " args: (,)\n", - " kwargs: None\n", - "\n", - "Converted call: \n", - " args: (,)\n", - " kwargs: {'depth': 4}\n", - "\n", - "Converted call: \n", - " args: (,)\n", - " kwargs: {'axis': 0}\n", - "\n", - "Converted call: .one_step_on_data_distributed at 0x7fdc71447740>\n", - " args: ([(, )],)\n", - " kwargs: {}\n", - "\n", - "Converted call: .one_step_on_data at 0x7fdc71447560>\n", - " args: ((, ),)\n", - " kwargs: {}\n", - "\n", - "\u001b[1m1385/1389\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 14ms/stepConverted call: .one_step_on_data_distributed at 0x7fdc71447740>\n", - " args: ([(, )],)\n", - " kwargs: {}\n", - "\n", - "Converted call: .one_step_on_data at 0x7fdc71447560>\n", - " args: ((, ),)\n", - " kwargs: {}\n", - "\n", - "\u001b[1m1389/1389\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m21s\u001b[0m 14ms/step\n", - "2024-06-21T12:54:06.955339+0200 INFO Adding predictions to anndata.layers[crested1].\n" + "\u001b[1m 13/1358\u001b[0m \u001b[37m━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m18s\u001b[0m 14ms/step" ] }, { - "ename": "AttributeError", - "evalue": "'numpy.ndarray' object has no attribute 'layers'", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mAttributeError\u001b[0m Traceback (most recent call last)", - "Cell \u001b[0;32mIn[55], line 5\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;66;03m# add predictions for model 1 to the adata\u001b[39;00m\n\u001b[1;32m 2\u001b[0m evaluator\u001b[38;5;241m.\u001b[39mload_model(\n\u001b[1;32m 3\u001b[0m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdeeppeak_benchmarking/crested_norm_TL/checkpoints/05.keras\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m 4\u001b[0m )\n\u001b[0;32m----> 5\u001b[0m adata \u001b[38;5;241m=\u001b[39m \u001b[43mevaluator\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpredict\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 6\u001b[0m \u001b[43m \u001b[49m\u001b[43madata\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmodel_name\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mcrested1\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\n\u001b[1;32m 7\u001b[0m \u001b[43m)\u001b[49m \u001b[38;5;66;03m# adds the predictions to the adata.layers[\"model_1\"]\u001b[39;00m\n", - "File \u001b[0;32m/data/projects/c04/cbd-saerts/nkemp/software/CREsted/src/crested/tl/_crested.py:384\u001b[0m, in \u001b[0;36mCrested.predict\u001b[0;34m(self, anndata, model_name)\u001b[0m\n\u001b[1;32m 382\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m anndata \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m model_name \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m 383\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mAdding predictions to anndata.layers[\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mmodel_name\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m].\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m--> 384\u001b[0m \u001b[43manndata\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlayers\u001b[49m[model_name] \u001b[38;5;241m=\u001b[39m predictions\u001b[38;5;241m.\u001b[39mT\n\u001b[1;32m 386\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m predictions\n", - "\u001b[0;31mAttributeError\u001b[0m: 'numpy.ndarray' object has no attribute 'layers'" + "name": "stderr", + "output_type": "stream", + "text": [ + "I0000 00:00:1719406065.390897 563997 device_compiler.h:188] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[1m1357/1358\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m━\u001b[0m \u001b[1m0s\u001b[0m 14ms/step" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n", + "I0000 00:00:1719406084.884494 564227 asm_compiler.cc:369] ptxas warning : Registers are spilled to local memory in function 'triton_gemm_dot_371', 64 bytes spill stores, 64 bytes spill loads\n", + "\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[1m1358/1358\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m25s\u001b[0m 16ms/step\n", + "2024-06-26T14:48:07.117265+0200 INFO Adding predictions to anndata.layers[crested1].\n" ] } ], "source": [ "# add predictions for model 1 to the adata\n", "evaluator.load_model(\n", - " \"deeppeak_benchmarking/crested_norm_TL/checkpoints/05.keras\"\n", + " \"deeppeak_benchmarking/crested_norm_TL/checkpoints/05.keras\", compile=False\n", ")\n", - "adata = evaluator.predict(\n", + "adata_p = evaluator.predict(\n", " adata, model_name=\"crested1\"\n", ") # adds the predictions to the adata.layers[\"model_1\"]" ] }, + { + "cell_type": "code", + "execution_count": 34, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 16ms/step\n", + "2024-06-26T14:58:27.262228+0200 INFO Plotting bar plots for region: chr18:4369383-4371497, models: ['crested1']\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "idx = 32\n", + "chrom=test_df.iloc[idx]['chr']\n", + "start=test_df.iloc[idx]['start']\n", + "end=test_df.iloc[idx]['end']\n", + "region = chrom+':'+str(start)+'-'+str(end)\n", + "region\n", + "\n", + "pred = evaluator.predict_regions(region)\n", + "crested.pl.bar.region_predictions(evaluator.anndatamodule.adata, region)" + ] + }, + { + "cell_type": "code", + "execution_count": 55, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 17ms/step\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "sequence_loader = crested.tl.data.SequenceLoader(genome_file='/home/VIB.LOCAL/niklas.kempynck/nkemp/software/dev_DeepPeak/DeepPeak/data/raw/genome.fa', chromsizes='/home/VIB.LOCAL/niklas.kempynck/nkemp/mouse/biccn/mm.chrom.sizes')\n", + "seq = sequence_loader.get_sequence(\"chr18:61107668-61109782\")\n", + "plt.figure(figsize=(25,3))\n", + "plt.bar(list(adata.obs_names),evaluator.predict_sequence(seq)[0])\n", + "plt.show()\n", + "#crested.pl.bar.region(adata_p, [\"chr18:61107668-61109782\"])" + ] + }, { "cell_type": "markdown", "metadata": {}, @@ -1003,7 +1543,7 @@ }, { "cell_type": "code", - "execution_count": 106, + "execution_count": 17, "metadata": {}, "outputs": [], "source": [ @@ -1012,37 +1552,50 @@ }, { "cell_type": "code", - "execution_count": 109, + "execution_count": 35, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "['L5ET']\n", + "2024-06-26T14:58:31.495910+0200 INFO Calculating contribution scores for 1 class(es) and 1 region(s).\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Region: 100%|██████████| 1/1 [00:02<00:00, 2.27s/it]\n" + ] + } + ], "source": [ - "# focus on two topics of interest\n", - "scores, one_hot_encoded_sequences = evaluator.calculate_contribution_scores(\n", - " [\"chr18:61107668-61109782\"], class_indices=[10], method='expected_integrated_grad'\n", - ")\n", - "\n", - "# calculate the contribution scores for two regions for all topics\n", - "# scores, one_hot_encoded_sequences = evaluator.calculate_contribution_scores(['chr1:1000-1500', 'chr1:2000-2500'], class_indices=range(len(adata.obs)))" + "cts =['L5ET']\n", + "scores, one_hot_encoded_sequences = evaluator.calculate_contribution_scores_regions(\n", + " region_idx = region, method='mutagenesis', class_names=cts\n", + ")" ] }, { "cell_type": "code", - "execution_count": 112, + "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "Crested(data=True, model=True, config=False)" + "0.650580644607544" ] }, - "execution_count": 112, + "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "evaluator" + "scores.max()" ] }, { @@ -1054,24 +1607,30 @@ }, { "cell_type": "code", - "execution_count": 113, + "execution_count": 38, "metadata": {}, "outputs": [ { - "ename": "AttributeError", - "evalue": "'numpy.ndarray' object has no attribute 'obs_names'", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mAttributeError\u001b[0m Traceback (most recent call last)", - "Cell \u001b[0;32mIn[113], line 2\u001b[0m\n\u001b[1;32m 1\u001b[0m crested\u001b[38;5;241m.\u001b[39mpl\u001b[38;5;241m.\u001b[39mcontribution_scores(\n\u001b[0;32m----> 2\u001b[0m scores, one_hot_encoded_sequences, class_indices\u001b[38;5;241m=\u001b[39m[\u001b[38;5;241m10\u001b[39m],class_names\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mlist\u001b[39m(\u001b[43madata\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mobs_names\u001b[49m)[\u001b[38;5;241m10\u001b[39m], zoom_n_bases\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m500\u001b[39m\n\u001b[1;32m 3\u001b[0m )\n", - "\u001b[0;31mAttributeError\u001b[0m: 'numpy.ndarray' object has no attribute 'obs_names'" + "name": "stdout", + "output_type": "stream", + "text": [ + "2024-06-26T14:59:13.091642+0200 INFO Plotting contribution scores for 1 sequence(s)\n" ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" } ], "source": [ - "crested.pl.contribution_scores(\n", - " scores, one_hot_encoded_sequences, class_indices=[10],class_names=list(adata.obs_names)[10], zoom_n_bases=500\n", + "crested.pl.patterns.contribution_scores(\n", + " scores, one_hot_encoded_sequences, labels=cts, zoom_n_bases=500, method='mutagenesis'\n", ")" ] }, @@ -2516,621 +3075,43 @@ "name": "stdout", "output_type": "stream", "text": [ - "2024-06-25T17:22:30.603439+0200 INFO After sorting and filtering, kept 950 regions.\n" + "2024-06-25T17:40:53.041985+0200 INFO After sorting and filtering, kept 9500 regions.\n" ] } ], "source": [ - "adata_spec = crested.pp.sort_and_filter_regions_on_specificity(adata, top_k=50, class_names=list(adata.obs_names), method='gini')" + "adata_spec = crested.pp.sort_and_filter_regions_on_specificity(adata, top_k=500, class_names=list(adata.obs_names), method='gini')" ] }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 9, "metadata": {}, "outputs": [ { "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
chrstartendClass namerankgini_score
chr10:96093683-96095797chr109609368396095797Astro10.756436
chr3:159593453-159595567chr3159593453159595567Astro20.755814
chr3:115410072-115412186chr3115410072115412186Astro30.736226
chr10:56898762-56900876chr105689876256900876Astro40.732570
chr1:143052747-143054861chr1143052747143054861Astro50.727982
chr18:66022269-66024383chr186602226966024383Astro60.727601
chrX:102587693-102589807chrX102587693102589807Astro70.725532
chrX:14301286-14303400chrX1430128614303400Astro80.724276
chrX:6452334-6454448chrX64523346454448Astro90.720401
chr6:141530067-141532181chr6141530067141532181Astro100.720374
chr18:8043916-8046030chr1880439168046030Astro110.718077
chrX:11157519-11159633chrX1115751911159633Astro120.717722
chr1:143052021-143054135chr1143052021143054135Astro130.714811
chr8:54791458-54793572chr85479145854793572Astro140.711594
chr1:5032334-5034448chr150323345034448Astro150.708968
chr3:115843758-115845872chr3115843758115845872Astro160.707991
chr6:141529548-141531662chr6141529548141531662Astro170.707742
chr6:49020327-49022441chr64902032749022441Astro180.707586
chr2:110134069-110136183chr2110134069110136183Astro190.705950
chrX:163660440-163662554chrX163660440163662554Astro200.705680
chr1:127159287-127161401chr1127159287127161401Astro210.705280
chr8:30201885-30203999chr83020188530203999Astro220.703659
chr18:8043408-8045522chr1880434088045522Astro230.703530
chr14:111252324-111254438chr14111252324111254438Astro240.703077
chr3:6427088-6429202chr364270886429202Astro250.701328
chr1:161272641-161274755chr1161272641161274755Astro260.701141
chr2:57521912-57524026chr25752191257524026Astro270.698021
chr4:154312378-154314492chr4154312378154314492Astro280.696996
chr2:65274604-65276718chr26527460465276718Astro290.696961
chr18:56440643-56442757chr185644064356442757Astro300.696608
chr15:18236980-18239094chr151823698018239094Astro310.695703
chr10:29896923-29899037chr102989692329899037Astro320.695475
chr3:47667963-47670077chr34766796347670077Astro330.694782
chr15:8814404-8816518chr1588144048816518Astro340.694436
chrX:130761597-130763711chrX130761597130763711Astro350.694203
chr5:9868290-9870404chr598682909870404Astro360.692700
chr12:90914962-90917076chr129091496290917076Astro370.692215
chr3:133051169-133053283chr3133051169133053283Astro380.691771
chr4:71483394-71485508chr47148339471485508Astro390.690659
chrX:110801591-110803705chrX110801591110803705Astro400.689756
chr14:78015493-78017607chr147801549378017607Astro410.689670
chr10:92247689-92249803chr109224768992249803Astro420.689451
chr10:56525742-56527856chr105652574256527856Astro430.689304
chr4:11761293-11763407chr41176129311763407Astro440.688151
chr1:149401421-149403535chr1149401421149403535Astro450.687990
chr2:137431719-137433833chr2137431719137433833Astro460.687783
chr1:161273161-161275275chr1161273161161275275Astro470.687692
chr15:8864466-8866580chr1588644668866580Astro480.687259
chr3:50421064-50423178chr35042106450423178Astro490.686798
chr13:15543014-15545128chr131554301415545128Astro500.684049
\n", - "
" - ], "text/plain": [ - " chr start end Class name rank \\\n", - "chr10:96093683-96095797 chr10 96093683 96095797 Astro 1 \n", - "chr3:159593453-159595567 chr3 159593453 159595567 Astro 2 \n", - "chr3:115410072-115412186 chr3 115410072 115412186 Astro 3 \n", - "chr10:56898762-56900876 chr10 56898762 56900876 Astro 4 \n", - "chr1:143052747-143054861 chr1 143052747 143054861 Astro 5 \n", - "chr18:66022269-66024383 chr18 66022269 66024383 Astro 6 \n", - "chrX:102587693-102589807 chrX 102587693 102589807 Astro 7 \n", - "chrX:14301286-14303400 chrX 14301286 14303400 Astro 8 \n", - "chrX:6452334-6454448 chrX 6452334 6454448 Astro 9 \n", - "chr6:141530067-141532181 chr6 141530067 141532181 Astro 10 \n", - "chr18:8043916-8046030 chr18 8043916 8046030 Astro 11 \n", - "chrX:11157519-11159633 chrX 11157519 11159633 Astro 12 \n", - "chr1:143052021-143054135 chr1 143052021 143054135 Astro 13 \n", - "chr8:54791458-54793572 chr8 54791458 54793572 Astro 14 \n", - "chr1:5032334-5034448 chr1 5032334 5034448 Astro 15 \n", - "chr3:115843758-115845872 chr3 115843758 115845872 Astro 16 \n", - "chr6:141529548-141531662 chr6 141529548 141531662 Astro 17 \n", - "chr6:49020327-49022441 chr6 49020327 49022441 Astro 18 \n", - "chr2:110134069-110136183 chr2 110134069 110136183 Astro 19 \n", - "chrX:163660440-163662554 chrX 163660440 163662554 Astro 20 \n", - "chr1:127159287-127161401 chr1 127159287 127161401 Astro 21 \n", - "chr8:30201885-30203999 chr8 30201885 30203999 Astro 22 \n", - "chr18:8043408-8045522 chr18 8043408 8045522 Astro 23 \n", - "chr14:111252324-111254438 chr14 111252324 111254438 Astro 24 \n", - "chr3:6427088-6429202 chr3 6427088 6429202 Astro 25 \n", - "chr1:161272641-161274755 chr1 161272641 161274755 Astro 26 \n", - "chr2:57521912-57524026 chr2 57521912 57524026 Astro 27 \n", - "chr4:154312378-154314492 chr4 154312378 154314492 Astro 28 \n", - "chr2:65274604-65276718 chr2 65274604 65276718 Astro 29 \n", - "chr18:56440643-56442757 chr18 56440643 56442757 Astro 30 \n", - "chr15:18236980-18239094 chr15 18236980 18239094 Astro 31 \n", - "chr10:29896923-29899037 chr10 29896923 29899037 Astro 32 \n", - "chr3:47667963-47670077 chr3 47667963 47670077 Astro 33 \n", - "chr15:8814404-8816518 chr15 8814404 8816518 Astro 34 \n", - "chrX:130761597-130763711 chrX 130761597 130763711 Astro 35 \n", - "chr5:9868290-9870404 chr5 9868290 9870404 Astro 36 \n", - "chr12:90914962-90917076 chr12 90914962 90917076 Astro 37 \n", - "chr3:133051169-133053283 chr3 133051169 133053283 Astro 38 \n", - "chr4:71483394-71485508 chr4 71483394 71485508 Astro 39 \n", - "chrX:110801591-110803705 chrX 110801591 110803705 Astro 40 \n", - "chr14:78015493-78017607 chr14 78015493 78017607 Astro 41 \n", - "chr10:92247689-92249803 chr10 92247689 92249803 Astro 42 \n", - "chr10:56525742-56527856 chr10 56525742 56527856 Astro 43 \n", - "chr4:11761293-11763407 chr4 11761293 11763407 Astro 44 \n", - "chr1:149401421-149403535 chr1 149401421 149403535 Astro 45 \n", - "chr2:137431719-137433833 chr2 137431719 137433833 Astro 46 \n", - "chr1:161273161-161275275 chr1 161273161 161275275 Astro 47 \n", - "chr15:8864466-8866580 chr15 8864466 8866580 Astro 48 \n", - "chr3:50421064-50423178 chr3 50421064 50423178 Astro 49 \n", - "chr13:15543014-15545128 chr13 15543014 15545128 Astro 50 \n", - "\n", - " gini_score \n", - "chr10:96093683-96095797 0.756436 \n", - "chr3:159593453-159595567 0.755814 \n", - "chr3:115410072-115412186 0.736226 \n", - "chr10:56898762-56900876 0.732570 \n", - "chr1:143052747-143054861 0.727982 \n", - "chr18:66022269-66024383 0.727601 \n", - "chrX:102587693-102589807 0.725532 \n", - "chrX:14301286-14303400 0.724276 \n", - "chrX:6452334-6454448 0.720401 \n", - "chr6:141530067-141532181 0.720374 \n", - "chr18:8043916-8046030 0.718077 \n", - "chrX:11157519-11159633 0.717722 \n", - "chr1:143052021-143054135 0.714811 \n", - "chr8:54791458-54793572 0.711594 \n", - "chr1:5032334-5034448 0.708968 \n", - "chr3:115843758-115845872 0.707991 \n", - "chr6:141529548-141531662 0.707742 \n", - "chr6:49020327-49022441 0.707586 \n", - "chr2:110134069-110136183 0.705950 \n", - "chrX:163660440-163662554 0.705680 \n", - "chr1:127159287-127161401 0.705280 \n", - "chr8:30201885-30203999 0.703659 \n", - "chr18:8043408-8045522 0.703530 \n", - "chr14:111252324-111254438 0.703077 \n", - "chr3:6427088-6429202 0.701328 \n", - "chr1:161272641-161274755 0.701141 \n", - "chr2:57521912-57524026 0.698021 \n", - "chr4:154312378-154314492 0.696996 \n", - "chr2:65274604-65276718 0.696961 \n", - "chr18:56440643-56442757 0.696608 \n", - "chr15:18236980-18239094 0.695703 \n", - "chr10:29896923-29899037 0.695475 \n", - "chr3:47667963-47670077 0.694782 \n", - "chr15:8814404-8816518 0.694436 \n", - "chrX:130761597-130763711 0.694203 \n", - "chr5:9868290-9870404 0.692700 \n", - "chr12:90914962-90917076 0.692215 \n", - "chr3:133051169-133053283 0.691771 \n", - "chr4:71483394-71485508 0.690659 \n", - "chrX:110801591-110803705 0.689756 \n", - "chr14:78015493-78017607 0.689670 \n", - "chr10:92247689-92249803 0.689451 \n", - "chr10:56525742-56527856 0.689304 \n", - "chr4:11761293-11763407 0.688151 \n", - "chr1:149401421-149403535 0.687990 \n", - "chr2:137431719-137433833 0.687783 \n", - "chr1:161273161-161275275 0.687692 \n", - "chr15:8864466-8866580 0.687259 \n", - "chr3:50421064-50423178 0.686798 \n", - "chr13:15543014-15545128 0.684049 " + "chr chr3\n", + "start 38667902\n", + "end 38670016\n", + "Class name Endo\n", + "rank 500\n", + "gini_score 0.537468\n", + "Name: chr3:38667902-38670016, dtype: object" ] }, - "execution_count": 7, + "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "adata_spec.var.head(n=50)" + "adata_spec.var.iloc[999]" ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 10, "metadata": {}, "outputs": [], "source": [ @@ -3145,33 +3126,32 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": null, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "2024-06-25T16:46:35.169387+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:47:37.441391+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:48:39.528406+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:49:41.675013+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:50:44.036741+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:51:46.093464+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:52:48.344770+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:53:50.088466+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:54:51.501521+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:55:53.001062+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:56:54.033865+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:57:55.182021+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:58:56.624829+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T16:59:57.935584+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T17:00:58.899692+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T17:02:00.035373+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T17:03:01.079742+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T17:04:02.123088+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "2024-06-25T17:05:02.980093+0200 INFO Calculating contribution scores for 1 class(es) and 50 region(s).\n", - "Contribution scores and one-hot encoded sequences saved to modisco_results\n" + "2024-06-25T17:42:59.294645+0200 INFO Calculating contribution scores for 1 class(es) and 500 region(s).\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "2024-06-25 17:42:59.782263: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:465] Loaded cuDNN version 8907\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "2024-06-25T17:53:24.646278+0200 INFO Calculating contribution scores for 1 class(es) and 500 region(s).\n", + "2024-06-25T18:03:38.499302+0200 INFO Calculating contribution scores for 1 class(es) and 500 region(s).\n", + "2024-06-25T18:13:50.544979+0200 INFO Calculating contribution scores for 1 class(es) and 500 region(s).\n", + "2024-06-25T18:24:02.370729+0200 INFO Calculating contribution scores for 1 class(es) and 500 region(s).\n", + "2024-06-25T18:34:14.363745+0200 INFO Calculating contribution scores for 1 class(es) and 500 region(s).\n" ] } ], @@ -3181,59 +3161,65 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": null, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "2024-06-25T22:00:20.552977+0200 INFO Running modisco for class: L6CT\n", + "Using 1493 positive seqlets\n" + ] + } + ], + "source": [ + "crested.tl.tfmodisco(window=1000, output_dir = 'modisco_results2')" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [], + "source": [ + "import numpy as np" + ] + }, + { + "cell_type": "code", + "execution_count": 29, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "2024-06-25T17:33:52.595025+0200 INFO Running modisco for class: L6CT\n", - "Using 142 positive seqlets\n", - "2024-06-25T17:34:00.034431+0200 INFO Running modisco for class: L6IT\n", - "Using 158 positive seqlets\n", - "2024-06-25T17:34:01.274526+0200 INFO Running modisco for class: Astro\n", - "Using 302 positive seqlets\n", - "2024-06-25T17:34:04.937461+0200 INFO Running modisco for class: L5IT\n", - "Using 128 positive seqlets\n", - "2024-06-25T17:34:05.687151+0200 INFO Running modisco for class: Lamp5\n", - "Using 158 positive seqlets\n", - "2024-06-25T17:34:06.992156+0200 INFO Running modisco for class: VLMC\n", - "Using 228 positive seqlets\n", - "2024-06-25T17:34:09.344291+0200 INFO Running modisco for class: Pvalb\n", - "Using 167 positive seqlets\n", - "2024-06-25T17:34:10.579415+0200 INFO Running modisco for class: Micro_PVM\n", - "Using 270 positive seqlets\n", - "2024-06-25T17:34:12.717802+0200 INFO Running modisco for class: Sncg\n", - "Using 145 positive seqlets\n", - "2024-06-25T17:34:13.627478+0200 INFO Running modisco for class: L2_3IT\n", - "Using 165 positive seqlets\n", - "2024-06-25T17:34:14.704486+0200 INFO Running modisco for class: L5ET\n", - "Using 125 positive seqlets\n", - "2024-06-25T17:34:15.472059+0200 INFO Running modisco for class: L6b\n", - "Using 148 positive seqlets\n", - "2024-06-25T17:34:16.704746+0200 INFO Running modisco for class: Vip\n", - "Using 131 positive seqlets\n", - "2024-06-25T17:34:17.608582+0200 INFO Running modisco for class: SstChodl\n", - "Using 271 positive seqlets\n", - "Extracted 167 negative seqlets\n", - "2024-06-25T17:34:21.328252+0200 INFO Running modisco for class: Oligo\n", - "Using 305 positive seqlets\n", - "Extracted 125 negative seqlets\n", - "2024-06-25T17:34:25.336564+0200 INFO Running modisco for class: Sst\n", - "Using 139 positive seqlets\n", - "2024-06-25T17:34:26.232381+0200 INFO Running modisco for class: L5_6NP\n", - "Using 125 positive seqlets\n", - "2024-06-25T17:34:27.018244+0200 INFO Running modisco for class: Endo\n", - "Extracted 105 negative seqlets\n", - "2024-06-25T17:34:27.667426+0200 INFO Running modisco for class: OPC\n", - "Using 246 positive seqlets\n", - "Extracted 152 negative seqlets\n" + "2024-06-26T11:51:51.428622+0200 INFO Starting genomic contributions plot for classes: ['L5ET']\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/data/projects/c04/cbd-saerts/nkemp/software/CREsted/src/crested/pl/_utils.py:52: UserWarning: The figure layout has changed to tight\n", + " plt.tight_layout()\n" ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" } ], "source": [ - "crested.tl.tfmodisco(window=1000)" + "%matplotlib inline\n", + "crested.pl.patterns.modisco_results(classes=['L5ET'], contribution='postive', contribution_dir='modisco_results2', num_seq=500, y_max=0.07, viz='contrib')" ] }, { diff --git a/src/crested/pl/patterns/_contribution_scores.py b/src/crested/pl/patterns/_contribution_scores.py index 4209994a..798269ea 100644 --- a/src/crested/pl/patterns/_contribution_scores.py +++ b/src/crested/pl/patterns/_contribution_scores.py @@ -9,7 +9,7 @@ from crested._logging import log_and_raise from crested.pl._utils import render_plot -from ._utils import _plot_attribution_map, grad_times_input_to_df +from ._utils import _plot_attribution_map, _plot_mutagenesis_map, grad_times_input_to_df, grad_times_input_to_df_mutagenesis @log_and_raise(ValueError) @@ -37,6 +37,7 @@ def contribution_scores( zoom_n_bases: int | None = None, highlight_positions: list[tuple[int, int]] | None = None, ylim: tuple | None = None, + method: str | None = None, **kwargs, ): """ @@ -58,6 +59,8 @@ def contribution_scores( List of tuples with start and end positions to highlight. Default is None. ylim Y-axis limits. Default is None. + method + Method used for calculating contribution scores. If mutagenesis, specify. Examples -------- @@ -80,20 +83,36 @@ def contribution_scores( start_idx = center - int(zoom_n_bases / 2) scores = scores[:, :, start_idx : start_idx + zoom_n_bases, :] - global_min = scores.min() - global_max = scores.max() # Plot logger.info(f"Plotting contribution scores for {seqs_one_hot.shape[0]} sequence(s)") for seq in range(seqs_one_hot.shape[0]): fig_height_per_class = 2 fig = plt.figure(figsize=(50, fig_height_per_class * scores.shape[1])) + seq_class_x = seqs_one_hot[seq, start_idx : start_idx + zoom_n_bases, :] + + if method == 'mutagenesis': + global_max = scores[seq].max()+0.25*np.abs(scores[seq].max()) + global_min = scores[seq].min()-0.25*np.abs(scores[seq].min()) + else: + mins = [] + maxs = [] + for i in range(scores.shape[1]): + seq_class_scores = scores[seq, i, :, :] + mins.append(np.min(seq_class_scores*seq_class_x)) + maxs.append(np.max(seq_class_scores*seq_class_x)) + global_max = np.array(maxs).max()+0.25*np.abs(np.array(maxs).max()) + global_min = np.array(mins).min()-0.25*np.abs(np.array(mins).min()) + for i in range(scores.shape[1]): seq_class_scores = scores[seq, i, :, :] - seq_class_x = seqs_one_hot[seq, :, :] - intgrad_df = grad_times_input_to_df(seq_class_x, seq_class_scores) ax = plt.subplot(scores.shape[1], 1, i + 1) - _plot_attribution_map(intgrad_df, ax=ax, return_ax=False) + if (method =='mutagenesis'): + mutagenesis_df = grad_times_input_to_df_mutagenesis(seq_class_x, seq_class_scores) + _plot_mutagenesis_map(mutagenesis_df, ax=ax) + else: + intgrad_df = grad_times_input_to_df(seq_class_x, seq_class_scores) + _plot_attribution_map(intgrad_df, ax=ax, return_ax=False) if labels: class_name = labels[i] else: @@ -102,11 +121,11 @@ def contribution_scores( if ylim: ax.set_ylim(ylim[0], ylim[1]) x_pos = 5 - y_pos = 0.75 * ylim[1] + y_pos = 0.5 * ylim[1] else: ax.set_ylim([global_min, global_max]) x_pos = 5 - y_pos = 0.75 * global_max + y_pos = 0.5 * global_max ax.text(x_pos, y_pos, text_to_add, fontsize=16, ha="left", va="center") # Draw rectangles to highlight positions diff --git a/src/crested/pl/patterns/_modisco_results.py b/src/crested/pl/patterns/_modisco_results.py index e050ee87..8d385616 100644 --- a/src/crested/pl/patterns/_modisco_results.py +++ b/src/crested/pl/patterns/_modisco_results.py @@ -45,7 +45,7 @@ def _trim_pattern_by_ic( contrib_scores = np.array(pattern["contrib_scores"]) if not pos_pattern: contrib_scores = -contrib_scores - contrib_scores[contrib_scores < 0] = 0 + contrib_scores[contrib_scores < 0] = 1e-9 # avoid division by zero ic = modisco.util.compute_per_position_ic( ppm=np.array(contrib_scores), background=background, pseudocount=pseudocount diff --git a/src/crested/pl/patterns/_utils.py b/src/crested/pl/patterns/_utils.py index ede20c84..8a57596f 100644 --- a/src/crested/pl/patterns/_utils.py +++ b/src/crested/pl/patterns/_utils.py @@ -71,3 +71,28 @@ def _plot_attribution_map( ax.spines["top"].set_visible(False) if return_ax: return ax + +def _plot_mutagenesis_map(mutagenesis_df, ax=None): + """Plot an attribution map for mutagenesis using different colored dots, with adjusted x-axis limits.""" + colors = {'A': 'green', 'C': 'blue', 'G': 'orange', 'T': 'red'} + if ax is None: + ax = plt.gca() + + # Add horizontal line at y=0 + ax.axhline(0, color='gray', linewidth=1, linestyle='--') + + # Scatter plot for each nucleotide type + for nuc, color in colors.items(): + # Filter out dots where the variant is the same as the original nucleotide + subset = mutagenesis_df[(mutagenesis_df['Nucleotide'] == nuc) & (mutagenesis_df['Nucleotide'] != mutagenesis_df['Original'])] + ax.scatter(subset['Position'], subset['Effect'], color=color, label=nuc, s=10) # s is the size of the dot + + # Set the limits of the x-axis to match exactly the first and last position + if not mutagenesis_df.empty: + ax.set_xlim(mutagenesis_df['Position'].min() - 0.5, mutagenesis_df['Position'].max() + 0.5) + + ax.legend(title="Nucleotide", loc='upper right') + ax.spines["right"].set_visible(False) + ax.spines["top"].set_visible(False) + ax.xaxis.set_ticks_position("none") + plt.xticks([]) # Optionally, hide x-axis ticks for a cleaner look diff --git a/src/crested/tl/_configs.py b/src/crested/tl/_configs.py index f5568386..09775bec 100644 --- a/src/crested/tl/_configs.py +++ b/src/crested/tl/_configs.py @@ -13,6 +13,7 @@ PearsonCorrelation, PearsonCorrelationLog, ZeroPenaltyMetric, + SpearmanCorrelationPerClass ) @@ -74,6 +75,9 @@ def metrics(self) -> list[tf.keras.metrics.Metric]: class PeakRegressionConfig(BaseConfig): """Default configuration for peak regression task.""" + def __init__(self, num_classes=None): + self.num_classes = num_classes + @property def loss(self) -> tf.keras.losses.Loss: return CosineMSELoss() @@ -84,15 +88,18 @@ def optimizer(self) -> tf.keras.optimizers.Optimizer: @property def metrics(self) -> list[tf.keras.metrics.Metric]: - return [ + metrics = [ tf.keras.metrics.MeanAbsoluteError(), tf.keras.metrics.MeanSquaredError(), tf.keras.metrics.CosineSimilarity(axis=1), PearsonCorrelation(), ConcordanceCorrelationCoefficient(), PearsonCorrelationLog(), - ZeroPenaltyMetric(), + ZeroPenaltyMetric() ] + #if self.num_classes is not None: + # metrics.append(SpearmanCorrelationPerClass(num_classes=self.num_classes)) + return metrics class TaskConfig(NamedTuple): @@ -156,7 +163,7 @@ def to_dict(self) -> dict: def default_configs( - task: str, + task: str, num_classes: int = None ) -> TaskConfig: """ Get default loss, optimizer, and metrics for an existing task. @@ -177,6 +184,8 @@ def default_configs( ---------- tasks Task for which to get default components. + num_classes + Number of output classes of model. Required for Spearman correlation metric. Returns ------- @@ -196,7 +205,7 @@ def default_configs( f"Task '{task}' not supported. Only {list(task_classes.keys())} are supported." ) - task_class = task_classes[task]() + task_class = task_classes[task](num_classes=num_classes) if task =='peak_regression' else task_classes[task]() loss = task_class.loss optimizer = task_class.optimizer metrics = task_class.metrics diff --git a/src/crested/tl/_crested.py b/src/crested/tl/_crested.py index 8c0207ad..c7d7cf8f 100644 --- a/src/crested/tl/_crested.py +++ b/src/crested/tl/_crested.py @@ -430,6 +430,31 @@ def predict_regions( return np.concatenate(all_predictions, axis=0) + def predict_sequence( + self, + sequence: str) -> np.ndarray: + """ + Make predictions using the model on the provided DNA sequence. + + Parameters + ---------- + model : a trained TensorFlow/Keras model + sequence : str + A string containing a DNA sequence (A, C, G, T). + + Returns + ------- + np.ndarray + Predictions for the provided sequence. + """ + # One-hot encode the sequence + x = one_hot_encode_sequence(sequence) + + # Make prediction + predictions = self.model.predict(x) + + return predictions + def calculate_contribution_scores( self, anndata: AnnData | None = None, @@ -584,12 +609,16 @@ def calculate_contribution_scores_regions( if isinstance(region_idx, str): region_idx = [region_idx] + if isinstance(class_names, str): + class_names = [class_names] + all_scores = [] all_one_hot_sequences = [] all_class_names = list(self.anndatamodule.adata.obs_names) if class_names is not None: + print(class_names) n_classes = len(class_names) class_indices = [ all_class_names.index(class_name) for class_name in class_names diff --git a/src/crested/tl/data/__init__.py b/src/crested/tl/data/__init__.py index a49456aa..95714fe8 100644 --- a/src/crested/tl/data/__init__.py +++ b/src/crested/tl/data/__init__.py @@ -1,3 +1,3 @@ from ._anndatamodule import AnnDataModule from ._dataloader import AnnDataLoader -from ._dataset import AnnDataset +from ._dataset import AnnDataset, SequenceLoader diff --git a/src/crested/tl/data/_dataset.py b/src/crested/tl/data/_dataset.py index 49002666..f220b64c 100644 --- a/src/crested/tl/data/_dataset.py +++ b/src/crested/tl/data/_dataset.py @@ -27,9 +27,9 @@ def __init__( self, genome_file: PathLike, chromsizes: dict | None, - in_memory: bool, - always_reverse_complement: bool, - max_stochastic_shift: int, + in_memory: bool = False, + always_reverse_complement: bool = False, + max_stochastic_shift: int = 0, regions: list[str] = None, ): self.genome = FastaFile(genome_file) @@ -82,7 +82,6 @@ def get_sequence(self, region: str, strand: str = "+", shift: int = 0) -> str: sequence = self.sequences[key] else: sequence = self._get_extended_sequence(region) - chrom, start_end = region.split(":") start, end = map(int, start_end.split("-")) start_idx = self.max_stochastic_shift + shift diff --git a/src/crested/tl/losses/_cosinemse.py b/src/crested/tl/losses/_cosinemse.py index c47ec019..3a7bd35a 100644 --- a/src/crested/tl/losses/_cosinemse.py +++ b/src/crested/tl/losses/_cosinemse.py @@ -6,10 +6,10 @@ class CosineMSELoss(tf.keras.losses.Loss): """Custom loss function that combines cosine similarity and mean squared error.""" - def __init__(self, max_weight=1.0, name="CustomMSELoss", reduction=None): + def __init__(self, max_weight=1.0, name="CustomMSELoss"): super().__init__(name=name) self.max_weight = max_weight - self.reduction=reduction + #self.reduction=reduction @tf.function def call(self, y_true, y_pred): @@ -39,8 +39,8 @@ def call(self, y_true, y_pred): def get_config(self): config = super().get_config() config.update({ - "max_weight": self.max_weight,#}) - "reduction":self.reduction}) + "max_weight": self.max_weight}) + #"reduction":self.reduction}) return config @classmethod diff --git a/src/crested/tl/metrics/__init__.py b/src/crested/tl/metrics/__init__.py index 38ad2875..deb4096b 100644 --- a/src/crested/tl/metrics/__init__.py +++ b/src/crested/tl/metrics/__init__.py @@ -2,3 +2,4 @@ from ._pearsoncorr import PearsonCorrelation from ._pearsoncorrlog import PearsonCorrelationLog from ._zeropenalty import ZeroPenaltyMetric +from ._spearmancorr import SpearmanCorrelationPerClass diff --git a/src/crested/tl/metrics/_spearmancorr.py b/src/crested/tl/metrics/_spearmancorr.py new file mode 100644 index 00000000..4ccaeee1 --- /dev/null +++ b/src/crested/tl/metrics/_spearmancorr.py @@ -0,0 +1,55 @@ +"""Spearman correlation metric.""" + +from __future__ import annotations +import tensorflow as tf + +@tf.keras.utils.register_keras_serializable(package="Metrics") +class SpearmanCorrelationPerClass(tf.keras.metrics.Metric): + def __init__(self, num_classes, name='spearman_correlation_per_class', **kwargs): + super(SpearmanCorrelationPerClass, self).__init__(name=name, **kwargs) + self.num_classes = num_classes + self.correlation_sums = self.add_weight(name='correlation_sums', shape=(num_classes,), initializer='zeros') + self.update_counts = self.add_weight(name='update_counts', shape=(num_classes,), initializer='zeros') + + def update_state(self, y_true, y_pred, sample_weight=None): + for i in range(self.num_classes): + y_true_class = tf.cast(y_true[:, i], tf.float32) + y_pred_class = tf.cast(y_pred[:, i], tf.float32) + + non_zero_indices = tf.where(tf.not_equal(y_true_class, 0)) + y_true_non_zero = tf.gather(y_true_class, non_zero_indices) + y_pred_non_zero = tf.gather(y_pred_class, non_zero_indices) + + # Ensure sizes are constant by checking them before the operation + num_elements = tf.size(y_true_non_zero) + proceed = num_elements > 1 + + def compute(): + return self.compute_correlation(y_true_non_zero, y_pred_non_zero) + + def skip(): + return 0.0 + + correlation = tf.cond(proceed, compute, skip) + self.correlation_sums[i].assign_add(correlation) + self.update_counts[i].assign_add(tf.cast(proceed, tf.float32)) + + def compute_correlation(self, y_true_non_zero, y_pred_non_zero): + ranks_true = tf.argsort(tf.argsort(y_true_non_zero)) + ranks_pred = tf.argsort(tf.argsort(y_pred_non_zero)) + + rank_diffs = tf.cast(ranks_true, tf.float32) - tf.cast(ranks_pred, tf.float32) + rank_diffs_squared_sum = tf.reduce_sum(tf.square(rank_diffs)) + n = tf.cast(tf.size(y_true_non_zero), tf.float32) + + correlation = 1 - (6 * rank_diffs_squared_sum) / (n * (n*n - 1)) + return tf.where(tf.math.is_nan(correlation), 0.0, correlation) + + def result(self): + valid_counts = self.update_counts + avg_correlations = self.correlation_sums / valid_counts + return tf.reduce_mean(avg_correlations) + + def reset_state(self): + self.correlation_sums.assign(tf.zeros_like(self.correlation_sums)) + self.update_counts.assign(tf.zeros_like(self.update_counts)) \ No newline at end of file