Training with [1,2,3] channels

bin/train.sh

@@ -4,7 +4,7 @@ export CUDA_VISIBLE_DEVICES=2,3
 RUN_CONFIG=config.yml
 
 
-LOGDIR=/raid/bac/kaggle/logs/recursion_cell/se_resnext50_32x4d/
+LOGDIR=/raid/bac/kaggle/logs/recursion_cell/test/rgb/se_resnext50_32x4d/
 catalyst-dl run \
     --config=./configs/${RUN_CONFIG} \
     --logdir=$LOGDIR \

configs/config.yml

@@ -1,7 +1,7 @@
 model_params:
   model: cell_senet
   model_name: se_resnext50_32x4d
-  n_channels: 6
+  n_channels: 3
   num_classes: 1108
 
 args:
@@ -40,6 +40,8 @@ stages:
     train_csv: "./csv/train_0.csv"
     valid_csv: "./csv/valid_0.csv"
     root: "/raid/data/kaggle/recursion-cellular-image-classification/"
+    site: 1
+    channels: [1, 2, 3]
 
   stage1:
 

preprocessing/combinations.py

@@ -0,0 +1,6 @@
+from itertools import combinations
+
+a = [1, 2, 3, 4, 5, 6]
+result = [list(i) for i in combinations(a, 3)]
+print(result)
+print(len(result))

preprocessing/convert_weight.py

@@ -0,0 +1,6 @@
+import torch
+
+import pdb
+pdb.set_trace()
+checkpoint = torch.load("../pretrained/results/se_resnext50.attention.per_image_norm.1024/checkpoint/swa.10.022.pth", map_location='cpu')
+

src/dataset.py

@@ -6,6 +6,35 @@
 
 NUM_CLASSES = 1108
 
+DEFAULT_CHANNELS = (1, 2, 3, 4, 5, 6)
+RGB_MAP = {
+    1: {
+        'rgb': np.array([19, 0, 249]),
+        'range': [0, 51]
+    },
+    2: {
+        'rgb': np.array([42, 255, 31]),
+        'range': [0, 107]
+    },
+    3: {
+        'rgb': np.array([255, 0, 25]),
+        'range': [0, 64]
+    },
+    4: {
+        'rgb': np.array([45, 255, 252]),
+        'range': [0, 191]
+    },
+    5: {
+        'rgb': np.array([250, 0, 253]),
+        'range': [0, 89]
+    },
+    6: {
+        'rgb': np.array([254, 255, 40]),
+        'range': [0, 191]
+    }
+}
+
+
 
 def load_image(path):
     image = cv2.imread(path, 0)
@@ -61,6 +90,39 @@ def load_images_as_tensor(image_paths, dtype=np.uint8):
     return data
 
 
+def convert_tensor_to_rgb(t, channels=DEFAULT_CHANNELS, vmax=255, rgb_map=RGB_MAP):
+    """
+    Converts and returns the image data as RGB image
+    Parameters
+    ----------
+    t : np.ndarray
+        original image data
+    channels : list of int
+        channels to include
+    vmax : int
+        the max value used for scaling
+    rgb_map : dict
+        the color mapping for each channel
+        See rxrx.io.RGB_MAP to see what the defaults are.
+    Returns
+    -------
+    np.ndarray the image data of the site as RGB channels
+    """
+    colored_channels = []
+    for i, channel in enumerate(channels):
+        x = (t[:, :, i] / vmax) / \
+            ((rgb_map[channel]['range'][1] - rgb_map[channel]['range'][0]) / 255) + \
+            rgb_map[channel]['range'][0] / 255
+        x = np.where(x > 1., 1., x)
+        x_rgb = np.array(
+            np.outer(x, rgb_map[channel]['rgb']).reshape(512, 512, 3),
+            dtype=int)
+        colored_channels.append(x_rgb)
+    im = np.array(np.array(colored_channels).sum(axis=0), dtype=int)
+    im = np.where(im > 255, 255, im)
+    return im
+
+
 class RecursionCellularSite(Dataset):
 
     def __init__(self,
@@ -111,6 +173,7 @@ def __getitem__(self, idx):
         ]
 
         image = load_images_as_tensor(channel_paths, dtype=np.float32)
+        # image = convert_tensor_to_rgb(image)
         image = image / 255
         if self.transform:
             image = self.transform(image=image)['image']

src/make_submission.py

@@ -34,15 +34,15 @@ def predict(model, loader):
 
 def predict_all():
     test_csv = '/raid/data/kaggle/recursion-cellular-image-classification/test.csv'
-    log_dir = "/raid/bac/kaggle/logs/recursion_cell/se_resnext50_32x4d/"
+    log_dir = "/raid/bac/kaggle/logs/recursion_cell/test/rgb/se_resnext50_32x4d/"
     root = "/raid/data/kaggle/recursion-cellular-image-classification/"
     site = 1
-    channels = [1,2,3,4,5,6]
+    channels = [1,2,3]
 
     model = cell_senet(
         model_name="se_resnext50_32x4d",
         num_classes=1108,
-        n_channels=6
+        n_channels=len(channels)
     )
 
     checkpoint = f"{log_dir}/checkpoints/best.pth"
@@ -74,7 +74,7 @@ def predict_all():
     submission = df.copy()
     submission['sirna'] = all_preds.astype(int)
     os.makedirs("submission", exist_ok=True)
-    submission.to_csv('./submission/submission_se_resnext50_32x4d.csv', index=False, columns=['id_code', 'sirna'])
+    submission.to_csv('./submission/submission_se_resnext50_32x4d_rgb.csv', index=False, columns=['id_code', 'sirna'])
 
 
 if __name__ == '__main__':

src/models/senet.py

@@ -10,7 +10,7 @@ def cell_senet(model_name='se_resnext50', num_classes=1108, n_channels=6):
         num_classes=num_classes,
         pretrained=True
     )
-    print(model)
+    # print(model)
     conv1 = model._features[0].conv1
     model._features[0].conv1 = nn.Conv2d(in_channels=n_channels,
                             out_channels=conv1.out_channels,