Adding more documentation

examples/mvp.ipynb

@@ -362,7 +362,7 @@
    "source": [
     "class FlaxKompressor(Kompressor):\n",
     "    def __init__(\n",
-    "        self, encode_fn, decode_fn, padding, model_fn, predictions_fn, seed=None\n",
+    "        self, encode_fn, decode_fn, padding, model_fn, predictions_fn\n",
     "    ):\n",
     "        super().__init__(encode_fn=encode_fn, decode_fn=decode_fn, padding=padding)\n",
     "        self.model_fn = model_fn\n",
@@ -541,7 +541,7 @@
     "                    )\n",
     "\n",
     "                    for label in level_encoded_maps.keys() & level_highres_maps.keys():\n",
-    "                        summaries[(f\"{writer_path}/{label}/run_length\")].extend(\n",
+    "                        summaries[f\"{writer_path}/{label}/run_length\"].extend(\n",
     "                            list(\n",
     "                                kom.image.metrics.mean_run_length(\n",
     "                                    level_encoded_maps[label]\n",
@@ -591,7 +591,7 @@
     "        A Numpy array containing the decoded data along with an additional axis to create Height, Width, Channel.\n",
     "\n",
     "    Examples:\n",
-    "        If we have an Data path array of (\"/tmp/0.mrc\",0) with shape (5,5) this will\n",
+    "        If we have a Data path array of (\"/tmp/0.mrc\",0) with shape (5,5) this will\n",
     "        return the numpy array with the shape (5,5,1)\n",
     "    \"\"\"\n",
     "    mrc_file = mrcfile.mmap(data_paths[0])\n",
@@ -616,9 +616,9 @@
     "        A list of paths and frames.\n",
     "\n",
     "    Example:\n",
-    "        If we have an MRC file file shape 1,2,3 at /tmp/0.mrc then::\n",
+    "        If we have an MRCfile shape 1,2,3 at /tmp/0.mrc then::\n",
     "\n",
-    "          data = get_data_paths_and_frames(\"/tmp/0.mrc)\n",
+    "          data = get_data_paths_and_frames(\"/tmp/0.mrc\")\n",
     "\n",
     "        data will be:\n",
     "          [(\"/tmp/0.mrc\",0), (\"/tmp/0.mrc\",1),(\"/tmp/0.mrc\",2)]\n",
@@ -742,7 +742,7 @@
     "    def __call__(self, frame: np.ndarray) -> np.ndarray:\n",
     "        \"\"\"Batch to downsample from a high-resolution image.\n",
     "        Args:\n",
-    "            batch: High resolution batch to downsample. Shape [height, width, channel]\n",
+    "            frame: High resolution batch to downsample. Shape [height, width, channel]\n",
     "        Return:\n",
     "            np.ndarray: Downsampled batch: Shape [height, width, channel]\n",
     "        \"\"\"\n",
@@ -1012,7 +1012,7 @@
    "source": [
     "# Train dataset gets transformed for the creation of maps\n",
     "train_dataset = MRCFileDataset(train_data, data_transforms)\n",
-    "# Test dataset does not need transformation as this is used to evalaute the performance of the model so only needs highres images\n",
+    "# Test dataset does not need transformation as this is used to evaluate the performance of the model so only needs highres images\n",
     "test_data = MRCFileDataset(train_data)"
    ],
    "outputs": [],
@@ -1116,7 +1116,7 @@
     "    padding=config[\"padding\"],\n",
     "    model_fn=ConvSRM(config[\"neighbourhood\"]),\n",
     "    predictions_fn=RegressionPredictor,\n",
-    ").init(train_dataloader)"
+    ").init(train_dataloader, seed = config[\"seed\"])"
    ],
    "outputs": [],
    "execution_count": null
@@ -1209,7 +1209,7 @@
     "        f\"baseline/{mode}/run_length\",\n",
     "    )\n",
     "    for highres in tqdm(dataset):\n",
-    "        summaries[(f\"{writer_path}\")].extend(\n",
+    "        summaries[f\"{writer_path}\"].extend(\n",
     "            list(kom.image.metrics.mean_run_length(highres).flatten())\n",
     "        )\n",
     "    return jnp.array(summaries[writer_path]).flatten().mean()"
@@ -1259,7 +1259,7 @@
     "            )\n",
     "\n",
     "            for label in level_encoded_maps.keys() & level_highres_maps.keys():\n",
-    "                summaries[(f\"{writer_path}/{label}/run_length\")].extend(\n",
+    "                summaries[f\"{writer_path}/{label}/run_length\"].extend(\n",
     "                    list(\n",
     "                        kom.image.metrics.mean_run_length(\n",
     "                            level_encoded_maps[label]\n",