Pre-commit fixes, optimizer loading fix

1 files changed, 109 insertions, 0 deletions

diff --git a/notebooks/00-testing-stuff-out.ipynb b/notebooks/00-testing-stuff-out.ipynb
index 8f2e3f8..4c93501 100644
--- a/notebooks/00-testing-stuff-out.ipynb
+++ b/notebooks/00-testing-stuff-out.ipynb
@@ -26,6 +26,115 @@
   {
    "cell_type": "code",
    "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from omegaconf import OmegaConf"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "path = \"../training/experiments/image_transformer.yaml\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "conf = OmegaConf.load(path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "network:\n",
+      "  type: ImageTransformer\n",
+      "  args:\n",
+      "    input_shape: None\n",
+      "    output_shape: None\n",
+      "    encoder:\n",
+      "      type: None\n",
+      "      args: None\n",
+      "    mapping: sentence_piece\n",
+      "    num_decoder_layers: 4\n",
+      "    hidden_dim: 256\n",
+      "    num_heads: 4\n",
+      "    expansion_dim: 1024\n",
+      "    dropout_rate: 0.1\n",
+      "    transformer_activation: glu\n",
+      "model:\n",
+      "  type: LitTransformerModel\n",
+      "  args:\n",
+      "    optimizer:\n",
+      "      type: MADGRAD\n",
+      "      args:\n",
+      "        lr: 0.01\n",
+      "        momentum: 0.9\n",
+      "        weight_decay: 0\n",
+      "        eps: 1.0e-06\n",
+      "    lr_scheduler:\n",
+      "      type: CosineAnnealingLR\n",
+      "      args:\n",
+      "        T_max: 512\n",
+      "    criterion:\n",
+      "      type: CrossEntropyLoss\n",
+      "      args:\n",
+      "        weight: None\n",
+      "        ignore_index: -100\n",
+      "        reduction: mean\n",
+      "    monitor: val_loss\n",
+      "    mapping: sentence_piece\n",
+      "data:\n",
+      "  type: IAMExtendedParagraphs\n",
+      "  args:\n",
+      "    batch_size: 16\n",
+      "    num_workers: 12\n",
+      "    train_fraction: 0.8\n",
+      "    augment: true\n",
+      "callbacks:\n",
+      "- type: ModelCheckpoint\n",
+      "  args:\n",
+      "    monitor: val_loss\n",
+      "    mode: min\n",
+      "- type: EarlyStopping\n",
+      "  args:\n",
+      "    monitor: val_loss\n",
+      "    mode: min\n",
+      "    patience: 10\n",
+      "trainer:\n",
+      "  args:\n",
+      "    stochastic_weight_avg: true\n",
+      "    auto_scale_batch_size: power\n",
+      "    gradient_clip_val: 0\n",
+      "    fast_dev_run: false\n",
+      "    gpus: 1\n",
+      "    precision: 16\n",
+      "    max_epocs: 512\n",
+      "    terminate_on_nan: true\n",
+      "    weights_summary: true\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(OmegaConf.to_yaml(conf))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
    "metadata": {
     "scrolled": true
    },