Address #5, thanks @Fquico1999

Signed-off-by: Jay Wang <[email protected]>

data-generation/README.md

@@ -36,4 +36,6 @@ If you change the dataset being visualized, should start by setting constants li
 
 ### Swap Dataset
 
-You should start by swapping out the loaded models in [`dataset_train`](https://github.com/poloclub/dodrio/blob/dd7a98bb26335cf960e7e508b683ff02d4a4a1ea/data-generation/dodrio-data-gen.py#L892), [`dataset_vali`](https://github.com/poloclub/dodrio/blob/dd7a98bb26335cf960e7e508b683ff02d4a4a1ea/data-generation/dodrio-data-gen.py#L891), and [`dataset_test`](https://github.com/poloclub/dodrio/blob/dd7a98bb26335cf960e7e508b683ff02d4a4a1ea/data-generation/dodrio-data-gen.py#L873). The primary change needed, so that the correct files can be generate across datasets are changes to the function [`collate_sst2`](https://github.com/poloclub/dodrio/blob/dd7a98bb26335cf960e7e508b683ff02d4a4a1ea/data-generation/dodrio-data-gen.py#L182) (which should be renamed as well). Changes to this function will affect how entries within the dataset are processed/tokenized. For non-classification NLP tasks, more significant changes will need to be made accross the file to support visualization in Dodrio. If you have any questions or problems, feel free to [open an issue](https://github.com/poloclub/dodrio/issues/new/choose).
+You should start by swapping out the loaded models in [`dataset_train`](https://github.com/poloclub/dodrio/blob/dd7a98bb26335cf960e7e508b683ff02d4a4a1ea/data-generation/dodrio-data-gen.py#L892), [`dataset_vali`](https://github.com/poloclub/dodrio/blob/dd7a98bb26335cf960e7e508b683ff02d4a4a1ea/data-generation/dodrio-data-gen.py#L891), and [`dataset_test`](https://github.com/poloclub/dodrio/blob/dd7a98bb26335cf960e7e508b683ff02d4a4a1ea/data-generation/dodrio-data-gen.py#L873). The primary change needed, so that the correct files can be generate across datasets are changes to the function [`collate_sst2`](https://github.com/poloclub/dodrio/blob/dd7a98bb26335cf960e7e508b683ff02d4a4a1ea/data-generation/dodrio-data-gen.py#L182) (which should be renamed as well). Changes to this function will affect how entries within the dataset are processed/tokenized. For non-classification NLP tasks, more significant changes will need to be made across the file to support visualization in Dodrio.
+For the demo, we choose the sample with `instanceID=1562`. You would need to change this variable across [different views](https://github.com/poloclub/dodrio/search?q=instanceID) for your custom dataset.
+If you have any questions or problems, feel free to [open an issue](https://github.com/poloclub/dodrio/issues/new/choose).

package.json

@@ -4,7 +4,7 @@
   "scripts": {
     "build": "rollup -c",
     "dev": "rollup -c -w",
-    "start": "sirv public"
+    "start": "sirv public --port 5005"
   },
   "devDependencies": {
     "@rollup/plugin-commonjs": "^14.0.0",
@@ -13,7 +13,7 @@
     "babel-eslint": "^10.1.0",
     "eslint": "^7.18.0",
     "eslint-plugin-svelte3": "^3.0.0",
-    "node-sass": "^5.0.0",
+    "node-sass": "^7.0.0",
     "rollup": "^2.3.4",
     "rollup-plugin-livereload": "^2.0.0",
     "rollup-plugin-svelte": "^6.0.0",