Improve data handling API and various other minor changes (#30)

asreview · Nov 22, 2021 · 799d676 · 799d676
1 parent fddacdd
commit 799d676
Show file tree

Hide file tree

Showing 6 changed files with 87 additions and 350 deletions.
diff --git a/README.md b/README.md
@@ -35,13 +35,8 @@ asreview semantic_clustering --help
 Other options are:
 
 ```shell
-asreview semantic_clustering -f <input.csv or url> -o <output.csv>
-asreview semantic_clustering --filepath <input.csv or url> --output <output.csv>
-```
-
-```shell
-asreview semantic_clustering -t -o <output.csv>
-asreview semantic_clustering --testfile --output <output.csv>
+asreview semantic_clustering -f <input> -o <output.csv>
+asreview semantic_clustering --filepath <input> --output <output.csv>
 ```
 
 ```shell
@@ -78,32 +73,23 @@ Using `-f` will process a file and store the results in the file specified in
 
 Semantic_clustering uses an [`ASReviewData`
 object](https://asreview.readthedocs.io/en/latest/API/generated/asreview.data.ASReviewData.html#asreview.data.ASReviewData),
-and can handle either a file or url:
+and can handle files, urls and benchmark sets:
 
 ```shell
-asreview semantic_clustering -f "https://raw.githubusercontent.com/asreview/systematic-review-datasets/master/datasets/van_de_Schoot_2017/output/van_de_Schoot_2017.csv" -o output.csv
+asreview semantic_clustering -f benchmark:van_de_schoot_2017 -o output.csv
 asreview semantic_clustering -f van_de_Schoot_2017.csv -o output.csv
 ```
 
 If an output file is not specified, `output.csv` is used as output file name.
 
-### Test file
-```shell
-asreview semantic_clustering -t -o <output_file.csv>
-```
-
-Using `-t` instead of `-f` uses the
-[`van_de_Schoot_2017`](https://asreview.readthedocs.io/en/latest/intro/datasets.html#featured-datasets)
-dataset as input file. This way, the plugin can easily be tested.
-
 ### Transformer
 Semantic Clustering uses the
 [`allenai/scibert_scivocab_uncased`](https://github.com/allenai/scibert)
 transformer model as default setting. Using the `--transformer <model>` option,
 another model can be selected for use instead:
 
 ```shell
-asreview semantic_clustering -t -o <output_file.csv> --transformer bert-base-uncased
+asreview semantic_clustering -f benchmark:van_de_schoot_2017 -o <output_file.csv> --transformer bert-base-uncased
 ```
 
 Any pretrained model will work.

diff --git a/asreviewcontrib/semantic_clustering/clustering.py b/asreviewcontrib/semantic_clustering/clustering.py
diff --git a/asreviewcontrib/semantic_clustering/dim_reduct.py b/asreviewcontrib/semantic_clustering/dim_reduct.py
diff --git a/asreviewcontrib/semantic_clustering/interactive.py b/asreviewcontrib/semantic_clustering/interactive.py
@@ -20,12 +20,24 @@ def run_app(filepath):
     # Read as STR for discrete colormap
     df['cluster_id'] = df['cluster_id'].astype(str)
 
+    # Set 'cluster_id' to max if 'inclusion_label' == 1
+    for row in df.itertuples():
+        if row.included == 1:
+            df.at[row.Index, 'cluster_id'] = 'included'
+
     # Show main figure
-    fig = px.scatter(df, x="x", y="y", color="cluster_id",
-                     color_discrete_sequence=px.colors.qualitative.Set1)
+    fig = px.scatter(df,
+                     x="x",
+                     y="y",
+                     color="cluster_id",
+                     color_discrete_sequence=px.colors.qualitative.Light24)
     fig.update_layout(dragmode="pan")
+    fig.update_layout(legend={'traceorder': 'normal'},
+                      plot_bgcolor='rgba(0,0,0,0.35)',
+                      height=400,)
     fig.update_layout(xaxis=dict(showticklabels=False, title=""),
                       yaxis=dict(showticklabels=False, ticks="", title=""))
+
     config = dict(
         {'scrollZoom': True,
             'displayModeBar': False,
@@ -46,8 +58,11 @@ def run_app(filepath):
 
             # Main semantic cluster graph
             html.Div([
-                dcc.Graph(figure=fig, id="cluster-div", config=config)
-            ], className="six columns"),
+                dcc.Graph(figure=fig, id="cluster-div", config=config,
+                          style={'width': '100%',
+                                 'height': '100%'
+                                 },)
+            ], className="six columns", style={'height': '80%'}),
 
             # Div for abstract window
             html.Div([
@@ -56,13 +71,21 @@ def run_app(filepath):
                     readOnly=True,
                     placeholder='Enter a value...',
                     value='This is a TextArea component',
-                    style={'width': '100%', 'height': '300px'},
+                    style={'width': '98%', 'height': '389px'},
                     id="abstract-div"
                 )
             ], className="six columns"),
 
-        ], className="row"),
-    ])
+        ], className="row", style={'height': '100%'}),
+    ], style={'backgroundColor': 'rgba(0,0,0,0.1)',
+              'position': 'fixed',
+              'width': '100%',
+              'height': '100%',
+              'top': '0',
+              'left': '0',
+              'z-index': '10',
+              'padding': '10px'
+              })
 
     # Allow global css - use chriddyp's time-tested external css
     app.css.config.serve_locally = False