initial app

.gitignore

@@ -0,0 +1 @@
+__pycache__

app.py

@@ -11,15 +11,52 @@ INTRODUCTION_TEXT = """
 
 📈 We evaluate models based on 4 datasets,
 
-1. CrossRef Melayu related DOI, https://huggingface.co/datasets/mesolitica/malaysian-ultrachat/resolve/main/ultrachat-crossref-melayu-malay.jsonl
+1. Research paper keyword `melayu` using Crossref, https://huggingface.co/datasets/mesolitica/malaysian-ultrachat/resolve/main/ultrachat-crossref-melayu-malay.jsonl
 2. Epenerbitan, https://huggingface.co/datasets/mesolitica/malaysian-ultrachat/resolve/main/ultrachat-epenerbitan-malay.jsonl
 3. gov.my PDF files, https://huggingface.co/datasets/mesolitica/malaysian-ultrachat/resolve/main/ultrachat-gov.my.jsonl
-4. lom.agc.gov.my PDF files, https://huggingface.co/datasets/mesolitica/malaysian-ultrachat/resolve/main/ultrachat-lom-agc.jsonl
 """
 
+close_source = [
+    {
+        'model': 'OpenAI ADA-002',
+        'Crossref Melayu top-1': 0.3155939351340496,
+        'Crossref Melayu top-3': 0.5120996083944171,
+        'Crossref Melayu top-5': 0.5878100210864544,
+        'Crossref Melayu top-10': 0.6721558389396526,
+    }
+]
+
+open_source = [
+    {
+        'model': '[llama2-embedding-600m-8k](https://huggingface.co/mesolitica/llama2-embedding-600m-8k)',
+        'Crossref Melayu top-1': 0.09549151521237072,
+        'Crossref Melayu top-3': 0.1834521538307059,
+        'Crossref Melayu top-5': 0.23375840947886334,
+        'Crossref Melayu top-10': 0.3098704689225826,
+        'lom.agc.gov.my top-1': 0.05215334826985312,
+        'lom.agc.gov.my top-3': 0.09932785660941,
+        'lom.agc.gov.my top-5': 0.12969878018421707,
+        'lom.agc.gov.my top-10': 0.1797361214836943,
+    },
+    {
+        'model': '[llama2-embedding-1b-8k](https://huggingface.co/mesolitica/llama2-embedding-1b-8k)',
+        'Crossref Melayu top-1': 0.06777788934631991,
+        'Crossref Melayu top-3': 0.142584596847073,
+        'Crossref Melayu top-5': 0.18817150316296816,
+        'Crossref Melayu top-10': 0.25715433276433375,
+        'lom.agc.gov.my top-1': 0.06870799103808813,
+        'lom.agc.gov.my top-3': 0.1343042071197411,
+        'lom.agc.gov.my top-5': 0.1717699775952203,
+        'lom.agc.gov.my top-10': 0.23089370176748816,
+    },
+]
+
+data = pd.DataFrame(close_source + open_source)
+
 demo = gr.Blocks(css=custom_css)
 with demo:
     gr.HTML(TITLE)
     gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
+    gr.DataFrame(data, datatype = 'markdown')
 
 demo.launch()