Upload with huggingface_hub

.gitattributes

@@ -1,34 +1,22 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+index/**/* filter=lfs diff=lfs merge=lfs -text
+data/data-00000-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00010-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00016-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00006-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00005-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00019-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00013-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00009-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00003-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00004-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00008-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00014-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00015-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00001-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00020-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00011-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00017-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00007-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00018-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00002-of-00021.arrow filter=lfs diff=lfs merge=lfs -text
+data/data-00012-of-00021.arrow filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,12 +1,13 @@
 ---
-title: Code Search
-emoji: 📊
-colorFrom: red
-colorTo: pink
+title: Code search
+emoji: 🐠
+colorFrom: blue
+colorTo: blue
 sdk: gradio
-sdk_version: 3.19.1
+sdk_version: 3.12.0
 app_file: app.py
 pinned: false
+license: apache-2.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,85 @@
+import gradio as gr
+from datasets import load_from_disk
+from pyserini.search.lucene import LuceneSearcher
+
+searcher = LuceneSearcher("index")
+ds = load_from_disk("data")
+NUM_PAGES = 10 # STATIC. THIS CAN'T CHANGE BECAUSE GRADIO CAN'T DYNAMICALLY CREATE COMPONENTS. 
+RESULTS_PER_PAGE = 5 
+
+TEXT_FIELD = "content"
+METADATA_FIELD = "docid"
+
+def result_html(result, meta):
+    return (
+    f"<div style=\"color:#2a5cb3;font-weight: 500\"><u>{meta}</u></div><br>"
+    f"<div><details><summary>{result[:250]}...</summary><p>{result[250:]}</p></details></div><br><hr><br>"
+    )
+
+def format_results(results):
+    return "\n".join([result_html(result, meta) for result,meta in zip(results[TEXT_FIELD], results[METADATA_FIELD])])
+    
+def page_0(query):
+    hits = searcher.search(query, k=NUM_PAGES*RESULTS_PER_PAGE)
+    ix = [int(hit.docid) for hit in hits]
+    results = ds.select(ix).shard(num_shards=NUM_PAGES, index=0, contiguous=True) # no need to shard. split ix in batches instead. (would make sense if results was cacheable)
+    results = format_results(results)
+    return results, [ix], gr.update(visible=True)
+
+def page_i(i, ix):
+    ix = ix[0]
+    results = ds.select(ix).shard(num_shards=NUM_PAGES, index=i, contiguous=True)
+    results = format_results(results)
+    return results, [ix]
+    
+with gr.Blocks(css="#b {min-width:15px;background:transparent;border:white;box-shadow:none;}") as demo: #
+    with gr.Row():
+        gr.Markdown(value="""## <p style="text-align: center;"> Code search </p>""")  
+    with gr.Row():
+        with gr.Column(scale=1):
+            result_list = gr.Dataframe(type="array", visible=False, col_count=1)      
+        with gr.Column(scale=13):
+            query = gr.Textbox(lines=1, max_lines=1, placeholder="Search…", label="")
+        with gr.Column(scale=1):
+            with gr.Row(scale=1):
+                pass
+            with gr.Row(scale=1):    
+                submit_btn = gr.Button("🔍", elem_id="b").style(full_width=False)
+            with gr.Row(scale=1):
+                pass
+                
+    with gr.Row():
+        with gr.Column(scale=1):
+            pass
+        with gr.Column(scale=13):
+            c = gr.HTML(label="Results")
+            with gr.Row(visible=False) as pagination:
+                # left = gr.Button(value="◀", elem_id="b", visible=False).style(full_width=True)
+                page_1 = gr.Button(value="1", elem_id="b").style(full_width=True)
+                page_2 = gr.Button(value="2", elem_id="b").style(full_width=True)
+                page_3 = gr.Button(value="3", elem_id="b").style(full_width=True)
+                page_4 = gr.Button(value="4", elem_id="b").style(full_width=True)
+                page_5 = gr.Button(value="5", elem_id="b").style(full_width=True)
+                page_6 = gr.Button(value="6", elem_id="b").style(full_width=True)
+                page_7 = gr.Button(value="7", elem_id="b").style(full_width=True)
+                page_8 = gr.Button(value="8", elem_id="b").style(full_width=True)
+                page_9 = gr.Button(value="9", elem_id="b").style(full_width=True)
+                page_10 = gr.Button(value="10", elem_id="b").style(full_width=True)
+                # right = gr.Button(value="▶", elem_id="b", visible=False).style(full_width=True)
+        with gr.Column(scale=1):
+            pass
+    query.submit(fn=page_0, inputs=[query], outputs=[c, result_list, pagination])
+    submit_btn.click(page_0, inputs=[query], outputs=[c, result_list, pagination])
+    with gr.Box(visible=False):
+        nums = [gr.Number(i, visible=False, precision=0) for i in range(NUM_PAGES)]
+    page_1.click(fn=page_i, inputs=[nums[0], result_list], outputs=[c, result_list])
+    page_2.click(fn=page_i, inputs=[nums[1], result_list], outputs=[c, result_list])
+    page_3.click(fn=page_i, inputs=[nums[2], result_list], outputs=[c, result_list])
+    page_4.click(fn=page_i, inputs=[nums[3], result_list], outputs=[c, result_list])
+    page_5.click(fn=page_i, inputs=[nums[4], result_list], outputs=[c, result_list])
+    page_6.click(fn=page_i, inputs=[nums[5], result_list], outputs=[c, result_list])
+    page_7.click(fn=page_i, inputs=[nums[6], result_list], outputs=[c, result_list])
+    page_8.click(fn=page_i, inputs=[nums[7], result_list], outputs=[c, result_list])
+    page_9.click(fn=page_i, inputs=[nums[8], result_list], outputs=[c, result_list])
+    page_10.click(fn=page_i, inputs=[nums[9], result_list], outputs=[c, result_list])
+demo.launch(enable_queue=True, debug=True)

data/data-00000-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e200bbda830ddb4e3587f85d6b545b12c07260f3c7767b9032076d4762675e2
+size 145373736

data/data-00001-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db20ca9a832d31e95f483bfe368b7ee4bb8c4364744cefbcf4df19843b6a00df
+size 142136480

data/data-00002-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95898c5154de5d9c3273dea6f448c6b83a0bdce2af310edee32a7b516ccd2d6b
+size 146489736

data/data-00003-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e68159c6759ddf4a22a903521d5b3f7c364221c54c013ff7c71d1a9d8a50584c
+size 144372320

data/data-00004-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3b9b642da8a57ad54fc8e6703b07f56ef6106dc9db1861aaad0aa1605fa0b2c
+size 143888944

data/data-00005-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2da425ec3e1b327114fdf59f93876cb8fc230f05e3b2ea035c3289841d24bcf3
+size 145977664

data/data-00006-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86c311615d0e275ac342fd2a4c455b9bbf24398d0f50fa914b677c5b44dd84ab
+size 148958672

data/data-00007-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a0ea29b1f6dd74ae3a4dd51bcf9620af4208959a1e3c613fbaabb8a34604860
+size 122847584

data/data-00008-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70426a18f9cbf99b83386226735a9506ec10398e1602b4e7d989bc42d9fd42b2
+size 126813336

data/data-00009-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a789b10a680501d8940b623fc02aa83de49bcb74c7c6a352abb5762ed38ad870
+size 123187024

data/data-00010-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:058175c967ca642193b41f66c760142e4b8e25d831c851d992558d4078085948
+size 124968032

data/data-00011-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2617fcc9e53b94f2871bbf5bb7e1e8f8cc5b03cb8728001fa2bfa2c43d8a81c9
+size 127244208

data/data-00012-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76ac5503c0c53249c1b5ddcd732fa5757de1a2390afedd5491b2dc45a179ddd9
+size 129710432

data/data-00013-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a7f05d892987c5060d1b41f346b941ad2882e41a503da9d058451dd9b8affa6
+size 128167184

data/data-00014-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:530578064d3d1083c557ff7fa7a0f0e8e32b867de09318306e16fed52a159f3e
+size 99745824

data/data-00015-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e47133749980a2d974db2d58f28c070fd8981a02e5f5fcbcfe688e15811c21a2
+size 108449752

data/data-00016-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6970d853767a5b6d859f8495bc40708cdf9aeb4132c08e78302bfa64b0c33df
+size 104164320

data/data-00017-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3421eab94a12425c319dabd1599c6e165255b36410989aba05b5113aeba5138
+size 103913216

data/data-00018-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85eff8c66076127b7a2f908930de15edda3be143b2338702a912bb0bb7e0244c
+size 97751296

data/data-00019-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a611d877b9de895ec1f9b4117d4d5c708593be6ef445da4bec4dfbeff2b5218
+size 97219672

data/data-00020-of-00021.arrow

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d12fea67d0dce6bba733cd336081a09c5a15b69800df4698332c984caf166e3a
+size 98667032

data/dataset_info.json

@@ -0,0 +1,16 @@
+{
+  "citation": "",
+  "description": "",
+  "features": {
+    "content": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "docid": {
+      "dtype": "int64",
+      "_type": "Value"
+    }
+  },
+  "homepage": "",
+  "license": ""
+}

data/state.json

@@ -0,0 +1,73 @@
+{
+  "_data_files": [
+    {
+      "filename": "data-00000-of-00021.arrow"
+    },
+    {
+      "filename": "data-00001-of-00021.arrow"
+    },
+    {
+      "filename": "data-00002-of-00021.arrow"
+    },
+    {
+      "filename": "data-00003-of-00021.arrow"
+    },
+    {
+      "filename": "data-00004-of-00021.arrow"
+    },
+    {
+      "filename": "data-00005-of-00021.arrow"
+    },
+    {
+      "filename": "data-00006-of-00021.arrow"
+    },
+    {
+      "filename": "data-00007-of-00021.arrow"
+    },
+    {
+      "filename": "data-00008-of-00021.arrow"
+    },
+    {
+      "filename": "data-00009-of-00021.arrow"
+    },
+    {
+      "filename": "data-00010-of-00021.arrow"
+    },
+    {
+      "filename": "data-00011-of-00021.arrow"
+    },
+    {
+      "filename": "data-00012-of-00021.arrow"
+    },
+    {
+      "filename": "data-00013-of-00021.arrow"
+    },
+    {
+      "filename": "data-00014-of-00021.arrow"
+    },
+    {
+      "filename": "data-00015-of-00021.arrow"
+    },
+    {
+      "filename": "data-00016-of-00021.arrow"
+    },
+    {
+      "filename": "data-00017-of-00021.arrow"
+    },
+    {
+      "filename": "data-00018-of-00021.arrow"
+    },
+    {
+      "filename": "data-00019-of-00021.arrow"
+    },
+    {
+      "filename": "data-00020-of-00021.arrow"
+    }
+  ],
+  "_fingerprint": "b4bd4106f941687d",
+  "_format_columns": null,
+  "_format_kwargs": {},
+  "_format_type": null,
+  "_output_all_columns": false,
+  "_split": null
+}

index/_x.fdt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c7d061745915f007c93eda53ed5819761ccb0074d95581dbd706f156a413f42
+size 3998795

index/_x_Lucene90_0.doc

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbedd0d3108bf9412bee31f35e70cf877d2836c4fb512be18118accf1e8c7d64
+size 1496033008

index/_x_Lucene90_0.dvd

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5497a5551af1944e772bf4fc7c6d3eb2c6a4394c5389fb4f881ada90167743c2
+size 4383858

index/_x_Lucene90_0.pos

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9797a15a462df4033bcba66c2525189d483673e667f90c7b3c4c2cda8ba63e36
+size 7582291303

index/_x_Lucene90_0.tim

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09f85b03a6ed0c57921e3f57642d33951fe730da67b529c1f23cfd3e4c1ee060
+size 101304744

index/_x_Lucene90_0.tip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8043c70e5342c3ec652c379e4599f5930dba1ed57c0947c9515e934a112e9bfb
+size 3361566

packages.txt

@@ -0,0 +1 @@
+openjdk-11-jdk

requirements.txt

@@ -0,0 +1,4 @@
+pyserini
+datasets
+faiss-cpu
+torch