stop encoder being read into memory on each call to /vectorize

docker/app.py

@@ -9,70 +9,76 @@
 from LASER.source.lib.text_processing import Token, BPEfastApply
 from LASER.source.embed import *
 
-app = Flask(__name__)
-app.config['JSON_AS_ASCII'] = False
 
-
-@app.route("/")
-def root():
-    print("/")
-    html = "<h3>Hello {name}!</h3>" \
-           "<b>Hostname:</b> {hostname}<br/>"
-    return html.format(name=os.getenv("LASER", "world"), hostname=socket.gethostname())
-
-
-@app.route("/vectorize")
-def vectorize():
-    content = request.args.get('q')
-    lang = request.args.get('lang')
-    embedding = ''
-    if lang is None or not lang:
-        lang = "en"
+def create_app():
+    app = Flask(__name__)
+    app.config['JSON_AS_ASCII'] = False
     # encoder
     model_dir = Path(__file__).parent / "LASER" / "models"
     encoder_path = model_dir / "bilstm.93langs.2018-12-26.pt"
-    bpe_codes_path = model_dir / "93langs.fcodes"
+    app.bpe_codes_path = model_dir / "93langs.fcodes"
     print(f' - Encoder: loading {encoder_path}')
     encoder = SentenceEncoder(encoder_path,
                               max_sentences=None,
                               max_tokens=12000,
                               sort_kind='mergesort',
                               cpu=True)
-    with tempfile.TemporaryDirectory() as tmp:
-        tmpdir = Path(tmp)
-        ifname = tmpdir / "content.txt"
-        bpe_fname = tmpdir / 'bpe'
-        bpe_oname = tmpdir / 'out.raw'
-        with ifname.open("w") as f:
-            f.write(content)
-        if lang != '--':
-            tok_fname = tmpdir / "tok"
-            Token(str(ifname),
-                  str(tok_fname),
-                  lang=lang,
-                  romanize=True if lang == 'el' else False,
-                  lower_case=True,
-                  gzip=False,
-                  verbose=True,
-                  over_write=False)
-            ifname = tok_fname
-        BPEfastApply(str(ifname),
+    app.encoder = encoder
+
+    @app.route("/")
+    def root():
+        print("/")
+        html = "<h3>Hello {name}!</h3>" \
+           "<b>Hostname:</b> {hostname}<br/>"
+        return html.format(name=os.getenv("LASER", "world"), hostname=socket.gethostname())
+
+    @app.route("/vectorize")
+    def vectorize():
+        content = request.args.get('q')
+        lang = request.args.get('lang')
+        embedding = ''
+        if lang is None or not lang:
+            lang = "en"
+
+        with tempfile.TemporaryDirectory() as tmp:
+            tmpdir = Path(tmp)
+            ifname = tmpdir / "content.txt"
+            bpe_fname = tmpdir / 'bpe'
+            bpe_oname = tmpdir / 'out.raw'
+            with ifname.open("w") as f:
+                f.write(content)
+            if lang != '--':
+                tok_fname = tmpdir / "tok"
+                Token(str(ifname),
+                    str(tok_fname),
+                    lang=lang,
+                    romanize=True if lang == 'el' else False,
+                    lower_case=True,
+                    gzip=False,
+                    verbose=True,
+                    over_write=False)
+                ifname = tok_fname
+            BPEfastApply(str(ifname),
                      str(bpe_fname),
-                     str(bpe_codes_path),
+                     str(app.bpe_codes_path),
                      verbose=True, over_write=False)
-        ifname = bpe_fname
-        EncodeFile(encoder,
+            ifname = bpe_fname
+            EncodeFile(app.encoder,
                    str(ifname),
                    str(bpe_oname),
                    verbose=True,
                    over_write=False,
                    buffer_size=10000)
-        dim = 1024
-        X = np.fromfile(str(bpe_oname), dtype=np.float32, count=-1)
-        X.resize(X.shape[0] // dim, dim)
-        embedding = X
-    body = {'content': content, 'embedding': embedding.tolist()}
-    return jsonify(body)
+            dim = 1024
+            X = np.fromfile(str(bpe_oname), dtype=np.float32, count=-1)
+            X.resize(X.shape[0] // dim, dim)
+            embedding = X
+        body = {'content': content, 'embedding': embedding.tolist()}
+        return jsonify(body)
+
+    return app
+
 
 if __name__ == "__main__":
+    app = create_app()
     app.run(debug=True, port=80, host='0.0.0.0')