inital commit

.gitattributes

@@ -26,3 +26,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.data- filter=lfs diff=lfs merge=lfs -text
+*.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,56 @@
+import transformers
+
+import gradio as gr
+import tensorflow as tf
+
+from official.nlp import optimization  # to create AdamW optimizer
+
+MODEL_DIRECTORY = 'save/modelV1'
+PRETRAINED_MODEL_NAME = 'dbmdz/bert-base-german-cased'
+TOKENIZER = transformers.BertTokenizer.from_pretrained(PRETRAINED_MODEL_NAME)
+MAX_SEQUENCE_LENGTH = 256
+EPOCHS = 2
+OPTIMIZER = 'adamw'
+INIT_LR = 3e-5
+LOSS = tf.keras.losses.BinaryCrossentropy(from_logits=False)
+METRICS = tf.metrics.BinaryAccuracy()
+
+def compile_model(model):
+    steps_per_epoch = 10
+    num_train_steps = steps_per_epoch * EPOCHS
+    num_warmup_steps = int(0.1*num_train_steps)
+
+    optimizer = optimization.create_optimizer(
+        init_lr=INIT_LR,
+        num_train_steps=steps_per_epoch,
+        num_warmup_steps=num_warmup_steps,
+        optimizer_type=OPTIMIZER
+    )
+
+    model.compile(optimizer=optimizer, loss=LOSS, metrics=[METRICS])
+    return model
+
+hs_detection_model = tf.keras.models.load_model(MODEL_DIRECTORY, compile=False)  #tf.keras.models.load_model('save/kerasmodel/model.h5') #tf.saved_model.load('save/model') #tf.keras.models.load_model('save/model')
+compile_model(hs_detection_model)
+
+def encode(sentences):
+    return TOKENIZER.batch_encode_plus(
+        sentences,
+        max_length=MAX_SEQUENCE_LENGTH, # set the length of the sequences
+        add_special_tokens=True, # add [CLS] and [SEP] tokens
+        return_attention_mask=True,
+        return_token_type_ids=False, # not needed for this type of ML task
+        pad_to_max_length=True, # add 0 pad tokens to the sequences less than max_length
+        return_tensors='tf'
+    )
+
+def inference(sentence):
+    print(sentence)
+    encoded_sentence = encode([sentence])
+    print(encoded_sentence)
+    predicition = hs_detection_model.predict(encoded_sentence.values())
+    print(predicition)
+    return predicition
+
+iface = gr.Interface(fn=inference, inputs="text", outputs="text") #, live=True)
+iface.launch()

save/modelV1/keras_metadata.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5972b15e2cc31b72e34a938e1d9614cf6a9d2b1872abbc654453ab5f613693a2
+size 155363

save/modelV1/saved_model.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccb2afb1ffb3c867ae345e7d43af0e20f3f476c83cb86c838a07bf703c216fa4
+size 6651882

save/modelV1/variables/variables.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7940b2394fa52153c18be7b6007423ca74bb4d346b1f808c5ca94e4951090140
+size 1319391849