update app.py, and evaluate_data.py to work with uploaded files

app.py

@@ -10,7 +10,7 @@ import json
 # from simcse import SimCSE # use for gpt
 from evaluate_data import store_sample_data, get_metrics_trf
 
-store_sample_data()
+# store_sample_data()
 
 
 
@@ -92,10 +92,12 @@ example = """
 My latest exclusive for The Hill : Conservative frustration over Republican efforts to force a House vote on reauthorizing the Export - Import Bank boiled over Wednesday during a contentious GOP meeting.
 
 """
-def find_orgs(sentence, choice):
-    return all_metrics
-radio_btn = gr.Radio(choices=['GPT', 'iSemantics'], value='iSemantics', label='Available models', show_label=True)
-textbox = gr.Textbox(label="Enter your text", placeholder=str(all_metrics), lines=8)
-
-iface = gr.Interface(fn=find_orgs, inputs=[textbox, radio_btn], outputs="text",  examples=[[example]])
+def find_orgs(uploaded_file):
+    uploaded_data = json.loads(uploaded_file)
+    return get_metrics_trf(uploaded_file)
+# radio_btn = gr.Radio(choices=['GPT', 'iSemantics'], value='iSemantics', label='Available models', show_label=True)
+# textbox = gr.Textbox(label="Enter your text", placeholder=str(all_metrics), lines=8)
+upload_btn = gr.UploadButton(label='Upload a json file.')
+
+iface = gr.Interface(fn=find_orgs, inputs=[upload_btn], outputs="text",  examples=[[example]])
 iface.launch(share=True)

evaluate_data.py

@@ -35,13 +35,6 @@ with open(feature_path, 'rb') as f:
 ner_model = AutoModelForTokenClassification.from_pretrained(checkpoint)
 
 
-tokenized_test = test.map(
-    tokenize_and_align_labels,
-    batched=True,
-    batch_size=None,
-    remove_columns=test.column_names[2:],
-    fn_kwargs={'tokenizer': tokenizer}
-)
 
 # tokenized_dataset.set_format('torch')
 
@@ -54,18 +47,28 @@ def collate_fn(data):
     
     return input_ids, token_type_ids, attention_mask, labels
 
-loader = torch.utils.data.DataLoader(tokenized_test, batch_size=16, collate_fn=collate_fn)
-device = torch.device('cuda:0') if torch.cuda.is_available() else torch.device('cpu')
 
 ner_model = ner_model.eval()
 
 
 
-def get_metrics_trf():
+def get_metrics_trf(data):
     print(device)
     
+    data = Dataset.from_dict(data)
+    
+    tokenized_data = data.map(
+        tokenize_and_align_labels,
+        batched=True,
+        batch_size=None,
+        remove_columns=data.column_names[2:],
+        fn_kwargs={'tokenizer': tokenizer}
+    )
+    
+    loader = torch.utils.data.DataLoader(tokenized_data, batch_size=16, collate_fn=collate_fn)
+    device = torch.device('cuda:0') if torch.cuda.is_available() else torch.device('cpu')
+    
     y_true, logits = [], []
-
     for input_ids, token_type_ids, attention_mask, labels in tqdm(loader):
         ner_model.to(device)
         with torch.no_grad():
@@ -110,10 +113,11 @@ def find_orgs(tokens, labels):
 
 
 
-def store_sample_data():
+def store_sample_data(data):
+    data = Dataset.from_dict(data)
     test_data = []
 
-    for sent in test:
+    for sent in data:
         labels = [ner_feature.feature.int2str(l) for l in sent['ner_tags']]
         # print(labels)
         sent_orgs = find_orgs(sent['tokens'], labels)