Upload app.py

app.py

@@ -12,10 +12,10 @@ loraModel = AutoPeftModelForSequenceClassification.from_pretrained("Intradiction
 tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased")
 
 tokenizer2 = AutoTokenizer.from_pretrained("microsoft/deberta-v3-xsmall")
-base_model = AutoModel.from_pretrained("microsoft/deberta-v3-xsmall")
-peft_model_id = "rajevan123/STS-Lora-Fine-Tuning-Capstone-Deberta-small"
-model = PeftModel.from_pretrained(base_model, peft_model_id)
-#merged_model = model.merge_and_unload()
+# base_model = AutoModel.from_pretrained("microsoft/deberta-v3-xsmall")
+# peft_model_id = "rajevan123/STS-Lora-Fine-Tuning-Capstone-Deberta-small"
+# model = PeftModel.from_pretrained(base_model, peft_model_id)
+# #merged_model = model.merge_and_unload()
 
 
 # Handle calls to DistilBERT------------------------------------------
@@ -53,15 +53,15 @@ def AlbertUntrained_fn(text1, text2):
 # Handle calls to Deberta--------------------------------------------
 DebertaUntrained_pipe = pipeline("text-classification", model="microsoft/deberta-v3-xsmall")
 DebertanoLORA_pipe = pipeline("text-classification", model="rajevan123/STS-Conventional-Fine-Tuning")
-DebertawithLORA_pipe = pipeline("text-classification",model=model, tokenizer=tokenizer2)
+#DebertawithLORA_pipe = pipeline("text-classification",model=model, tokenizer=tokenizer2)
 
 #STS models
 def DebertanoLORA_fn(text1, text2):
     return DebertanoLORA_pipe({'text': text1, 'text_pair': text2})
 
 def DebertawithLORA_fn(text1, text2):
-    return DebertawithLORA_pipe({'text': text1, 'text_pair': text2})
-    #return ("working2")
+    #return DebertawithLORA_pipe({'text': text1, 'text_pair': text2})
+    return ("working2")
 
 def DebertaUntrained_fn(text1, text2):
     return DebertaUntrained_pipe({'text': text1, 'text_pair': text2})
@@ -73,6 +73,50 @@ def displayMetricStatsUntrained():
     return "No statistics to display for untrained models"
 
 def displayMetricStatsText():
+    file_name = 'events.out.tfevents.distilbertSA-conventional.0'
+    event_acc = event_accumulator.EventAccumulator(file_name,
+    size_guidance={
+    event_accumulator.COMPRESSED_HISTOGRAMS: 500,
+    event_accumulator.IMAGES: 4,
+    event_accumulator.AUDIO: 4,
+    event_accumulator.SCALARS: 0,
+    event_accumulator.HISTOGRAMS: 1,
+})
+   
+    event_acc.Reload()
+    accuracy_data = event_acc.Scalars('eval/accuracy')
+    loss_data = event_acc.Scalars('eval/loss')
+    metrics = ''
+    for i in range(0, len(loss_data)):
+        metrics = metrics + 'Epoch Number: ' + str(i) + '\n'
+        metrics = metrics + 'Accuracy (%): ' + str(round(accuracy_data[i].value * 100, 3)) + '\n'
+        metrics = metrics + 'Loss (%): ' + str(round(loss_data[i].value * 100, 3)) + '\n\n'
+    
+    return metrics
+
+def displayMetricStatsTextTCLora():
+    file_name = 'events.out.tfevents.distilbertSA-LORA.0'
+    event_acc = event_accumulator.EventAccumulator(file_name,
+    size_guidance={
+    event_accumulator.COMPRESSED_HISTOGRAMS: 500,
+    event_accumulator.IMAGES: 4,
+    event_accumulator.AUDIO: 4,
+    event_accumulator.SCALARS: 0,
+    event_accumulator.HISTOGRAMS: 1,
+})
+   
+    event_acc.Reload()
+    accuracy_data = event_acc.Scalars('eval/accuracy')
+    loss_data = event_acc.Scalars('eval/loss')
+    metrics = ''
+    for i in range(0, len(loss_data)):
+        metrics = metrics + 'Epoch Number: ' + str(i) + '\n'
+        metrics = metrics + 'Accuracy (%): ' + str(round(accuracy_data[i].value * 100, 3)) + '\n'
+        metrics = metrics + 'Loss (%): ' + str(round(loss_data[i].value * 100, 3)) + '\n\n'
+    
+    return metrics
+
+def displayMetricStatsTextNLINoLora():
     file_name = 'events.out.tfevents.1701212945.784ae33ab242.985.0'
     event_acc = event_accumulator.EventAccumulator(file_name,
     size_guidance={
@@ -94,6 +138,71 @@ def displayMetricStatsText():
     
     return metrics
 
+def displayMetricStatsTextNLILora():
+    file_name = 'events.out.tfevents.1701212945.784ae33ab242.985.0'
+    event_acc = event_accumulator.EventAccumulator(file_name,
+    size_guidance={
+    event_accumulator.COMPRESSED_HISTOGRAMS: 500,
+    event_accumulator.IMAGES: 4,
+    event_accumulator.AUDIO: 4,
+    event_accumulator.SCALARS: 0,
+    event_accumulator.HISTOGRAMS: 1,
+})
+   
+    event_acc.Reload()
+    accuracy_data = event_acc.Scalars('eval/accuracy')
+    loss_data = event_acc.Scalars('eval/loss')
+    metrics = ''
+    for i in range(0, len(loss_data)):
+        metrics = metrics + 'Epoch Number: ' + str(i) + '\n'
+        metrics = metrics + 'Accuracy (%): ' + str(round(accuracy_data[i].value * 100, 3)) + '\n'
+        metrics = metrics + 'Loss (%): ' + str(round(loss_data[i].value * 100, 3)) + '\n\n'
+    
+    return metrics
+
+def displayMetricStatsTextSTSLora():
+    file_name = 'events.out.tfevents.STS-Lora.2'
+    event_acc = event_accumulator.EventAccumulator(file_name,
+    size_guidance={
+    event_accumulator.COMPRESSED_HISTOGRAMS: 500,
+    event_accumulator.IMAGES: 4,
+    event_accumulator.AUDIO: 4,
+    event_accumulator.SCALARS: 0,
+    event_accumulator.HISTOGRAMS: 1,
+})
+   
+    event_acc.Reload()
+    accuracy_data = event_acc.Scalars('eval/accuracy')
+    loss_data = event_acc.Scalars('eval/loss')
+    metrics = ''
+    for i in range(0, len(loss_data)):
+        metrics = metrics + 'Epoch Number: ' + str(i) + '\n'
+        metrics = metrics + 'Accuracy (%): ' + str(round(accuracy_data[i].value * 100, 3)) + '\n'
+        metrics = metrics + 'Loss (%): ' + str(round(loss_data[i].value * 100, 3)) + '\n\n'
+    
+    return metrics
+def displayMetricStatsTextSTSNoLora():
+    file_name = 'events.out.tfevents.STS-Conventional.0'
+    event_acc = event_accumulator.EventAccumulator(file_name,
+    size_guidance={
+    event_accumulator.COMPRESSED_HISTOGRAMS: 500,
+    event_accumulator.IMAGES: 4,
+    event_accumulator.AUDIO: 4,
+    event_accumulator.SCALARS: 0,
+    event_accumulator.HISTOGRAMS: 1,
+})
+   
+    event_acc.Reload()
+    accuracy_data = event_acc.Scalars('eval/accuracy')
+    loss_data = event_acc.Scalars('eval/loss')
+    metrics = ''
+    for i in range(0, len(loss_data)):
+        metrics = metrics + 'Epoch Number: ' + str(i) + '\n'
+        metrics = metrics + 'Accuracy (%): ' + str(round(accuracy_data[i].value * 100, 3)) + '\n'
+        metrics = metrics + 'Loss (%): ' + str(round(loss_data[i].value * 100, 3)) + '\n\n'
+    
+    return metrics
+
 def displayMetricStatsGraph():
    file_name = 'events.out.tfevents.1701212945.784ae33ab242.985.0'
    event_acc = event_accumulator.EventAccumulator(file_name,
@@ -199,7 +308,7 @@ with gr.Blocks(
         btn.click(fn=distilBERTwithLORA_fn, inputs=inp, outputs=TextClassOut2)
         btnTextClassStats.click(fn=displayMetricStatsUntrained, outputs=TextClassUntrained)
         btnTextClassStats.click(fn=displayMetricStatsText, outputs=TextClassNoLoraStats)
-        btnTextClassStats.click(fn=DebertawithLORA_fn, outputs=TextClassLoraStats) #to be changed
+        btnTextClassStats.click(fn=displayMetricStatsTextTCLora, outputs=TextClassLoraStats) #to be changed
 
     with gr.Tab("Natural Language Inferencing"):
          with gr.Row():
@@ -313,8 +422,8 @@ with gr.Blocks(
          sts_btn.click(fn=DebertanoLORA_fn, inputs=[sts_p1,sts_p2], outputs=sts_out1)
          sts_btn.click(fn=DebertawithLORA_fn, inputs=[sts_p1,sts_p2], outputs=sts_out2)
          btnSTSStats.click(fn=displayMetricStatsUntrained, outputs=STSUntrained)
-         #btnSTSStats.click(fn=displayMetricStatsUntrained, outputs=STSNoLoraStats)
-         #btnSTSStats.click(fn=displayMetricStatsUntrained, outputs=STSLoraStats)
+         btnSTSStats.click(fn=displayMetricStatsTextSTSNoLora, outputs=STSNoLoraStats)
+         btnSTSStats.click(fn=displayMetricStatsTextSTSLora, outputs=STSLoraStats)
 
     with gr.Tab("More informatioen"):
         gr.Markdown("stuff to add")