new classifier

milestone-3.py

@@ -25,7 +25,7 @@ dataset = filtered_dataset.shuffle(seed=42).select(range(20))
 dataset = dataset.sort("patent_number")
 
 # Create pipeline using model trainned on Colab
-model = torch.load("patent_classifier.pt", map_location=torch.device("cpu"))
+model = torch.load("patent_classifier_v2.pt", map_location=torch.device("cpu"))
 tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased")
 classifier = pipeline("text-classification", model=model, tokenizer=tokenizer)
 
@@ -74,4 +74,3 @@ if submitted:
             pred, score
         )
     )
-    check = st.markdown("Actual Label: **{}**.".format(label))

milestone_2.py

@@ -1,26 +0,0 @@
-import streamlit as st
-from transformers import (AutoTokenizer, TFAutoModelForSequenceClassification,
-                          pipeline)
-
-st.title("CS-GY-6613 Project Milestone 2")
-model_choices = (
-    "distilbert-base-uncased-finetuned-sst-2-english",
-    "j-hartmann/emotion-english-distilroberta-base",
-    "joeddav/distilbert-base-uncased-go-emotions-student",
-)
-
-with st.form("Input Form"):
-    text = st.text_area("Write your text here:", "CS-GY-6613 is a great course!")
-    model_name = st.selectbox("Select a model:", model_choices)
-    submitted = st.form_submit_button("Submit")
-
-if submitted:
-    model = TFAutoModelForSequenceClassification.from_pretrained(model_name)
-    tokenizer = AutoTokenizer.from_pretrained(model_name)
-    classifier = pipeline("sentiment-analysis", model=model, tokenizer=tokenizer)
-    res = classifier(text)
-    label = res[0]["label"].upper()
-    score = res[0]["score"]
-    st.markdown(
-        f"This text was classified as **{label}** with a confidence score of **{score}**."
-    )

patent_classification_v2.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fbbdc470f673703431aa31cc7451af0d0608df3bd6e7006ab32866803f4eece
+size 267882633