Duplicate from asimokby/cv-parser-huggingface

Co-authored-by: Asem <asimokby@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,27 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bin.* filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.xz 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

Main.py

@@ -0,0 +1,23 @@
+from ResumeReader import ResumeReader
+from ResumeParser import ResumeParser
+from Models import Models
+import json
+import os
+
+
+class Main:
+    def __init__(self):
+        models = Models()
+        ner, ner_dates, zero_shot_classifier, tagger = models.load_trained_models()
+        self.reader = ResumeReader()
+        self.parser = ResumeParser(ner, ner_dates, zero_shot_classifier, tagger) 
+
+    def parse_cv(self, file_path):
+        resume_lines = self.reader.read_file(file_path)
+        output = self.parser.parse(resume_lines)
+        return output
+        
+    def save_parse_as_json(self, dict, file_name):
+        print("Saving the parse...")
+        with open(file_name, 'w', encoding="utf-8") as f:
+            json.dump(dict, f, indent=4, default=str, ensure_ascii=False) 

Models.py

@@ -0,0 +1,58 @@
+from transformers import AutoTokenizer, AutoModelForTokenClassification, AutoModelForSequenceClassification
+from transformers import pipeline
+from flair.data import Sentence
+from flair.models import SequenceTagger
+import pickle
+
+
+
+class Models:
+
+    def pickle_it(self, obj, file_name):
+        with open(f'{file_name}.pickle', 'wb') as f:
+            pickle.dump(obj, f)
+
+    def unpickle_it(self, file_name):
+        with open(f'{file_name}.pickle', 'rb') as f:
+            return pickle.load(f)
+
+    def load_trained_models(self, pickle=False):
+        #NER (dates)
+        tokenizer = AutoTokenizer.from_pretrained("Jean-Baptiste/camembert-ner-with-dates")
+        model = AutoModelForTokenClassification.from_pretrained("Jean-Baptiste/camembert-ner-with-dates")
+        self.ner_dates = pipeline('ner', model=model, tokenizer=tokenizer, aggregation_strategy="simple")
+
+        #Zero Shot Classification
+        # self.zero_shot_classifier = pipeline("zero-shot-classification", model='facebook/bart-large-mnli')
+        self.zero_shot_classifier = pipeline("zero-shot-classification", model='valhalla/distilbart-mnli-12-6')
+
+        # Ner
+        tokenizer = AutoTokenizer.from_pretrained("dslim/bert-base-NER")
+        model = AutoModelForTokenClassification.from_pretrained("dslim/bert-base-NER")
+        self.ner = pipeline('ner', model=model, tokenizer=tokenizer, grouped_entities=True)
+
+        # Pos Tagging
+        self.tagger = SequenceTagger.load("flair/pos-english-fast")
+
+
+        if pickle:
+            self.pickle_models()
+        
+        return self.ner, self.ner_dates, self.zero_shot_classifier, self.tagger
+    
+    def pickle_models(self):
+        self.pickle_it(self.ner, "ner")
+        self.pickle_it(self.zero_shot_classifier, "zero_shot_classifier_6")
+        self.pickle_it(self.ner_dates, "ner_dates")
+        self.pickle_it(self.tagger, "pos_tagger_fast")
+
+
+    def load_pickled_models(self):
+        ner_dates = self.unpickle_it('ner_dates')
+        ner = self.unpickle_it('ner')
+        zero_shot_classifier = self.unpickle_it('zero_shot_classifier_6')
+        tagger = self.unpickle_it("pos_tagger_fast")
+        return ner_dates, ner, zero_shot_classifier, tagger
+    
+    def get_flair_sentence(self, sent):
+        return Sentence(sent)

README.md

@@ -0,0 +1,47 @@
+---
+title: Cv Parser
+emoji: 💩
+colorFrom: green
+colorTo: red
+sdk: gradio
+app_file: app.py
+pinned: false
+license: mit
+duplicated_from: asimokby/cv-parser-huggingface
+---
+
+# Configuration
+
+`title`: _string_
+Display title for the Space
+
+`emoji`: _string_
+Space emoji (emoji-only character allowed)
+
+`colorFrom`: _string_
+Color for Thumbnail gradient (red, yellow, green, blue, indigo, purple, pink, gray)
+
+`colorTo`: _string_
+Color for Thumbnail gradient (red, yellow, green, blue, indigo, purple, pink, gray)
+
+`sdk`: _string_
+Can be either `gradio`, `streamlit`, or `static`
+
+`sdk_version` : _string_
+Only applicable for `streamlit` SDK.  
+See [doc](https://hf.co/docs/hub/spaces) for more info on supported versions.
+
+`app_file`: _string_
+Path to your main application file (which contains either `gradio` or `streamlit` Python code, or `static` html code).
+Path is relative to the root of the repository.
+
+`models`: _List[string]_
+HF model IDs (like "gpt2" or "deepset/roberta-base-squad2") used in the Space.
+Will be parsed automatically from your code if not specified here.
+
+`datasets`: _List[string]_
+HF dataset IDs (like "common_voice" or "oscar-corpus/OSCAR-2109") used in the Space.
+Will be parsed automatically from your code if not specified here.
+
+`pinned`: _boolean_
+Whether the Space stays on top of your list.

ResumeParser.py

@@ -0,0 +1,241 @@
+from Models import Models
+from ResumeSegmenter import ResumeSegmenter
+from datetime import datetime
+from dateutil import parser
+import re
+from string import punctuation
+
+class ResumeParser:
+    def __init__(self, ner, ner_dates, zero_shot_classifier, tagger):
+        self.models = Models()
+        self.segmenter = ResumeSegmenter(zero_shot_classifier)
+        self.ner, self.ner_dates, self.zero_shot_classifier, self.tagger = ner, ner_dates, zero_shot_classifier, tagger 
+        self.parsed_cv = {}
+
+    def parse(self, resume_lines):
+        resume_segments = self.segmenter.segment(resume_lines)
+        print("Parsing the Resume...")
+        for segment_name in resume_segments:
+            if segment_name == "contact_info":
+                contact_info = resume_segments[segment_name]
+                self.parse_contact_info(contact_info)
+            elif segment_name == "work_and_employment":
+                resume_segment = resume_segments[segment_name]
+                self.parse_job_history(resume_segment)
+        return self.parsed_cv
+
+
+    def parse_contact_info(self, contact_info):
+        contact_info_dict = {}
+        name = self.find_person_name(contact_info)
+        email = self.find_contact_email(contact_info)
+        self.parsed_cv['Name'] = name
+        contact_info_dict["Email"] = email
+        self.parsed_cv['Contact Info'] = contact_info_dict
+
+    def find_person_name(self, items):
+        class_score = []
+        splitter = re.compile(r'[{}]+'.format(re.escape(punctuation.replace("&", "") )))
+        classes = ["person name", "address", "email", "title"]
+        for item in items: 
+            elements = splitter.split(item)
+            for element in elements:
+                element = ''.join(i for i in element.strip() if not i.isdigit())
+                if not len(element.strip().split()) > 1: continue
+                out = self.zero_shot_classifier(element, classes)
+                highest = sorted(zip(out["labels"], out["scores"]), key=lambda x: x[1])[-1]
+                if highest[0] == "person name":
+                    class_score.append((element, highest[1]))
+        if len(class_score):
+            return sorted(class_score, key=lambda x: x[1], reverse=True)[0][0]
+        return ""
+    
+    def find_contact_email(self, items):
+        for item in items: 
+            match = re.search(r'[\w.+-]+@[\w-]+\.[\w.-]+', item)
+            if match:
+                return match.group(0)
+        return ""
+
+    def parse_job_history(self, resume_segment):
+        idx_job_title = self.get_job_titles(resume_segment)
+        current_and_below = False
+        if not len(idx_job_title): 
+            self.parsed_cv["Job History"] = [] 
+            return
+        if idx_job_title[0][0] == 0: current_and_below = True
+        job_history = []
+        for ls_idx, (idx, job_title) in enumerate(idx_job_title): 
+            job_info = {}
+            job_info["Job Title"] = self.filter_job_title(job_title) 
+            # company 
+            if current_and_below: line1, line2 = idx, idx+1
+            else: line1, line2 = idx, idx-1 
+            job_info["Company"] = self.get_job_company(line1, line2, resume_segment)
+            if current_and_below: st_span = idx
+            else: st_span = idx-1
+            # Dates 
+            if ls_idx == len(idx_job_title) - 1: end_span = len(resume_segment) 
+            else: end_span = idx_job_title[ls_idx+1][0]
+            start, end = self.get_job_dates(st_span, end_span, resume_segment)
+            job_info["Start Date"] = start
+            job_info["End Date"] = end
+            job_history.append(job_info)
+        self.parsed_cv["Job History"] = job_history 
+
+    def get_job_titles(self, resume_segment):
+        classes = ["organization", "institution", "company", "job title", "work details"]
+        idx_line = []
+        for idx, line in enumerate(resume_segment):
+            has_verb = False
+            line_modifed = ''.join(i for i in line if not i.isdigit())
+            sentence = self.models.get_flair_sentence(line_modifed)
+            self.tagger.predict(sentence)
+            tags = []
+            for entity in sentence.get_spans('pos'):
+                tags.append(entity.tag)
+                if entity.tag.startswith("V"): 
+                    has_verb = True
+
+            most_common_tag = max(set(tags), key=tags.count)
+            if most_common_tag == "NNP":
+                if not has_verb:
+                    out = self.zero_shot_classifier(line, classes)
+                    class_score = zip(out["labels"], out["scores"])
+                    highest = sorted(class_score, key=lambda x: x[1])[-1]
+
+                    if highest[0] == "job title":
+                        idx_line.append((idx, line))
+
+        return idx_line
+
+    def get_job_dates(self, st, end, resume_segment):
+        search_span = resume_segment[st:end]
+        dates = []
+        for line in search_span:
+            for dt in self.get_ner_in_line(line, "DATE"):
+                if self.isvalidyear(dt.strip()):
+                    dates.append(dt)
+        if len(dates): first = dates[0]
+        exists_second = False
+        if len(dates) > 1:
+            exists_second = True
+            second = dates[1]
+        
+        if len(dates) > 0:
+            if self.has_two_dates(first):
+                d1, d2 = self.get_two_dates(first)
+                return self.format_date(d1), self.format_date(d2)
+            elif exists_second and self.has_two_dates(second): 
+                d1, d2 = self.get_two_dates(second)
+                return self.format_date(d1), self.format_date(d2)
+            else: 
+                if exists_second: 
+                    st = self.format_date(first)
+                    end = self.format_date(second)
+                    return st, end
+                else: 
+                    return (self.format_date(first), "") 
+        else: return ("", "")
+
+    
+    
+    def filter_job_title(self, job_title):
+        job_title_splitter = re.compile(r'[{}]+'.format(re.escape(punctuation.replace("&", "") )))
+        job_title = ''.join(i for i in job_title if not i.isdigit())
+        tokens = job_title_splitter.split(job_title)
+        tokens = [''.join([i for i in tok.strip() if (i.isalpha() or i.strip()=="")]) for tok in tokens if tok.strip()] 
+        classes = ["company", "organization", "institution", "job title", "responsibility",  "details"]
+        new_title = []
+        for token in tokens:
+            if not token: continue
+            res = self.zero_shot_classifier(token, classes)
+            class_score = zip(res["labels"], res["scores"])
+            highest = sorted(class_score, key=lambda x: x[1])[-1]
+            if highest[0] == "job title":
+                new_title.append(token.strip())
+        if len(new_title):
+            return ', '.join(new_title)
+        else: return ', '.join(tokens)
+
+    def has_two_dates(self, date):
+        years = self.get_valid_years()
+        count = 0
+        for year in years:
+            if year in str(date):
+                count+=1
+        return count == 2
+    
+    def get_two_dates(self, date):
+        years = self.get_valid_years()
+        idxs = []
+        for year in years:
+            if year in date: 
+                idxs.append(date.index(year))
+        min_idx = min(idxs)  
+        first = date[:min_idx+4]
+        second = date[min_idx+4:]
+        return first, second
+    def get_valid_years(self):
+        current_year = datetime.today().year
+        years = [str(i) for i in range(current_year-100, current_year)]
+        return years
+
+    def format_date(self, date):
+        out = self.parse_date(date)
+        if out: 
+            return out
+        else: 
+            date = self.clean_date(date)
+            out = self.parse_date(date)
+            if out: 
+                return out
+            else: 
+                return date
+
+    def clean_date(self, date): 
+        try:
+            date = ''.join(i for i in date if i.isalnum() or i =='-' or i == '/')
+            return date
+        except:
+            return date
+
+    def parse_date(self, date):
+        try:
+            date = parser.parse(date)
+            return date.strftime("%m-%Y")
+        except: 
+            try:
+                date = datetime(date)
+                return date.strftime("%m-%Y")
+            except: 
+                return 0 
+
+
+    def isvalidyear(self, date):
+        current_year = datetime.today().year
+        years = [str(i) for i in range(current_year-100, current_year)]
+        for year in years:
+            if year in str(date):
+                return True 
+        return False
+
+    def get_ner_in_line(self, line, entity_type):
+        if entity_type == "DATE": ner = self.ner_dates
+        else: ner = self.ner
+        return [i['word'] for i in ner(line) if i['entity_group'] == entity_type]
+        
+
+    def get_job_company(self, idx, idx1, resume_segment):
+        job_title = resume_segment[idx]
+        if not idx1 <= len(resume_segment)-1: context = ""
+        else:context = resume_segment[idx1]
+        candidate_companies = self.get_ner_in_line(job_title, "ORG") + self.get_ner_in_line(context, "ORG")
+        classes = ["organization", "company", "institution", "not organization", "not company", "not institution"]
+        scores = []
+        for comp in candidate_companies:
+            res = self.zero_shot_classifier(comp, classes)['scores']
+            scores.append(max(res[:3]))
+        sorted_cmps = sorted(zip(candidate_companies, scores), key=lambda x: x[1], reverse=True)
+        if len(sorted_cmps): return sorted_cmps[0][0]
+        return context

ResumeReader.py

@@ -0,0 +1,99 @@
+import re
+import os
+import logging
+import pdfplumber
+
+class ResumeReader:
+
+    def convert_docx_to_txt(self, docx_file,docx_parser):
+        """
+            A utility function to convert a Microsoft docx files to raw text.
+
+            This code is largely borrowed from existing solutions, and does not match the style of the rest of this repo.
+            :param docx_file: docx file with gets uploaded by the user
+            :type docx_file: InMemoryUploadedFile
+            :return: The text contents of the docx file
+            :rtype: str
+        """
+      
+        # doc = docx.Document(docx_file)
+        # allText = []
+        # for docpara in doc.paragraphs:
+            # allText.append(docpara.text)
+        # text = ' '.join(allText)
+        text = ""
+        try:
+            clean_text = re.sub(r'\n+', '\n', text)
+            clean_text = clean_text.replace("\r", "\n").replace("\t", " ")  # Normalize text blob
+            resume_lines = clean_text.splitlines()  # Split text blob into individual lines
+            resume_lines = [re.sub('\s+', ' ', line.strip()) for line in resume_lines if
+                            line.strip()]  # Remove empty strings and whitespaces
+            return resume_lines, text
+        except Exception as e:
+            logging.error('Error in docx file:: ' + str(e))
+            return [], " "
+
+    def convert_pdf_to_txt(self, pdf_file):
+        """
+        A utility function to convert a machine-readable PDF to raw text.
+
+        This code is largely borrowed from existing solutions, and does not match the style of the rest of this repo.
+        :param input_pdf_path: Path to the .pdf file which should be converted
+        :type input_pdf_path: str
+        :return: The text contents of the pdf
+        :rtype: str
+        """
+
+        pdf = pdfplumber.open(pdf_file)
+        raw_text= ""
+       
+        for page in pdf.pages:
+            raw_text += page.extract_text() + "\n"
+
+        pdf.close()                
+      
+        try:
+            full_string = re.sub(r'\n+', '\n', raw_text)
+            full_string = full_string.replace("\r", "\n")
+            full_string = full_string.replace("\t", " ")
+
+            # Remove awkward LaTeX bullet characters
+            full_string = re.sub(r"\uf0b7", " ", full_string)
+            full_string = re.sub(r"\(cid:\d{0,3}\)", " ", full_string)
+            full_string = re.sub(r'• ', " ", full_string)
+
+            # Split text blob into individual lines
+            resume_lines = full_string.splitlines(True)
+
+            # Remove empty strings and whitespaces
+            resume_lines = [re.sub('\s+', ' ', line.strip()) for line in resume_lines if line.strip()]
+           
+            return resume_lines, raw_text 
+        except Exception as e:
+            logging.error('Error in docx file:: ' + str(e))
+            return [], " "
+
+    def read_file(self, file,docx_parser = "tika"):
+        """
+        file : Give path of resume file
+        docx_parser : Enter docx2txt or tika, by default is tika
+        """
+        print("Reading the Resume...")
+        # file = "/content/Asst Manager Trust Administration.docx"
+        file = os.path.join(file)
+        if file.endswith('docx') or file.endswith('doc'):
+            # if file.endswith('doc') and docx_parser == "docx2txt":
+                # docx_parser = "tika"
+                # logging.error("doc format not supported by the docx2txt changing back to tika")
+            resume_lines, raw_text = self.convert_docx_to_txt(file,docx_parser)
+        elif file.endswith('pdf'):
+            resume_lines, raw_text = self.convert_pdf_to_txt(file)
+        elif file.endswith('txt'):
+            with open(file, 'r', encoding='utf-8') as f:
+                resume_lines = f.readlines()
+
+        else:
+            resume_lines = None
+        
+      
+        return resume_lines 

ResumeSegmenter.py

@@ -0,0 +1,259 @@
+from Models import Models 
+
+class ResumeSegmenter:
+
+    def __init__(self, zero_shot_classifier):
+        self.zero_shot_classifier = zero_shot_classifier
+
+    objective = (
+        'career goal',
+        'objective',
+        'career objective',
+        'employment objective',
+        'professional objective',
+        'summary',
+        'summary of qualifications',
+        'digital'
+    )
+
+    work_and_employment = (
+        'employment history',
+        'employment data',
+        'career summary',
+        'work history',
+        'work experience',
+        'experience',
+        'professional experience',
+        'professional background',
+        'professional employment',
+        'additional experience',
+        'career related experience',
+        "professional employment history",
+        'related experience',
+        'programming experience',
+        'freelance',
+        'freelance experience',
+        'army experience',
+        'military experience',
+        'military background',
+    )
+
+    education_and_training = (
+        'academic background',
+        'academic experience',
+        'programs',
+        'courses',
+        'related courses',
+        'education',
+        'educational background',
+        'educational qualifications',
+        'educational training',
+        'education and training',
+        'training',
+        'academic training',
+        'professional training',
+        'course project experience',
+        'related course projects',
+        'internship experience',
+        'internships',
+        'apprenticeships',
+        'college activities',
+        'certifications',
+        'special training',
+    )
+
+    skills_header = (
+        'credentials',
+        'qualifications',
+        'areas of experience',
+        'areas of expertise',
+        'areas of knowledge',
+        'skills',
+        "other skills",
+        "other abilities",
+        'career related skills',
+        'professional skills',
+        'specialized skills',
+        'technical skills',
+        'computer skills',
+        'personal skills',
+        'computer knowledge',        
+        'technologies',
+        'technical experience',
+        'proficiencies',
+        'languages',
+        'language competencies and skills',
+        'programming languages',
+        'competencies'
+    )
+
+    misc = (
+        'activities and honors',
+        'activities',
+        'affiliations',
+        'professional affiliations',
+        'associations',
+        'professional associations',
+        'memberships',
+        'professional memberships',
+        'athletic involvement',
+        'community involvement',
+        'refere',
+        'civic activities',
+        'extra-Curricular activities',
+        'professional activities',
+        'volunteer work',
+        'volunteer experience',
+        'additional information',
+        'interests'
+    )
+
+    accomplishments = (
+        'achievement',
+        'awards and achievements',
+        'licenses',
+        'presentations',
+        'conference presentations',
+        'conventions',
+        'dissertations',
+        'exhibits',
+        'papers',
+        'publications',
+        'professional publications',
+        'research experience',
+        'research grants',
+        'project',
+        'research projects',
+        'personal projects',
+        'current research interests',
+        'thesis',
+        'theses',
+    )
+
+
+    def find_segment_indices(self, string_to_search, resume_segments, resume_indices):
+        for i, line in enumerate(string_to_search):
+
+            if line[0].islower():
+                continue
+
+            header = line.lower()
+
+            if [o for o in self.objective if header.startswith(o)]:
+                try:
+                    resume_segments['objective'][header]
+                except:
+                    resume_indices.append(i)
+                    header = [o for o in self.objective if header.startswith(o)][0]
+                    resume_segments['objective'][header] = i
+            elif [w for w in self.work_and_employment if header.startswith(w)]:
+                try:
+                    resume_segments['work_and_employment'][header]
+                except:
+                    resume_indices.append(i)
+                    header = [w for w in self.work_and_employment if header.startswith(w)][0]
+                    resume_segments['work_and_employment'][header] = i
+            elif [e for e in self.education_and_training if header.startswith(e)]:
+                try:
+                    resume_segments['education_and_training'][header]
+                except:
+                    resume_indices.append(i)
+                    header = [e for e in self.education_and_training if header.startswith(e)][0]
+                    resume_segments['education_and_training'][header] = i
+            elif [s for s in self.skills_header if header.startswith(s)]:
+                try:
+                    resume_segments['skills'][header]
+                except:
+                    resume_indices.append(i)
+                    header = [s for s in self.skills_header if header.startswith(s)][0]
+                    resume_segments['skills'][header] = i
+            elif [m for m in self.misc if header.startswith(m)]:
+                try:
+                    resume_segments['misc'][header]
+                except:
+                    resume_indices.append(i)
+                    header = [m for m in self.misc if header.startswith(m)][0]
+                    resume_segments['misc'][header] = i
+            elif [a for a in self.accomplishments if header.startswith(a)]:
+                try:
+                    resume_segments['accomplishments'][header]
+                except:
+                    resume_indices.append(i)
+                    header = [a for a in self.accomplishments if header.startswith(a)][0]
+                    resume_segments['accomplishments'][header] = i
+
+    def slice_segments(self, string_to_search, resume_segments, resume_indices):
+        resume_segments['contact_info'] = string_to_search[:resume_indices[0]]
+        sec_idxs = {} 
+        for section, value in resume_segments.items():
+            if section == 'contact_info':
+                continue
+            
+            for sub_section, start_idx in value.items():
+                end_idx = len(string_to_search)
+                if (resume_indices.index(start_idx) + 1) != len(resume_indices):
+                    end_idx = resume_indices[resume_indices.index(start_idx) + 1]
+                
+                sec_idxs[section] = (start_idx, end_idx)
+                # print(start_idx, end_idx)
+
+                resume_segments[section][sub_section] = string_to_search[start_idx:end_idx]
+        return sec_idxs
+
+    def find_true_segment(self, dict_of_segments, segment_name):
+        segment_classes = {
+            'objective': ["objective", "other"],
+            'work_and_employment':["employment history", "other"],
+            'education_and_training': ["education", "other"],
+            'skills': ["skills", "other"],
+            'accomplishments': ["accomplishments", "other"],
+            'misc': ["misc", "other"],
+            'contact_info': ["contact information", "other"]
+        }
+        classes = segment_classes[segment_name]
+        scores = []
+        segs = dict_of_segments.keys()
+        for seg in segs:
+            sequence = dict_of_segments[seg]
+            score = self.zero_shot_classifier(' '.join(sequence), classes)["scores"][0]
+            scores.append(score)
+        
+        res = sorted(zip(dict_of_segments.keys(), scores), key=lambda x: x[1], reverse=True)
+        if len(res):
+            return res[0][0]
+        else: return 0 
+
+    def segment(self, string_to_search):
+        print("Segmenting the Resume..")
+        resume_segments = {
+            'objective': {},
+            'work_and_employment': {},
+            'education_and_training': {},
+            'skills': {},
+            'accomplishments': {},
+            'misc': {}
+        }
+
+        resume_indices = []
+
+        self.find_segment_indices(string_to_search, resume_segments, resume_indices)
+        if len(resume_indices) != 0:
+            sec_idx = self.slice_segments(string_to_search, resume_segments, resume_indices)
+        else:
+            resume_segments['contact_info'] = []
+
+        for segment in resume_segments:
+            if segment == "contact_info": continue
+            if not len(resume_segments[segment]) > 1:
+                if len(resume_segments[segment]) == 1: 
+                    only_key = list(resume_segments[segment].keys())[0]
+                    resume_segments[segment] = resume_segments[segment][only_key][1:]
+                    continue
+            if segment != "work_and_employment": continue
+            true_seg = self.find_true_segment(resume_segments[segment], segment)
+            if not true_seg: 
+                resume_segments[segment] = []
+            else:     
+                resume_segments[segment] = resume_segments[segment][true_seg][1:]
+            
+        return resume_segments

app.py

@@ -0,0 +1,18 @@
+from pydoc import describe
+import gradio as gr
+from Main import Main
+
+
+main = Main()
+
+def parse_cv(cv):
+    return main.parse_cv(cv.name)
+    
+
+description = """A demo for a CV parser built with HuggingFace's transformers."""
+article = "Find the code on GitHub 🚀: https://github.com/asimokby/cv-parser-huggingface"
+file_input = gr.inputs.File(file_count="single", type="file", label="Upload a CV: .PDF Or .TXT", optional=False)
+iface = gr.Interface(fn=parse_cv, inputs=file_input, outputs="json", allow_flagging="never",
+                    allow_screenshot=False, title="CV Parser", theme="dark", description=description, article=article)
+
+iface.launch()

requirements.txt

@@ -0,0 +1,109 @@
+aiohttp==3.8.1
+aiosignal==1.2.0
+analytics-python==1.4.0
+anyio==3.5.0
+asgiref==3.5.0
+async-timeout==4.0.2
+attrs==21.4.0
+backoff==1.10.0
+bcrypt==3.2.0
+bpemb==0.3.3
+certifi==2021.10.8
+cffi==1.15.0
+chardet==4.0.0
+charset-normalizer==2.0.11
+click==8.0.3
+colorama==0.4.4
+coloredlogs==15.0.1
+conllu==4.4.1
+cryptography==36.0.1
+cycler==0.11.0
+Cython==0.29.23
+Deprecated==1.2.13
+doc2text==0.2.4
+fastapi==0.73.0
+ffmpy==0.3.0
+filelock==3.4.2
+flair==0.10
+flatbuffers==2.0
+fonttools==4.29.1
+frozenlist==1.3.0
+ftfy==6.0.3
+future==0.18.2
+gdown==3.12.2
+gensim==4.1.2
+gradio==2.7.5.2
+h11==0.13.0
+huggingface-hub==0.4.0
+humanfriendly==10.0
+idna==3.3
+importlib-metadata==3.10.1
+Janome==0.4.1
+Jinja2==3.0.3
+joblib==1.1.0
+kiwisolver==1.3.2
+konoha==4.6.5
+langdetect==1.0.9
+markdown2==2.4.2
+MarkupSafe==2.0.1
+matplotlib==3.5.1
+mime==0.1.0
+monotonic==1.6
+more-itertools==8.8.0
+mpld3==0.3
+multidict==6.0.2
+numpy==1.22.1
+overrides==3.1.0
+packaging==21.3
+pandas==1.4.0
+paramiko==2.9.2
+pdfminer.six==20211012
+pdfplumber==0.6.0
+Pillow==9.0.1
+protobuf==3.19.4
+psutil==5.9.0
+py-cpuinfo==8.0.0
+py3nvml==0.2.7
+pycparser==2.21
+pycryptodome==3.14.1
+pydantic==1.9.0
+pydub==0.25.1
+PyNaCl==1.5.0
+pyparsing==3.0.7
+PyPDF2==1.26.0
+pyreadline3==3.4.1
+PySocks==1.7.1
+pytesseract==0.3.8
+python-dateutil==2.8.2
+python-multipart==0.0.5
+pytz==2021.3
+PyYAML==6.0
+regex==2022.1.18
+requests==2.27.1
+sacremoses==0.0.47
+scikit-learn==1.0.2
+scipy==1.7.3
+segtok==1.5.11
+sentencepiece==0.1.95
+six==1.16.0
+smart-open==5.2.1
+sniffio==1.2.0
+sqlitedict==1.7.0
+starlette==0.17.1
+tabulate==0.8.9
+threadpoolctl==3.1.0
+tokenizers==0.10.3
+torch==1.10.2
+tqdm==4.62.3
+transformers==4.15.0
+typing_extensions==4.0.1
+urllib3==1.26.8
+uvicorn==0.17.4
+Wand==0.6.7
+wcwidth==0.2.5
+Wikipedia-API==0.5.4
+wincertstore==0.2
+wrapt==1.13.3
+xmltodict==0.12.0
+yarl==1.7.2
+zipp==3.7.0