Hi there, I'm Tim 👋

I'm an AI Developer and recent graduate of a master's degree in AI at Utrecht University. I'm passionate about all things AI, with a specific interest in Natural Language Processing (NLP) and Computer Vision.

📚 My Highlighted Repositories

Here are some of the projects I've worked on that I'm most proud of:

SAURON: This repository contains the code for my master's thesis. By running this code, you can create a writing style representation transformer. It's a project I'm particularly proud of, as it represents the culmination of my studies in AI.
Automatic Piano Fingering: This is the code for my bachelor's thesis. It creates a Q-Learning algorithm from scratch to determine the most optimal piano fingering. This project was a great opportunity to apply AI to a unique and interesting problem.
TiML: This repository is a from-scratch implementation of the most common machine learning methods. It includes simple explanations and is implemented in Python + NumPy. This project was a great opportunity to delve deep into the inner workings of machine learning algorithms.

🤝 My Collaboration Efforts

I believe in the power of collaboration and have had the opportunity to work with some amazing teams. Here are a few collaborations I'm proud of:

FinTwit_Bot: This is a Discord bot written in Python, with the purpose of providing an overview of the financial markets discussed on Twitter. The bot is able to distinguish multiple markets based on the tickers mentioned in the tweets and provides detailed information of the financial data discussed in a Tweet.
Axie_Manager_Bot: This is a Discord bot written in Python, with the purpose of helping our guild manage the scholars in our Discord server. The purpose of this bot is that it can be used for guilds with multiple scholars and different managers, who each have their own scholars and wallets.

More collaboration projects to come...

📫 How to reach me

The best way to reach me is by email at [email protected]. I'm always open to new opportunities and collaborations.

Thanks for stopping by!

Add .ipynb notebook for analysis

SHAP -> only works using notebook...

import shap
from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline
from datasets import load_dataset

# Load pre-trained model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("StephanAkkerman/FinTwitBERT-sentiment")
model = AutoModelForSequenceClassification.from_pretrained(
    "StephanAkkerman/FinTwitBERT-sentiment"
)

classifier = pipeline(
    "text-classification",
    model=model,
    tokenizer=tokenizer,
    device=0,
    top_k=None,
)


dataset = load_dataset(
    "financial_phrasebank",
    cache_dir="datasets/",
    split="train",
    name="sentences_50agree",
)

# Rename sentence to text
dataset = dataset.rename_column("sentence", "text")

short_data = [v[:500] for v in dataset["text"][:20]]

# define the explainer
explainer = shap.Explainer(classifier)

# explain the predictions of the pipeline on the first two samples
shap_values = explainer(short_data[:2])

# Try visualizing the SHAP values without indexing
shap.plots.text(shap_values)

LIME

import torch.nn.functional as F
from lime.lime_text import LimeTextExplainer

from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained("StephanAkkerman/FinTwitBERT-sentiment")
model = AutoModelForSequenceClassification.from_pretrained(
    "StephanAkkerman/FinTwitBERT-sentiment"
)
class_names = ["BULLISH", "BEARISH", "NEUTRAL"]


def predictor(texts):
    outputs = model(**tokenizer(texts, return_tensors="pt", padding=True))
    probas = F.softmax(outputs.logits).detach().numpy()
    return probas


explainer = LimeTextExplainer(class_names=class_names)

str_to_predict = "surprising increase in revenue in spite of decrease in market share"
exp = explainer.explain_instance(
    str_to_predict, predictor, num_features=20, num_samples=2000
)

exp.save_to_file("temp.html")
fig = exp.as_pyplot_figure()
fig.savefig("lime_report.jpg")

timkoornstra / fintwitbert Goto Github PK

fintwitbert's Introduction

Hi there, I'm Tim 👋

📚 My Highlighted Repositories

🤝 My Collaboration Efforts

📫 How to reach me

fintwitbert's People

Contributors

Stargazers

Watchers

fintwitbert's Issues

Recommend Projects

Recommend Topics

Recommend Org

Jobs