SkyLeopard/placement-season

Placement Season Prediction Model

Model Details

• Model Name: Placement Season Prediction
• Model Type: Random Forest Classifier
• Framework: scikit-learn
• Language: Python
• Task: Binary Classification
• License: Apache 2.0
• Author: SkyLeopard

This model predicts whether a college student will get placed based on academic and experiential features.

Model Description

This is a supervised machine learning model trained to predict student placement outcome (Yes/No) using structured tabular data. The model uses a Random Forest classifier trained on preprocessed student attributes such as:

• IQ
• CGPA
• Academic performance
• Internship experience
• Projects
• Certifications
• Technical and soft skills

The final output is a binary prediction:

• 1 → Placed
• 0 → Not Placed

Example Applications

• Colleges analyzing placement trends
• Students estimating placement probability
• Career guidance systems

Limitations

• Trained on a single dataset, may not generalize to all institutions.
• Performance depends heavily on feature quality.
• Should not be used for high-stakes real-world decisions (e.g., hiring, admissions).

Training Data

The model was trained on a structured CSV dataset containing student records with features such as:

Column Name	Type	Description
College_ID	Categorical (string)	Unique college/student identifier (dropped during training)
IQ	Integer	IQ score of the student
Prev_Sem_Result	Float	Previous semester result (GPA/percentage)
CGPA	Float	Cumulative Grade Point Average
Academic_Performance	Integer	Academic performance rating
Internship_Experience	Categorical	Internship experience (Yes / No)
Extra_Curricular_Score	Integer	Score for extracurricular activities
Communication_Skills	Integer	Communication skills rating
Projects_Completed	Integer	Number of projects completed

Target column:

• Placement (mapped to 0/1)

Preprocessing

The following preprocessing steps were applied:

• Dropped non-informative ID column.

• Categorical mapping:

  • Yes → 1
  • No → 0

• Feature scaling using StandardScaler.

• Train-test split.

Evaluation Results

Evaluation performed on a held-out test set.

Metrics used:

Accuracy: ~0.99 (approx)

How to Use

Get model from the hub

from huggingface_hub import hf_hub_url, cached_download
import joblib
import pandas as pd

REPO_ID = "SkyLeopard/placement-season"
FILENAME = "rf-placement.pkl"

model = joblib.load(
    cached_download(
        hf_hub_url(REPO_ID, FILENAME)
    )
)

# model is a sklearn RandomForestClassifier

Get sample data from this repo

data_file = cached_download(
    hf_hub_url(REPO_ID, "college_student_placement_dataset.csv")
)

df = pd.read_csv(data_file)

X = df.drop(["College_ID", "Placement"], axis=1)
y = df["Placement"]

print(X.head(3))

Predict

# Encode Internship_Experience manually
X["Internship_Experience"] = X["Internship_Experience"].map({"Yes": 1, "No": 0})

labels = model.predict(X[:3])
print(labels)

Eval

model.score(X, y)

Ethical Considerations

• Predictions may reflect biases in the training dataset.
• Should not be used to deny opportunities.
• Use as a decision-support tool only.

---

Model Card Authors

• SkyLeopard

Citation

If you use this model:

@misc{placement_season_model,
  author = {SkyLeopard},
  title = {Placement Season Prediction Model},
  year = {2026},
  publisher = {Hugging Face}
}