# sarandharshansp/amazon-ml-challenge-2025 architecture

> Architecture diagram of the GitHub repository sarandharshansp/amazon-ml-challenge-2025, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-09-26). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/sarandharshansp/amazon-ml-challenge-2025
- Repository: https://github.com/sarandharshansp/amazon-ml-challenge-2025

## Overview

This repository is a collection of experiments and submissions for the Amazon ML Challenge 2025 Smart Product Pricing Challenge. The principal workflow combines product text and images using CLIP and DistilBERT, trains a contrastive multimodal price regressor, produces predictions, and averages the two best models for the final submission. The README also documents CLIP-only, BERT-only, embedding-based, and feature-engineered approaches. No source excerpts were available, so links follow documented model designs and named notebook artifacts rather than inspected runtime calls.

Main entry points: [`13 Oct 25 0859 PM IST/amritha.ipynb`](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb) (CLIP Features), [`13 Oct 25 1139 PM IST - Best/Averaging_code.ipynb`](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/Averaging_code.ipynb) (Best-Model Average), [`13 Oct 25 1139 PM IST - Best/predictions_amritha.csv`](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/predictions_amritha.csv) (Model Predictions), [`13 Oct 25 1139 PM IST - Best/test_predictions_clip_big_epoch-10_full_v2_saran.csv`](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/test_predictions_clip_big_epoch-10_full_v2_saran.csv) (CLIP Predictions)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_winning["Winning Multimodal Model"]
  node_clip["CLIP Features<br/>[amritha.ipynb]"]
  node_distil["DistilBERT Features<br/>[amritha.ipynb]"]
  node_contrast["Contrastive Learning<br/>[amritha.ipynb]"]
  node_fusion["Feature Fusion<br/>[amritha.ipynb]"]
  node_regression["Price Regression<br/>[amritha.ipynb]"]
end

subgraph group_alternatives["Alternative Models"]
  node_cliponly["CLIP Regression<br/>[clip-big-v2.ipynb]"]
  node_bert["BERT Regression<br/>[large-bert.ipynb]"]
  node_distilonly["DistilBERT Regression"]
  node_embeddings["Embedding Regression<br/>[Qwen3-0.6B.ipynb]"]
  node_features["Engineered Features<br/>[Untitled &#40;6&#41;.ipynb]"]
end

subgraph group_artifacts["Prediction Artifacts"]
  node_predictions["Model Predictions"]
  node_clip_predictions["CLIP Predictions"]
  node_average["Best-Model Average"]
  node_submission["Final Predictions"]
end

node_products(("Product Text + Images"))

node_products -->|"encodes text and images"| node_clip
node_products -->|"encodes text views"| node_distil
node_clip -->|"aligns image-text features"| node_contrast
node_distil -->|"aligns augmented views"| node_contrast
node_clip -->|"provides features"| node_fusion
node_distil -->|"provides projection"| node_fusion
node_fusion -->|"predicts price"| node_regression
node_regression -->|"writes predictions"| node_predictions
node_products -.->|"processes text and images"| node_cliponly
node_products -.->|"processes text"| node_bert
node_products -.->|"processes text"| node_distilonly
node_products -.->|"embeds text"| node_embeddings
node_products -.->|"extracts text features"| node_features
node_predictions -->|"averages model output"| node_average
node_clip_predictions -->|"averages model output"| node_average
node_average -->|"writes final output"| node_submission

click node_clip "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb"
click node_distil "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb"
click node_contrast "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb"
click node_fusion "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb"
click node_regression "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb"
click node_cliponly "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200827%20AM%20IST/clip-big-v2.ipynb"
click node_bert "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/12%20Oct%2025%200902%20PM%20IST/large-bert.ipynb"
click node_distilonly "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/12%20Oct%2025%200820%20PM%20IST/distilbert-base-uncased-finetuned-sst-2-english.ipynb"
click node_embeddings "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/11%20Oct%2025%200506%20PM%20IST/qwen_06B_raw/Qwen3-0.6B.ipynb"
click node_features "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/11%20Oct%2025%201129%20PM%20IST/Untitled%20(6).ipynb"
click node_predictions "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/predictions_amritha.csv"
click node_clip_predictions "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/test_predictions_clip_big_epoch-10_full_v2_saran.csv"
click node_average "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/Averaging_code.ipynb"
click node_submission "https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/final_prediction.csv"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_clip,node_distil,node_contrast,node_fusion,node_regression toneBlue
class node_cliponly,node_bert,node_distilonly,node_embeddings,node_features toneAmber
class node_predictions,node_clip_predictions,node_average,node_submission toneMint
class node_products toneIndigo
```

## Components

### Winning Multimodal Model

- [CLIP Features](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb): `13 Oct 25 0859 PM IST/amritha.ipynb`
- [DistilBERT Features](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb): `13 Oct 25 0859 PM IST/amritha.ipynb`
- [Contrastive Learning](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb): `13 Oct 25 0859 PM IST/amritha.ipynb`
- [Feature Fusion](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb): `13 Oct 25 0859 PM IST/amritha.ipynb`
- [Price Regression](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200859%20PM%20IST/amritha.ipynb): `13 Oct 25 0859 PM IST/amritha.ipynb`

### Alternative Models

- [CLIP Regression](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%200827%20AM%20IST/clip-big-v2.ipynb): `13 Oct 25 0827 AM IST/clip-big-v2.ipynb`
- [BERT Regression](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/12%20Oct%2025%200902%20PM%20IST/large-bert.ipynb): `12 Oct 25 0902 PM IST/large-bert.ipynb`
- [DistilBERT Regression](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/12%20Oct%2025%200820%20PM%20IST/distilbert-base-uncased-finetuned-sst-2-english.ipynb): `12 Oct 25 0820 PM IST/distilbert-base-uncased-finetuned-sst-2-english.ipynb`
- [Embedding Regression](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/11%20Oct%2025%200506%20PM%20IST/qwen_06B_raw/Qwen3-0.6B.ipynb): `11 Oct 25 0506 PM IST/qwen_06B_raw/Qwen3-0.6B.ipynb`
- [Engineered Features](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/11%20Oct%2025%201129%20PM%20IST/Untitled%20(6).ipynb): `11 Oct 25 1129 PM IST/Untitled (6).ipynb`

### Prediction Artifacts

- [Model Predictions](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/predictions_amritha.csv): `13 Oct 25 1139 PM IST - Best/predictions_amritha.csv`
- [CLIP Predictions](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/test_predictions_clip_big_epoch-10_full_v2_saran.csv): `13 Oct 25 1139 PM IST - Best/test_predictions_clip_big_epoch-10_full_v2_saran.csv`
- [Best-Model Average](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/Averaging_code.ipynb): `13 Oct 25 1139 PM IST - Best/Averaging_code.ipynb`
- [Final Predictions](https://github.com/sarandharshansp/amazon-ml-challenge-2025/blob/main/13%20Oct%2025%201139%20PM%20IST%20-%20Best/final_prediction.csv): `13 Oct 25 1139 PM IST - Best/final_prediction.csv`

### Other components

- **Product Text + Images**

## Connections

- Product Text + Images → CLIP Features: encodes text and images
- Product Text + Images → DistilBERT Features: encodes text views
- CLIP Features → Contrastive Learning: aligns image-text features
- DistilBERT Features → Contrastive Learning: aligns augmented views
- CLIP Features → Feature Fusion: provides features
- DistilBERT Features → Feature Fusion: provides projection
- Feature Fusion → Price Regression: predicts price
- Price Regression → Model Predictions: writes predictions
- Product Text + Images → CLIP Regression: processes text and images
- Product Text + Images → BERT Regression: processes text
- Product Text + Images → DistilBERT Regression: processes text
- Product Text + Images → Embedding Regression: embeds text
- Product Text + Images → Engineered Features: extracts text features
- Model Predictions → Best-Model Average: averages model output
- CLIP Predictions → Best-Model Average: averages model output
- Best-Model Average → Final Predictions: writes final output