Publish the model trained 2026-10-18

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+model.keras filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,112 @@
+---
+license: mit
+library_name: keras
+pipeline_tag: tabular-regression
+tags:
+- tabular-regression
+- leetcode
+- rating-prediction
+- numpy
+---
+
+# LeetCode Rating Predictor
+
+Predicts how much a LeetCode contest placement moves a user's rating, from the
+rank and the user's contest history. It is a small Dense network
+(`15 -> 128 -> 64 -> 32 -> 1`, 12,417 parameters) that runs on plain
+NumPy or JavaScript; no ML framework is needed to serve it.
+
+It powers the [web app](https://leetcode-rating-predictor.onrender.com) and the
+[browser extension](https://github.com/Sagargupta16/LeetCode_Rating_Predictor#browser-extension), which loads `model.json`
+from this repo and adds a Predicted column to LeetCode ranking pages.
+Source, training and feature code: [GitHub](https://github.com/Sagargupta16/LeetCode_Rating_Predictor).
+
+## Files
+
+| File | What |
+|---|---|
+| `model.json` | Network and scaler in one JSON file (what the extension loads) |
+| `weights.npz`, `scaler.json` | The same network as NumPy arrays (what the API loads) |
+| `model.keras` | The trained Keras model these are exported from |
+| `metrics.json` | Split sizes and held-out scores of this model |
+| `backtest.md` | Accuracy on recent contests of users outside the training data |
+
+## Results
+
+Held-out users (818 users, 27,698 contest
+records never seen in training), trained 2026-10-18:
+
+| Metric | Value |
+|---|---|
+| MAE | 14.14 |
+| RMSE | 22.01 |
+| MAE of always predicting 0 | 27.79 |
+
+Backtest on recent real contests of users outside the training data
+(`scripts/backtest.py`, 2026-10-19):
+
+| Metric | Value |
+|---|---|
+| Contests scored | 173 |
+| MAE | 9.11 |
+| Direction right (gain vs loss) | 90.8% |
+| MAE of always predicting 0 | 29.34 |
+
+## Inputs
+
+One row per contest result, in this order. `scaler` in `model.json` maps them
+to the network's range (`x * scale + min`).
+
+| | Feature |
+|---|---|
+| f1 | `rating_before` |
+| f2 | `rank` |
+| f3 | `participants` |
+| f4 | `rank_percent` |
+| f5 | `contest_index` |
+| f6 | `avg_solve_rate` |
+| f7 | `avg_finish_time` |
+| f8 | `recent_solve_rate` |
+| f9 | `recent_finish_time` |
+| f10 | `rating_trend` |
+| f11 | `max_rating` |
+| f12 | `log_rank` |
+| f13 | `rating_x_rank_share` |
+| f14 | `solve_rate_x_rating` |
+| f15 | `finish_time_share` |
+
+`participants` is `max(rank * 1.5, 10000)`, the form the training data was
+built with, not the registered count. See `app/services/features.py` on GitHub
+for the exact definitions.
+
+## Use
+
+```python
+import json, numpy as np
+from huggingface_hub import hf_hub_download
+
+model = json.load(open(hf_hub_download("Sagargupta1610/leetcode-rating-predictor", "model.json")))
+x = np.asarray(features, dtype=float)  # the 15 inputs above
+x = x * np.asarray(model["scaler"]["scale"]) + np.asarray(model["scaler"]["min"])
+for layer in model["layers"]:
+    x = x @ np.asarray(layer["kernel"]) + np.asarray(layer["bias"])
+    if layer["activation"] == "relu":
+        x = np.maximum(x, 0)
+rating_change = float(x[0])
+```
+
+## Training
+
+Trained on the contest records collected by `scripts/update_data.py`: 295,899 records from 8,182 users,
+split by user so no user is on both sides (242,108 train,
+26,093 validation, 27,698 test), huber loss,
+seed 42, 62 epochs with early stopping.
+
+## Limits
+
+- A user's first few contests can move the rating by 100+ points; those are the
+  largest errors.
+- The participant count is synthetic (see Inputs), so the model does not know
+  how large a contest really was.
+- LeetCode can change its rating formula; the backtest is re-run weekly to
+  catch drift.

backtest.md

@@ -0,0 +1,23 @@
+# Backtest
+
+Generated by `python scripts/backtest.py` on 2026-10-19. Each prediction
+uses only the history a user had before that contest, built by the same feature
+code the API serves, and is compared with the rating change LeetCode applied.
+
+| Metric | Value |
+|---|---|
+| Contests scored | 173 |
+| Users | 50 |
+| MAE | **9.11** |
+| RMSE | 19.85 |
+| Median absolute error | 5.59 |
+| Predictions within 10 points | 73.4% |
+| Direction right (gain vs loss) | 90.8% |
+| MAE of always predicting 0 | 29.34 |
+
+Users were sampled with seed 7 from `data/usernames.json` past index
+12000, the slice `scripts/update_data.py` never fetches, so none of
+these contests is in the training data. Up to the last 5 rated contests
+per user count, if they started within 365 days of the run.
+869 sampled accounts had no contest in that window.
+Model: `models/weights.npz`.

metrics.json

@@ -0,0 +1,19 @@
+{
+  "trained": "2026-10-18",
+  "split": "by user",
+  "loss": "huber",
+  "seed": 42,
+  "records": 295899,
+  "users": 8182,
+  "train_records": 242108,
+  "val_records": 26093,
+  "test_records": 27698,
+  "test_users": 818,
+  "epochs_run": 62,
+  "params": 12417,
+  "test": {
+    "mae": 14.14,
+    "rmse": 22.01,
+    "zero_baseline_mae": 27.79
+  }
+}

model.keras

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e557df8d962cbf075e533c0692183ba5bc6372818547e1579195f3c9104c0528
+size 184012

scaler.json

@@ -0,0 +1,38 @@
+{
+  "type": "MinMaxScaler",
+  "n_features_in": 15,
+  "scale": [
+    0.0004001544596214139,
+    2.491839226533104e-05,
+    1.9921112394916133e-05,
+    0.015001500150015003,
+    0.0015772870662460567,
+    1.0,
+    5.1148278860416345e-05,
+    1.0,
+    5.1148278860416345e-05,
+    0.0017399474535869017,
+    0.0004595343262951056,
+    0.10093974906378382,
+    0.000510551230420807,
+    0.0002867677557417787,
+    0.2761973153620947
+  ],
+  "min": [
+    -0.4302804882684717,
+    -2.491839226533104e-05,
+    -0.19921112394916132,
+    -0.00015001500150015003,
+    0.0,
+    0.0,
+    -0.005012531328320802,
+    0.0,
+    -0.005012531328320802,
+    0.1451672959476624,
+    -0.6425231926974482,
+    -0.06996134007610857,
+    -7.658268456312105e-05,
+    0.0,
+    -0.004999171408053915
+  ]
+}

weights.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37600eaa49d5a96e1e6b478332b3e5d1cc94ff245259241f33d38e8fae7154a2
+size 52142