Upload README.md with huggingface_hub

README.md

@@ -1,68 +1,72 @@
 ---
-library_name: transformers
-base_model: seyonec/ChemBERTa-zinc-base-v1
+license: apache-2.0
 tags:
-- generated_from_trainer
-metrics:
-- accuracy
-- precision
-- recall
-- f1
+  - chemistry
+  - precite
+  - chemberta
+datasets:
+  - blainetrain/precite-dataset-FLP-Test-v10
+base_model: seyonec/ChemBERTa-zinc-base-v1
 model-index:
-- name: FLP-Test-v10
-  results: []
+  - name: FLP-Test-v10
+    results:
+      - task:
+          type: molecular-property-prediction
+        metrics:
+          - name: Accuracy
+            type: accuracy
+            value: 0.5000
+          - name: F1
+            type: f1
+            value: 0.3333
+          - name: Precision
+            type: precision
+            value: 0.2500
+          - name: Recall
+            type: recall
+            value: 0.5000
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
-# FLP-Test-v10
-
-This model is a fine-tuned version of [seyonec/ChemBERTa-zinc-base-v1](https://huggingface.co/seyonec/ChemBERTa-zinc-base-v1) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 1.3126
-- Accuracy: 0.5
-- Precision: 0.25
-- Recall: 0.5
-- F1: 0.3333
-
-## Model description
+# FLP Test v10
 
-More information needed
+A chemistry prediction model fine-tuned on Precite platform.
 
-## Intended uses & limitations
+## Model Details
 
-More information needed
+- **Base Model**: [seyonec/ChemBERTa-zinc-base-v1](https://huggingface.co/seyonec/ChemBERTa-zinc-base-v1)
+- **Fine-tuned On**: 8 training samples, 2 validation samples (80/20 split)
+- **Task**: Molecular property prediction (4 classes)
+- **Epochs**: 2
+- **Training Date**: 2026-02-04
 
-## Training and evaluation data
+## Performance Metrics (20% Holdout Test Set)
 
-More information needed
+| Metric | Value |
+|--------|-------|
+| **Accuracy** | 0.5000 |
+| **F1 Score** | 0.3333 |
+| **Precision** | 0.2500 |
+| **Recall** | 0.5000 |
+| Training Loss | 1.4379 |
 
-## Training procedure
+## Label Classes
 
-### Training hyperparameters
+- `high`
+- `low`
+- `medium`
+- `very_low`
 
-The following hyperparameters were used during training:
-- learning_rate: 5e-05
-- train_batch_size: 4
-- eval_batch_size: 4
-- seed: 42
-- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 100
-- num_epochs: 2
+## Usage
 
-### Training results
+This model can be queried through the Precite platform for FLP chemistry predictions.
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
-| No log        | 1.0   | 2    | 1.3126          | 0.5      | 0.25      | 0.5    | 0.3333 |
-| No log        | 2.0   | 4    | 1.3204          | 0.5      | 0.25      | 0.5    | 0.3333 |
+```python
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
 
+model = AutoModelForSequenceClassification.from_pretrained("blainetrain/FLP-Test-v10")
+tokenizer = AutoTokenizer.from_pretrained("blainetrain/FLP-Test-v10")
+```
 
-### Framework versions
+## Training Data
 
-- Transformers 5.0.0
-- Pytorch 2.10.0+cu128
-- Datasets 4.5.0
-- Tokenizers 0.22.2
+See the associated dataset: [blainetrain/precite-dataset-FLP-Test-v10](https://huggingface.co/datasets/blainetrain/precite-dataset-FLP-Test-v10)