Text Classification
Transformers
ONNX
Safetensors
GLiClass
rlcd
typesafe-ai
jev
decision-engine
system-1
modernbert
gliclass
non-autoregressive
zero-token-generation
structured-outputs
calibration
expected-calibration-error
ece
brier-score
proper-scoring-rules
webgpu
edge-ai
fast-inference
banking77
Instructions to use heman10x/rlcd-modernbert-151m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use heman10x/rlcd-modernbert-151m with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="heman10x/rlcd-modernbert-151m")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("heman10x/rlcd-modernbert-151m", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload train_manifest.json with huggingface_hub
Browse files- train_manifest.json +25 -19
train_manifest.json
CHANGED
|
@@ -5,38 +5,44 @@
|
|
| 5 |
"batch_size": 8,
|
| 6 |
"effective_batch_size": 32,
|
| 7 |
"loss_function": "CE + 1.0 * Brier (Proper Scoring Rule)",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
"best_val_accuracy": 0.916,
|
| 9 |
-
"best_val_brier": 0.
|
| 10 |
-
"
|
|
|
|
|
|
|
| 11 |
"device": "mps",
|
| 12 |
"seed": 42,
|
| 13 |
"history": [
|
| 14 |
{
|
| 15 |
"epoch": 1,
|
| 16 |
-
"loss": 0.
|
| 17 |
-
"duration_seconds":
|
| 18 |
"val_accuracy": 0.912,
|
| 19 |
-
"val_nll": 0.
|
| 20 |
-
"val_brier": 0.
|
| 21 |
-
"val_abstention_recall": 0.
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 2,
|
| 25 |
-
"loss": 0.
|
| 26 |
-
"duration_seconds":
|
| 27 |
-
"val_accuracy": 0.
|
| 28 |
-
"val_nll": 0.
|
| 29 |
-
"val_brier": 0.
|
| 30 |
-
"val_abstention_recall": 0.
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 3,
|
| 34 |
-
"loss": 0.
|
| 35 |
-
"duration_seconds":
|
| 36 |
-
"val_accuracy": 0.
|
| 37 |
-
"val_nll": 0.
|
| 38 |
-
"val_brier": 0.
|
| 39 |
-
"val_abstention_recall": 0.
|
| 40 |
}
|
| 41 |
]
|
| 42 |
}
|
|
|
|
| 5 |
"batch_size": 8,
|
| 6 |
"effective_batch_size": 32,
|
| 7 |
"loss_function": "CE + 1.0 * Brier (Proper Scoring Rule)",
|
| 8 |
+
"selection_metric": "val_nll",
|
| 9 |
+
"selected_epoch": 2,
|
| 10 |
+
"scheduler": "get_cosine_schedule_with_warmup",
|
| 11 |
+
"warmup_ratio": 0.1,
|
| 12 |
"best_val_accuracy": 0.916,
|
| 13 |
+
"best_val_brier": 0.11916223463324897,
|
| 14 |
+
"best_val_nll": 0.22156117179827198,
|
| 15 |
+
"best_val_abstention_recall": 0.81,
|
| 16 |
+
"total_training_seconds": 645.6283350839803,
|
| 17 |
"device": "mps",
|
| 18 |
"seed": 42,
|
| 19 |
"history": [
|
| 20 |
{
|
| 21 |
"epoch": 1,
|
| 22 |
+
"loss": 0.8398726179961356,
|
| 23 |
+
"duration_seconds": 158.79704595799558,
|
| 24 |
"val_accuracy": 0.912,
|
| 25 |
+
"val_nll": 0.2697804118457134,
|
| 26 |
+
"val_brier": 0.13645900255650575,
|
| 27 |
+
"val_abstention_recall": 0.72
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 2,
|
| 31 |
+
"loss": 0.298750960456596,
|
| 32 |
+
"duration_seconds": 215.9080920839915,
|
| 33 |
+
"val_accuracy": 0.916,
|
| 34 |
+
"val_nll": 0.22156117179827198,
|
| 35 |
+
"val_brier": 0.11916223463324897,
|
| 36 |
+
"val_abstention_recall": 0.81
|
| 37 |
},
|
| 38 |
{
|
| 39 |
"epoch": 3,
|
| 40 |
+
"loss": 0.12417868667540549,
|
| 41 |
+
"duration_seconds": 224.52253008299158,
|
| 42 |
+
"val_accuracy": 0.93,
|
| 43 |
+
"val_nll": 0.23878168585024195,
|
| 44 |
+
"val_brier": 0.11045744162180102,
|
| 45 |
+
"val_abstention_recall": 0.8
|
| 46 |
}
|
| 47 |
]
|
| 48 |
}
|