Upload TFT5ForConditionalGeneration

README.md

@@ -14,15 +14,15 @@ probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [deanna-emery/ASL_t5_movinet_sentence](https://huggingface.co/deanna-emery/ASL_t5_movinet_sentence) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.2383
-- Train Top 1: 0.9491
-- Train Top 5: 0.9730
-- Validation Loss: 0.3164
-- Validation Top 1: 0.9416
-- Validation Top 5: 0.9642
-- Train Bleu: 0.5761
-- Train Gen Len: 13.5547
-- Epoch: 2
+- Train Loss: 0.1776
+- Train Top 1: 0.9586
+- Train Top 5: 0.9815
+- Validation Loss: 0.3415
+- Validation Top 1: 0.9408
+- Validation Top 5: 0.9633
+- Train Bleu: 1.984
+- Train Gen Len: 12.7656
+- Epoch: 3
 
 ## Model description
 
@@ -41,16 +41,17 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- optimizer: {'name': 'Adafactor', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'CosineDecay', 'config': {'initial_learning_rate': 0.0005, 'decay_steps': 1998, 'alpha': 0.0, 'name': None, 'warmup_target': 0.0005, 'warmup_steps': 600}, 'registered_name': None}, 'beta_2_decay': -0.8, 'epsilon_1': 1e-30, 'epsilon_2': 0.001, 'clip_threshold': 1.0, 'relative_step': True}
+- optimizer: {'name': 'Adafactor', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'CosineDecay', 'config': {'initial_learning_rate': 0.0005, 'decay_steps': 3330, 'alpha': 0.0, 'name': None, 'warmup_target': 0.0005, 'warmup_steps': 600}, 'registered_name': None}, 'beta_2_decay': -0.8, 'epsilon_1': 1e-30, 'epsilon_2': 0.001, 'clip_threshold': 1.0, 'relative_step': True}
 - training_precision: float32
 
 ### Training results
 
 | Train Loss | Train Top 1 | Train Top 5 | Validation Loss | Validation Top 1 | Validation Top 5 | Train Bleu | Train Gen Len | Epoch |
 |:----------:|:-----------:|:-----------:|:---------------:|:----------------:|:----------------:|:----------:|:-------------:|:-----:|
-| 0.2845     | 0.9432      | 0.9669      | 0.3092          | 0.9416           | 0.9644           | 0.4712     | 14.4505       | 0     |
-| 0.2619     | 0.9459      | 0.9697      | 0.3116          | 0.9417           | 0.9643           | 0.6262     | 13.9115       | 1     |
-| 0.2383     | 0.9491      | 0.9730      | 0.3164          | 0.9416           | 0.9642           | 0.5761     | 13.5547       | 2     |
+| 0.2351     | 0.9496      | 0.9736      | 0.3181          | 0.9413           | 0.9640           | 1.1898     | 13.0052       | 0     |
+| 0.2173     | 0.9522      | 0.9760      | 0.3255          | 0.9409           | 0.9638           | 1.525      | 13.0182       | 1     |
+| 0.1965     | 0.9554      | 0.9789      | 0.3372          | 0.9405           | 0.9630           | 1.4326     | 13.8776       | 2     |
+| 0.1776     | 0.9586      | 0.9815      | 0.3415          | 0.9408           | 0.9633           | 1.984      | 12.7656       | 3     |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:856df96167528cc61d42046f91f4b8b5a2e18690219278131e9904ab2844c631
+oid sha256:52f925a56ff4e2fea8cf01fb0c0e951e1db7f152a0899a24637e3e2f9431467c
 size 1089544048