Push model using huggingface_hub.

README.md

@@ -1,69 +1,43 @@
 ---
 license: apache-2.0
-base_model: facebook/bart-base
 tags:
-- generated_from_trainer
-metrics:
-- rouge
-model-index:
-- name: wiki_asp-software_7902_bart-base
-  results: []
+- trl
+- ppo
+- transformers
+- reinforcement-learning
 ---
 
-<!-- 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. -->
+# TRL Model
 
-# wiki_asp-software_7902_bart-base
+This is a [TRL language model](https://github.com/huggingface/trl) that has been fine-tuned with reinforcement learning to
+ guide the model outputs according to a value, function, or human feedback. The model can be used for text generation.
 
-This model is a fine-tuned version of [facebook/bart-base](https://huggingface.co/facebook/bart-base) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 3.0237
-- Rouge1: 0.1523
-- Rouge2: 0.0499
-- Rougel: 0.1269
-- Rougelsum: 0.1266
-- Gen Len: 19.9987
+## Usage
 
-## Model description
+To use this model for inference, first install the TRL library:
 
-More information needed
+```bash
+python -m pip install trl
+```
 
-## Intended uses & limitations
+You can then generate text as follows:
 
-More information needed
+```python
+from transformers import pipeline
 
-## Training and evaluation data
+generator = pipeline("text-generation", model="baek26//tmp/tmp82h_y_48/baek26/wiki_asp-software_7902_bart-base")
+outputs = generator("Hello, my llama is cute")
+```
 
-More information needed
+If you want to use the model for training or to obtain the outputs from the value head, load the model as follows:
 
-## Training procedure
+```python
+from transformers import AutoTokenizer
+from trl import AutoModelForCausalLMWithValueHead
 
-### Training hyperparameters
+tokenizer = AutoTokenizer.from_pretrained("baek26//tmp/tmp82h_y_48/baek26/wiki_asp-software_7902_bart-base")
+model = AutoModelForCausalLMWithValueHead.from_pretrained("baek26//tmp/tmp82h_y_48/baek26/wiki_asp-software_7902_bart-base")
 
-The following hyperparameters were used during training:
-- learning_rate: 5e-05
-- train_batch_size: 4
-- eval_batch_size: 4
-- seed: 42
-- gradient_accumulation_steps: 16
-- total_train_batch_size: 64
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 500
-- num_epochs: 10
-
-### Training results
-
-| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
-|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
-| No log        | 2.51  | 500  | 3.1584          | 0.1516 | 0.0485 | 0.1263 | 0.126     | 20.0    |
-| No log        | 5.03  | 1000 | 3.0483          | 0.1474 | 0.0466 | 0.1226 | 0.1224    | 20.0    |
-| No log        | 7.54  | 1500 | 3.0237          | 0.1523 | 0.0499 | 0.1269 | 0.1266    | 19.9987 |
-
-
-### Framework versions
-
-- Transformers 4.38.2
-- Pytorch 2.0.0+cu117
-- Datasets 2.18.0
-- Tokenizers 0.15.2
+inputs = tokenizer("Hello, my llama is cute", return_tensors="pt")
+outputs = model(**inputs, labels=inputs["input_ids"])
+```

config.json

@@ -1,75 +1,49 @@
 {
-  "_name_or_path": "facebook/bart-base",
-  "activation_dropout": 0.1,
-  "activation_function": "gelu",
-  "add_bias_logits": false,
-  "add_final_layer_norm": false,
-  "architectures": [
-    "BartForConditionalGeneration"
-  ],
-  "attention_dropout": 0.1,
-  "bos_token_id": 0,
-  "classif_dropout": 0.1,
-  "classifier_dropout": 0.0,
-  "d_model": 768,
-  "decoder_attention_heads": 12,
-  "decoder_ffn_dim": 3072,
-  "decoder_layerdrop": 0.0,
-  "decoder_layers": 6,
-  "decoder_start_token_id": 2,
-  "dropout": 0.1,
+  "exp_name": "rlqaf",
+  "seed": 0,
+  "log_with": null,
+  "task_name": null,
+  "model_name": "facebook/bart-base",
+  "query_dataset": "imdb",
+  "reward_model": "sentiment-analysis:lvwerra/distilbert-imdb",
+  "remove_unused_columns": true,
+  "tracker_kwargs": {},
+  "accelerator_kwargs": {},
+  "project_kwargs": {},
+  "tracker_project_name": "trl",
+  "push_to_hub_if_best_kwargs": {},
+  "steps": 20000,
+  "learning_rate": 1.41e-06,
+  "adap_kl_ctrl": true,
+  "init_kl_coef": 0.2,
+  "kl_penalty": "kl",
+  "target": 6,
+  "horizon": 10000,
+  "gamma": 0.9,
+  "lam": 0.95,
+  "cliprange": 0.2,
+  "cliprange_value": 0.2,
+  "vf_coef": 0.1,
+  "batch_size": 1,
+  "forward_batch_size": null,
+  "mini_batch_size": 1,
+  "gradient_accumulation_steps": 1,
+  "world_size": 1,
+  "ppo_epochs": 4,
+  "max_grad_norm": null,
+  "optimize_cuda_cache": null,
+  "optimize_device_cache": false,
   "early_stopping": true,
-  "encoder_attention_heads": 12,
-  "encoder_ffn_dim": 3072,
-  "encoder_layerdrop": 0.0,
-  "encoder_layers": 6,
-  "eos_token_id": 2,
-  "forced_bos_token_id": 0,
-  "forced_eos_token_id": 2,
-  "gradient_checkpointing": false,
-  "id2label": {
-    "0": "LABEL_0",
-    "1": "LABEL_1",
-    "2": "LABEL_2"
-  },
-  "init_std": 0.02,
+  "target_kl": 1,
+  "compare_steps": 1,
+  "ratio_threshold": 10.0,
+  "use_score_scaling": false,
+  "use_score_norm": false,
+  "score_clip": null,
+  "whiten_rewards": false,
   "is_encoder_decoder": true,
-  "label2id": {
-    "LABEL_0": 0,
-    "LABEL_1": 1,
-    "LABEL_2": 2
-  },
-  "max_position_embeddings": 1024,
-  "model_type": "bart",
-  "no_repeat_ngram_size": 3,
-  "normalize_before": false,
-  "normalize_embedding": true,
-  "num_beams": 4,
-  "num_hidden_layers": 6,
-  "pad_token_id": 1,
-  "scale_embedding": false,
-  "task_specific_params": {
-    "summarization": {
-      "length_penalty": 1.0,
-      "max_length": 128,
-      "min_length": 12,
-      "num_beams": 4
-    },
-    "summarization_cnn": {
-      "length_penalty": 2.0,
-      "max_length": 142,
-      "min_length": 56,
-      "num_beams": 4
-    },
-    "summarization_xsum": {
-      "length_penalty": 1.0,
-      "max_length": 62,
-      "min_length": 11,
-      "num_beams": 6
-    }
-  },
-  "torch_dtype": "float32",
-  "transformers_version": "4.38.2",
-  "use_cache": true,
-  "vocab_size": 50265
-}
+  "is_peft_model": false,
+  "backward_batch_size": 1,
+  "global_backward_batch_size": 1,
+  "global_batch_size": 1
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:789a7c0d8c55085c4524ba7b939b95466942bfcdb72a52550099036bdb274fdc
-size 557912620
+oid sha256:def604a43dd0fb64dc25b02587e7c8b57e7e5180f15ddb9c765e6dd08d87439c
+size 557915872