huggingartists

README.md

@@ -0,0 +1,97 @@
+---
+language: en
+datasets:
+- huggingartists/big-russian-boss
+tags:
+- huggingartists
+- lyrics
+- lm-head
+- causal-lm
+widget:
+- text: "I am"
+---
+
+<div class="inline-flex flex-col" style="line-height: 1.5;">
+    <div class="flex">
+        <div
+			style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://images.genius.com/d66eeeef006738708df1e52b84c34c14.403x403x1.jpg&#39;)">
+        </div>
+    </div>
+    <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 HuggingArtists Model 🤖</div>
+    <div style="text-align: center; font-size: 16px; font-weight: 800">Big Russian Boss</div>
+    <a href="https://genius.com/artists/big-russian-boss">
+    	<div style="text-align: center; font-size: 14px;">@big-russian-boss</div>
+    </a>
+</div>
+
+I was made with [huggingartists](https://github.com/AlekseyKorshuk/huggingartists).
+
+Create your own bot based on your favorite artist with [the demo](https://colab.research.google.com/github/AlekseyKorshuk/huggingartists/blob/master/huggingartists-demo.ipynb)!
+
+## How does it work?
+
+To understand how the model was developed, check the [W&B report](https://wandb.ai/huggingartists/huggingartists/reportlist).
+
+## Training data
+
+The model was trained on lyrics from Big Russian Boss.
+
+Dataset is available [here](https://huggingface.co/datasets/huggingartists/big-russian-boss).
+And can be used with:
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("huggingartists/big-russian-boss")
+```
+
+[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1ju9bqqi/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
+
+## Training procedure
+
+The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Big Russian Boss's lyrics.
+
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3820n7qx) for full transparency and reproducibility.
+
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3820n7qx/artifacts) is logged and versioned.
+
+## How to use
+
+You can use this model directly with a pipeline for text generation:
+
+```python
+from transformers import pipeline
+generator = pipeline('text-generation',
+                     model='huggingartists/big-russian-boss')
+generator("I am", num_return_sequences=5)
+```
+
+Or with Transformers library:
+
+```python
+from transformers import AutoTokenizer, AutoModelWithLMHead
+  
+tokenizer = AutoTokenizer.from_pretrained("huggingartists/big-russian-boss")
+
+model = AutoModelWithLMHead.from_pretrained("huggingartists/big-russian-boss")
+```
+
+## Limitations and bias
+
+The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
+
+In addition, the data present in the user's tweets further affects the text generated by the model.
+
+## About
+
+*Built by Aleksey Korshuk*
+
+[![Follow](https://img.shields.io/github/followers/AlekseyKorshuk?style=social)](https://github.com/AlekseyKorshuk)
+
+[![Follow](https://img.shields.io/twitter/follow/alekseykorshuk?style=social)](https://twitter.com/intent/follow?screen_name=alekseykorshuk)
+
+[![Follow](https://img.shields.io/badge/dynamic/json?color=blue&label=Telegram%20Channel&query=%24.result&url=https%3A%2F%2Fapi.telegram.org%2Fbot1929545866%3AAAFGhV-KKnegEcLiyYJxsc4zV6C-bdPEBtQ%2FgetChatMemberCount%3Fchat_id%3D-1001253621662&style=social&logo=telegram)](https://t.me/joinchat/_CQ04KjcJ-4yZTky)
+
+For more details, visit the project repository.
+
+[![GitHub stars](https://img.shields.io/github/stars/AlekseyKorshuk/huggingartists?style=social)](https://github.com/AlekseyKorshuk/huggingartists)

config.json

@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "gpt2",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "gradient_checkpointing": false,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "resid_pdrop": 0.1,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 200,
+      "min_length": 100,
+      "temperature": 1.0,
+      "top_p": 0.95
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.10.2",
+  "use_cache": true,
+  "vocab_size": 50257
+}

evaluation.txt

@@ -0,0 +1 @@
+{"eval_loss": 1.9292813539505005, "eval_runtime": 8.8697, "eval_samples_per_second": 20.519, "eval_steps_per_second": 2.593, "epoch": 2.0}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d7078748de8d1dec30781a7ade52b843feeff6c609b8fe3b79de80ce5ccb01b
+size 497764120

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c78f845755671930de872d23d761a1bf2eb80ae70333ec08006232f028e4f13f
+size 995603825

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c90fcbd897010b174500212e9c722c022eef9b86040281f34f4ea0e8ff63282
+size 510403817

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df4003aeeda8e472a41875f1db1ee7fb106906695db02639a5ced4a76e68a877
+size 14503

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71dece44af9bc94414a578525516fd968bcf7cd964f733d507bddca799880cea
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}

trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_metric": 1.9292813539505005,
+  "best_model_checkpoint": "output/big-russian-boss/checkpoint-248",
+  "epoch": 2.0,
+  "global_step": 248,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00013665032034568096,
+      "loss": 2.5025,
+      "step": 5
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0001350100903542216,
+      "loss": 2.3418,
+      "step": 10
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0001323055957706572,
+      "loss": 2.1866,
+      "step": 15
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00012858017786751834,
+      "loss": 2.2697,
+      "step": 20
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00012389353887294475,
+      "loss": 2.131,
+      "step": 25
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00011832078520391764,
+      "loss": 2.0523,
+      "step": 30
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00011195122383741646,
+      "loss": 2.0523,
+      "step": 35
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00010488693110842961,
+      "loss": 2.0586,
+      "step": 40
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 9.724111687075534e-05,
+      "loss": 2.1215,
+      "step": 45
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 8.913631023597238e-05,
+      "loss": 2.0016,
+      "step": 50
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 8.070239596528567e-05,
+      "loss": 2.0287,
+      "step": 55
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 7.207453298233569e-05,
+      "loss": 1.9277,
+      "step": 60
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 6.339098836414812e-05,
+      "loss": 1.9488,
+      "step": 65
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 5.4790921521917906e-05,
+      "loss": 1.9926,
+      "step": 70
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.641215408156227e-05,
+      "loss": 1.8875,
+      "step": 75
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.838896120314628e-05,
+      "loss": 1.9105,
+      "step": 80
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 3.084991973471496e-05,
+      "loss": 1.8968,
+      "step": 85
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 2.3915847685255567e-05,
+      "loss": 1.9608,
+      "step": 90
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7697868038067536e-05,
+      "loss": 1.8985,
+      "step": 95
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.2295627933180862e-05,
+      "loss": 1.9667,
+      "step": 100
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 7.795701757575202e-06,
+      "loss": 2.0107,
+      "step": 105
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.270203734710291e-06,
+      "loss": 1.7971,
+      "step": 110
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.7756322475164567e-06,
+      "loss": 2.0397,
+      "step": 115
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 3.519644153159924e-07,
+      "loss": 1.9473,
+      "step": 120
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.9632209539413452,
+      "eval_runtime": 8.8591,
+      "eval_samples_per_second": 20.544,
+      "eval_steps_per_second": 2.596,
+      "step": 124
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.2015434664866374e-08,
+      "loss": 2.0051,
+      "step": 125
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 7.910729510915516e-07,
+      "loss": 1.9063,
+      "step": 130
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 2.6468123211622144e-06,
+      "loss": 1.9292,
+      "step": 135
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 5.559494122852188e-06,
+      "loss": 1.8741,
+      "step": 140
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 9.482440748992502e-06,
+      "loss": 1.9671,
+      "step": 145
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.4352784446073873e-05,
+      "loss": 1.9278,
+      "step": 150
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 2.009247481060283e-05,
+      "loss": 1.8776,
+      "step": 155
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 2.660952959723034e-05,
+      "loss": 1.8322,
+      "step": 160
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.379950879364461e-05,
+      "loss": 1.9671,
+      "step": 165
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.154718833922634e-05,
+      "loss": 1.8845,
+      "step": 170
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 4.972840666504281e-05,
+      "loss": 1.9002,
+      "step": 175
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.8212054463186014e-05,
+      "loss": 1.8912,
+      "step": 180
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 6.686217579811531e-05,
+      "loss": 1.9383,
+      "step": 185
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 7.554014688833781e-05,
+      "loss": 1.9313,
+      "step": 190
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 8.410689764204419e-05,
+      "loss": 1.8199,
+      "step": 195
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.242514034515462e-05,
+      "loss": 1.8346,
+      "step": 200
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.0001003615697856113,
+      "loss": 1.8638,
+      "step": 205
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.00010778899955550273,
+      "loss": 1.8885,
+      "step": 210
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.0001145884002954007,
+      "loss": 1.8695,
+      "step": 215
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00012065080721672542,
+      "loss": 1.8793,
+      "step": 220
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00012587906633286313,
+      "loss": 1.8785,
+      "step": 225
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00013018939141455285,
+      "loss": 1.8644,
+      "step": 230
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00013351270671832803,
+      "loss": 1.85,
+      "step": 235
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.00013579575396992113,
+      "loss": 1.8876,
+      "step": 240
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 0.0001370019458625086,
+      "loss": 1.8228,
+      "step": 245
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.9292813539505005,
+      "eval_runtime": 8.8397,
+      "eval_samples_per_second": 20.589,
+      "eval_steps_per_second": 2.602,
+      "step": 248
+    }
+  ],
+  "max_steps": 248,
+  "num_train_epochs": 2,
+  "total_flos": 258156527616000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8e32e9b3301c0343a4b1e8de43fbb14e5765d03eff46f9ce78a63cf6a8f20a6
+size 2671