huggingartists

README.md

@@ -0,0 +1,97 @@
+---
+language: en
+datasets:
+- huggingartists/kasta
+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/4fb42a447843eee46b0b77439ecd8fd2.800x800x1.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">Каста (Kasta)</div>
+    <a href="https://genius.com/artists/kasta">
+    	<div style="text-align: center; font-size: 14px;">@kasta</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 Каста (Kasta).
+
+Dataset is available [here](https://huggingface.co/datasets/huggingartists/kasta).
+And can be used with:
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("huggingartists/kasta")
+```
+
+[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/3k79xvbx/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 Каста (Kasta)'s lyrics.
+
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1rphmch0) for full transparency and reproducibility.
+
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1rphmch0/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/kasta')
+generator("I am", num_return_sequences=5)
+```
+
+Or with Transformers library:
+
+```python
+from transformers import AutoTokenizer, AutoModelWithLMHead
+  
+tokenizer = AutoTokenizer.from_pretrained("huggingartists/kasta")
+
+model = AutoModelWithLMHead.from_pretrained("huggingartists/kasta")
+```
+
+## 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": 1.798671007156372, "eval_runtime": 14.7803, "eval_samples_per_second": 20.974, "eval_steps_per_second": 2.639, "epoch": 1.0}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ceb793009ae679222e47bfdf19881b0cc81e692f752afa140730c7a769885f3
+size 497764120

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9e32a544f9f2aeecb3ecb2b4340d16cad5293bc68beaa1af22f7c3012329c1e
+size 995603825

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce22b96aae02342a72d32fa6b896875d50aadd63db23f34a8b4db927bc20d1a3
+size 510403817

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9aca1f70765d22678b6811d3c02285661e257eaf875fa97d081e3cbf73b51318
+size 14567

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98ff732203d181d58cdfc4e2fc4604923740c775fe8b58ca023ef5a380de4639
+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 +1,318 @@
+{
+  "best_metric": 1.798671007156372,
+  "best_model_checkpoint": "output/kasta/checkpoint-246",
+  "epoch": 1.0,
+  "global_step": 246,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.00013706019712792517,
+      "loss": 2.7165,
+      "step": 5
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00013664135833219747,
+      "loss": 2.3564,
+      "step": 10
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00013594519075178427,
+      "loss": 2.2801,
+      "step": 15
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00013497453188602036,
+      "loss": 2.095,
+      "step": 20
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00013373333802928601,
+      "loss": 2.1713,
+      "step": 25
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0001322266681456037,
+      "loss": 2.0746,
+      "step": 30
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00013046066324888032,
+      "loss": 2.1268,
+      "step": 35
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00012844252137283782,
+      "loss": 2.1756,
+      "step": 40
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00012618046823265178,
+      "loss": 2.0614,
+      "step": 45
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00012368372369787862,
+      "loss": 1.9928,
+      "step": 50
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00012096246421332296,
+      "loss": 2.0832,
+      "step": 55
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00011802778132101399,
+      "loss": 1.9764,
+      "step": 60
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00011489163645235038,
+      "loss": 2.0674,
+      "step": 65
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00011156681217467561,
+      "loss": 2.0935,
+      "step": 70
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00010806686009099738,
+      "loss": 1.9101,
+      "step": 75
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00010440604560520553,
+      "loss": 1.9785,
+      "step": 80
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00010059928977791948,
+      "loss": 1.9156,
+      "step": 85
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 9.666210850995393e-05,
+      "loss": 2.0267,
+      "step": 90
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.261054930128376e-05,
+      "loss": 1.8933,
+      "step": 95
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 8.846112584327212e-05,
+      "loss": 1.9379,
+      "step": 100
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 8.423075071075525e-05,
+      "loss": 2.0375,
+      "step": 105
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 7.993666642832438e-05,
+      "loss": 2.014,
+      "step": 110
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 7.55963751917701e-05,
+      "loss": 1.9119,
+      "step": 115
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 7.122756753113643e-05,
+      "loss": 1.9155,
+      "step": 120
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 6.684805020614639e-05,
+      "loss": 1.9652,
+      "step": 125
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 6.247567362788848e-05,
+      "loss": 1.8941,
+      "step": 130
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 5.81282591025852e-05,
+      "loss": 1.8734,
+      "step": 135
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 5.382352619398988e-05,
+      "loss": 1.866,
+      "step": 140
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.957902050047381e-05,
+      "loss": 1.9107,
+      "step": 145
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.541204214117672e-05,
+      "loss": 1.8795,
+      "step": 150
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.1339575242702164e-05,
+      "loss": 1.9314,
+      "step": 155
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.7378218713762616e-05,
+      "loss": 1.8708,
+      "step": 160
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.354411858992822e-05,
+      "loss": 1.9518,
+      "step": 165
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.985290222423505e-05,
+      "loss": 1.9226,
+      "step": 170
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.6319614591883445e-05,
+      "loss": 1.9602,
+      "step": 175
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 2.2958656968642224e-05,
+      "loss": 1.8641,
+      "step": 180
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.9783728232897674e-05,
+      "loss": 1.8891,
+      "step": 185
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.6807769030594122e-05,
+      "loss": 1.7877,
+      "step": 190
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.4042909030642942e-05,
+      "loss": 1.8873,
+      "step": 195
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.1500417485781092e-05,
+      "loss": 1.9452,
+      "step": 200
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.190657300387505e-06,
+      "loss": 1.8256,
+      "step": 205
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 7.123042792471594e-06,
+      "loss": 1.817,
+      "step": 210
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 5.306001321991061e-06,
+      "loss": 1.9063,
+      "step": 215
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 3.7469389418978793e-06,
+      "loss": 1.9345,
+      "step": 220
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.4522102119145282e-06,
+      "loss": 1.8102,
+      "step": 225
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.42709229807627e-06,
+      "loss": 1.8263,
+      "step": 230
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 6.757634636067098e-07,
+      "loss": 1.9225,
+      "step": 235
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 2.0128603879540573e-07,
+      "loss": 1.8347,
+      "step": 240
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 5.593939290255423e-09,
+      "loss": 1.8634,
+      "step": 245
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.798671007156372,
+      "eval_runtime": 14.7214,
+      "eval_samples_per_second": 21.058,
+      "eval_steps_per_second": 2.649,
+      "step": 246
+    }
+  ],
+  "max_steps": 246,
+  "num_train_epochs": 1,
+  "total_flos": 256588775424000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e047fecf95a94d1f9df8385bfee3cd947a122d95ad40757299ed3f7c842d878
+size 2671