huggingartists

README.md

@@ -0,0 +1,93 @@
+---
+language: en
+datasets:
+- huggingartists/tom-waits
+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/505d2d5d1d43304dca446fd2e788a0f8.750x750x1.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">Tom Waits</div>
+    <a href="https://genius.com/artists/tom-waits">
+    	<div style="text-align: center; font-size: 14px;">@tom-waits</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 Tom Waits.
+
+Dataset is available [here](https://huggingface.co/datasets/huggingartists/tom-waits).
+And can be used with:
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("huggingartists/tom-waits")
+```
+
+Or with Transformers library:
+
+```python
+from transformers import AutoTokenizer, AutoModelWithLMHead
+  
+tokenizer = AutoTokenizer.from_pretrained("huggingartists/tom-waits")
+
+model = AutoModelWithLMHead.from_pretrained("huggingartists/tom-waits")
+```
+
+[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/216zw2jw/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 Tom Waits's lyrics.
+
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/16iei9vt) for full transparency and reproducibility.
+
+At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/16iei9vt/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/tom-waits')
+generator("I am", num_return_sequences=5)
+```
+
+## 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)
+
+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": "4.9.1",
+  "use_cache": true,
+  "vocab_size": 50257
+}

flax_model.msgpack

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5970c2b5d3f46666cabb8962860ab904a617c379a81705faf5d7a6fb088f040d
+size 497764120

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab9aa653691ed6d06e3476bc97c31686e2e7de9bae139faaad168caa70c9c3c3
+size 995603825

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c740f283a3819ddb62bc794f49e76fe5fb597109db3672aa74cf62d0c37a13fd
+size 510403817

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b6cc17bd3c2a0e7a4670f3b8e5841d9f4bc2b0a91d85a50187653eae4d73a5e
+size 14503

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:210db157b55dd747a9007cf5cf1259093551c56de3cafadcc170777b60990c64
+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 +1,136 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "global_step": 103,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.0001364038084940016,
+      "loss": 3.7165,
+      "step": 5
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0001340337156353132,
+      "loss": 3.7345,
+      "step": 10
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00013014473739485363,
+      "loss": 3.4567,
+      "step": 15
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0001248271469925082,
+      "loss": 3.512,
+      "step": 20
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00011820437942275874,
+      "loss": 3.5706,
+      "step": 25
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00011043016620995614,
+      "loss": 3.4168,
+      "step": 30
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00010168496690295909,
+      "loss": 3.5556,
+      "step": 35
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 9.21717801433564e-05,
+      "loss": 3.3815,
+      "step": 40
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 8.211143154318805e-05,
+      "loss": 3.2048,
+      "step": 45
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 7.173744775268251e-05,
+      "loss": 3.3812,
+      "step": 50
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 6.12906357041245e-05,
+      "loss": 3.4762,
+      "step": 55
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 5.101349286159054e-05,
+      "loss": 3.2324,
+      "step": 60
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.114457822907933e-05,
+      "loss": 3.3225,
+      "step": 65
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.1912974780463516e-05,
+      "loss": 3.5544,
+      "step": 70
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 2.3532971852551605e-05,
+      "loss": 3.3748,
+      "step": 75
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.6199090936639392e-05,
+      "loss": 3.3532,
+      "step": 80
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.0081570332763395e-05,
+      "loss": 3.4422,
+      "step": 85
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 5.322413479482709e-06,
+      "loss": 3.556,
+      "step": 90
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.032092687651423e-06,
+      "loss": 3.3998,
+      "step": 95
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.869847930266052e-07,
+      "loss": 3.5112,
+      "step": 100
+    }
+  ],
+  "max_steps": 103,
+  "num_train_epochs": 1,
+  "total_flos": 107260379136000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8b55cdf2da34a0eeb37dd5726256b79632c6d1223ef65e3b609be48669597b4
+size 2671