CodeT5-small-custom-functions-dataset-python Open Source Model - Free Generation of Python Custom Function Code

Codet5 Small Custom Functions Dataset Python

Developed by sharoz

A Python code generation model fine-tuned based on Salesforce/codet5-small, specializing in custom function generation tasks

Large Language Model

Transformers

Open Source License:Apache-2.0 #Python code generation #Function-level fine-tuning #Small model efficiency

Downloads 43

Release Time : 5/14/2023

Model Overview

This model is a small version of the CodeT5 architecture, specifically fine-tuned for Python custom function generation tasks. Suitable for development assistance scenarios such as code completion and function generation.

Model Features

Specialized in Python function generation

Specifically optimized for Python custom function generation tasks

Lightweight architecture

Based on CodeT5-small architecture, suitable for resource-constrained environments

Fine-tuning performance optimization

Achieved a validation loss of 0.2103 after fine-tuning on a specific dataset

Model Capabilities

Python code generation

Function auto-completion

Code transformation

Use Cases

Development assistance tools

IDE plugin

Integrated into development environments to provide real-time function suggestions

Improves development efficiency

Code completion system

Generates complete functions based on partial code snippets

🚀 codet5-small-custom-functions-dataset-python

This model is a fine - tuned version of Salesforce/codet5-small on the None dataset. It can achieve a loss of 0.2103 on the evaluation set, which shows its good performance in relevant tasks.

🚀 Quick Start

This model is a fine - tuned version of Salesforce/codet5-small on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.2103

🔧 Technical Details

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e - 05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
6.8821	0.03	1	4.9003
5.1641	0.06	2	4.1876
4.5747	0.09	3	3.5772
3.985	0.12	4	3.0527
4.0255	0.15	5	2.5962
3.1963	0.18	6	2.2589
3.01	0.21	7	1.9755
2.5837	0.24	8	1.7736
2.6645	0.27	9	1.6032
1.8825	0.3	10	1.4620
2.282	0.33	11	1.3621
1.9555	0.36	12	1.2926
2.0374	0.39	13	1.2261
1.6276	0.42	14	1.1631
1.937	0.45	15	1.1053
1.4738	0.48	16	1.0512
1.5335	0.52	17	1.0016
1.5224	0.55	18	0.9554
1.5048	0.58	19	0.9175
1.3983	0.61	20	0.8806
1.2506	0.64	21	0.8495
1.186	0.67	22	0.8243
1.1824	0.7	23	0.7988
1.29	0.73	24	0.7728
1.159	0.76	25	0.7468
0.9893	0.79	26	0.7193
1.2054	0.82	27	0.7013
1.0004	0.85	28	0.6850
0.7918	0.88	29	0.6704
1.0357	0.91	30	0.6570
1.0648	0.94	31	0.6452
1.0679	0.97	32	0.6336
0.9296	1.0	33	0.6227
0.8459	1.03	34	0.6123
0.8312	1.06	35	0.6000
0.9367	1.09	36	0.5844
0.8813	1.12	37	0.5724
0.9134	1.15	38	0.5608
0.6967	1.18	39	0.5509
0.8654	1.21	40	0.5416
0.784	1.24	41	0.5324
0.7623	1.27	42	0.5237
0.739	1.3	43	0.5145
0.8273	1.33	44	0.5064
0.7384	1.36	45	0.4968
0.6936	1.39	46	0.4882
0.7078	1.42	47	0.4807
0.6214	1.45	48	0.4740
0.6983	1.48	49	0.4662
0.6328	1.52	50	0.4588
0.663	1.55	51	0.4533
0.6518	1.58	52	0.4476
0.5782	1.61	53	0.4343
0.6361	1.64	54	0.4296
0.5804	1.67	55	0.4249
0.6557	1.7	56	0.4210
0.6801	1.73	57	0.4173
0.6682	1.76	58	0.4132
0.6346	1.79	59	0.4090
0.6421	1.82	60	0.4028
0.6318	1.85	61	0.3969
0.6914	1.88	62	0.3942
0.5953	1.91	63	0.3920
0.7016	1.94	64	0.3894
0.5728	1.97	65	0.3839
0.5417	2.0	66	0.3738
0.5502	2.03	67	0.3705
0.5167	2.06	68	0.3668
0.6452	2.09	69	0.3629
0.4713	2.12	70	0.3583
0.5239	2.15	71	0.3553
0.6125	2.18	72	0.3527
0.4548	2.21	73	0.3414
0.5705	2.24	74	0.3389
0.4912	2.27	75	0.3374
0.4566	2.3	76	0.3316
0.5642	2.33	77	0.3288
0.4212	2.36	78	0.3260
0.3808	2.39	79	0.3236
0.4833	2.42	80	0.3214
0.4775	2.45	81	0.3193
0.5598	2.48	82	0.3175
0.5144	2.52	83	0.3162
0.4554	2.55	84	0.3152
0.4811	2.58	85	0.3141
0.4545	2.61	86	0.3130
0.438	2.64	87	0.3117
0.4071	2.67	88	0.3104
0.4635	2.7	89	0.3090
0.5118	2.73	90	0.3077
0.4043	2.76	91	0.3059
0.4675	2.79	92	0.3044
0.4551	2.82	93	0.3021
0.497	2.85	94	0.2987
0.4334	2.88	95	0.2932
0.4087	2.91	96	0.2901
0.477	2.94	97	0.2888
0.4834	2.97	98	0.2871
0.4513	3.0	99	0.2856
0.4172	3.03	100	0.2845
0.3827	3.06	101	0.2837
0.3851	3.09	102	0.2830
0.3976	3.12	103	0.2823
0.4909	3.15	104	0.2833
0.5409	3.18	105	0.2830
0.4039	3.21	106	0.2808
0.4057	3.24	107	0.2789
0.4214	3.27	108	0.2779
0.4209	3.3	109	0.2768
0.5044	3.33	110	0.2759
0.3457	3.36	111	0.2750
0.394	3.39	112	0.2744
0.4008	3.42	113	0.2739
0.3837	3.45	114	0.2736
0.3843	3.48	115	0.2734
0.4458	3.52	116	0.2730
0.4417	3.55	117	0.2725
0.4274	3.58	118	0.2719
0.4129	3.61	119	0.2712
0.421	3.64	120	0.2702
0.3625	3.67	121	0.2692
0.3785	3.7	122	0.2683
0.4023	3.73	123	0.2671
0.416	3.76	124	0.2663
0.3661	3.79	125	0.2654
0.373	3.82	126	0.2647
0.4045	3.85	127	0.2640
0.3955	3.88	128	0.2633
0.3796	3.91	129	0.2627
0.3682	3.94	130	0.2621
0.4195	3.97	131	0.2614
0.4135	4.0	132	0.2609
0.3244	4.03	133	0.2601
0.411	4.06	134	0.2597
0.4019	4.09	135	0.2599
0.451	4.12	136	0.2592
0.3948	4.15	137	0.2584
0.3375	4.18	138	0.2577
0.3687	4.21	139	0.2567
0.3946	4.24	140	0.2557
0.4181	4.27	141	0.2547
0.2949	4.3	142	0.2540
0.3621	4.33	143	0.2530
0.4134	4.36	144	0.2523
0.3366	4.39	145	0.2516
0.3798	4.42	146	0.2510
0.3519	4.45	147	0.2505
0.2999	4.48	148	0.2501
0.4096	4.52	149	0.2495
0.4736	4.55	150	0.2485
0.3481	4.58	151	0.2481
0.3683	4.61	152	0.2479
0.325	4.64	153	0.2476
0.3746	4.67	154	0.2473
0.3394	4.7	155	0.2468
0.3653	4.73	156	0.2463
0.3222	4.76	157	0.2458
0.3496	4.79	158	0.2453
0.368	4.82	159	0.2450
0.3473	4.85	160	0.2447
0.3712	4.88	161	0.2445
0.3542	4.91	162	0.2443
0.3249	4.94	163	0.2436
0.3135	4.97	164	0.2431
0.3603	5.0	165	0.2427
0.3345	5.03	166	0.2424
0.3385	5.06	167	0.2428
0.3939	5.09	168	0.2422
0.334	5.12	169	0.2414
0.3482	5.15	170	0.2401
0.3323	5.18	171	0.2396
0.3603	5.21	172	0.2391
0.354	5.24	173	0.2385
0.3241	5.27	174	0.2379
0.4134	5.3	175	0.2373
0.3726	5.33	176	0.2369
0.2997	5.36	177	0.2364
0.3317	5.39	178	0.2360
0.3692	5.42	179	0.2356
0.3411	5.45	180	0.2347
0.274	5.48	181	0.2342
0.3714	5.52	182	0.2337
0.442	5.55	183	0.2332
0.3262	5.58	184	0.2327
0.2929	5.61	185	0.2323
0.3435	5.64	186	0.2315
0.3921	5.67	187	0.2311
0.3609	5.7	188	0.2306
0.3585	5.73	189	0.2302
0.3323	5.76	190	0.2298
0.3205	5.79	191	0.2295
0.3407	5.82	192	0.2293
0.3109	5.85	193	0.2290
0.3075	5.88	194	0.2287
0.3538	5.91	195	0.2285
0.2968	5.94	196	0.2283
0.34	5.97	197	0.2281
0.3608	6.0	198	0.2279
0.2768	6.03	199	0.2277
0.3783	6.06	200	0.2275
0.3024	6.09	201	0.2272
0.3221	6.12	202	0.2269
0.3432	6.15	203	0.2266
0.3497	6.18	204	0.2264
0.3174	6.21	205	0.2261
0.3034	6.24	206	0.2259
0.3035	6.27	207	0.2257
0.3185	6.3	208	0.2255
0.3851	6.33	209	0.2252
0.3612	6.36	210	0.2249
0.2838	6.39	211	0.2247
0.3452	6.42	212	0.2245
0.3358	6.45	213	0.2243
0.3181	6.48	214	0.2241
0.329	6.52	215	0.2240
0.2819	6.55	216	0.2238
0.3283	6.58	217	0.2237
0.2752	6.61	218	0.2235
0.3194	6.64	219	0.2233
0.2981	6.67	220	0.2230
0.2954	6.7	221	0.2229
0.2762	6.73	222	0.2228
0.3206	6.76	223	0.2223
0.3017	6.79	224	0.2221
0.3219	6.82	225	0.2219
0.2929	6.85	226	0.2215
0.3576	6.88	227	0.2212
0.2712	6.91	228	0.2210
0.2682	6.94	229	0.2207
0.3412	6.97	230	0.2205
0.3136	7.0	231	0.2203
0.3161	7.03	232	0.2200
0.2902	...	...	...

📄 License

This model is licensed under the Apache - 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご